Missouri Department of Social Services

PREDECISIONAL

Rapid Response Review –

Assessment of Missouri Medicaid Program

FINAL REPORT

February 11, 2019

Missouri Department of Social Services      PREDECISIONAL  
Rapid Response Review – Assessment of Missouri Medicaid Program  
 2 
 
 
 
Contents 
Executive Summary  3 
Overview of Medicaid Program  7 
The Current Program  7 
Major Industry Trends Affecting the Program  11 
Projected Spending Without Significant Course Corrections  13 
Opportunities and Potential Initiatives  16 
Acute Care Services  17 
Long-Term Services and Supports  43 
Pharmacy  57 
Managed Care  62 
Program Integrity  73 
Federal Financing  79 
Medicaid Management Information System (MMIS)  83 
Operations  93 
Implementation Considerations  110 
 
   

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Executive Summary

Missouri’s Medicaid program is an important safety net for Missouri’s most vulnerable

populations, providing health care and support for activities of daily living for nearly one million

Missourians. Children in low-income families comprise 63.5% of participants, while persons with

disabilities comprise the largest share (46%) of spending. In State Fiscal Year (SFY) 2018, total

spending for the program was approximately $10 billion, funded 53% by federal funds, 21% by

state general revenues, and the balance by provider taxes and other funds.

Analysis of historical trends indicates that the financial sustainability of Missouri’s Medicaid

program is currently under pressure: Medicaid spending has grown from 17% of state general

revenues in SFY2009 to 24% in SFY2018. Based on continuation of these trends, spending

could grow to 26% by SFY2023. This number could be even higher (30%) if Missouri were to

experience an economic downturn, given the potential for such a downturn to increase Medicaid

enrollment while also reducing growth in state general revenues.

Under any of the scenarios described in the pages that follow, significant changes in the

structure and performance of Missouri’s Medicaid program would be necessary to bring

Medicaid spending growth in line with projected economic growth of the state.

Opportunities for Improvement

Since October, a detailed assessment of the Missouri Medicaid program has uncovered a wide

range of opportunities for improvement, which may collectively deliver substantial reductions in

the rate of growth of Medicaid spending. These opportunities, detailed in the pages that follow,

are based on extensive interviews with state agency leaders and staff, detailed analysis of

claims- and non-claims data, review of agency operations, and benchmarking against other

states as well as Medicaid managed care organizations (MCOs) and other health insurers.

High-level summaries are provided below for each of eight programmatic and functional areas.

Further details follow in the report, including descriptions of opportunities (with supporting facts)

and potential initiatives that may be considered by the state in shaping its approach to Medicaid

transformation. Potential initiatives include possible changes to provider payment methods, care

management, and contracting with MCOs and other vendors, as well as improvements in

agency and vendor operations. A selection of potential initiatives outlined in the following pages

could collectively enable the Medicaid program to achieve significant cost savings while

maintaining or improving access to high-quality care, without broad-based cuts in provider rates,

or reductions in eligibility or covered services.

Were Missouri to effectively address the opportunities and potential initiatives outlined in this

report, total savings to the program (including federal and state share) could total up to $0.5-1.0

billion by SFY2023. This level of savings, while significant, does not represent an absolute

reduction in the size of the Medicaid program but rather a meaningful reduction in the rate of

growth of Medicaid spending, to bring it more closely in line with growth of the economy.

Implementing changes at this scale would make the program more financially sustainable under

all future financing scenarios, allowing for the state to continue to make investments to protect

the program’s essential role in serving the state’s most vulnerable populations.

Acute Care Services. Missouri spent ~$4.2 billion in SFY2018 on acute care services, including

hospital, clinic, physician, and diagnostics services, across both the managed care and non-

managed care populations. (Pharmaceutical services are discussed in a separate section).

Provider payment for acute care providers in Missouri is currently almost exclusively fee-for-

service. Fee schedules are based on historical costs; in some cases, these are adjusted each

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

year based on changes in operating costs. Accordingly, the payment methods used offer few

incentives for providers to contain costs. A significant proportion of Missouri Medicaid acute

care expenditures is associated with potentially avoidable exacerbations and complications

(PECs) and inefficiencies in the choice of provider, site, or treatment. In addition, Missouri

Medicaid is unique in making “add-on payments” to hospitals for services provided by Missouri

hospitals to non-Missouri residents. Potential initiatives to improve incentives and reduce costs

include adjusting rate setting methodologies, moving to value-based payment models, and

investing in the rural and safety net heath care infrastructure, including primary care and

behavioral health. In total, the gross impact of Acute Care initiatives could range anywhere from

$250 million to $500 million, depending on choices made by the state.

Long-Term Services and Supports. Missouri spent ~$2.9 billion in SFY2018 on long-term

services and supports (LTSS) for approximately 106,000 Medicaid participants accessing these

services. LTSS in Missouri consist of institutional services (e.g., nursing homes for frail elderly,

intermediate care facilities for individuals with intellectual disabilities), and home and

community-based services (both residential and non-residential) covered by the Medicaid State

Plan and nine waivers. Nursing facilities are reimbursed using a cost-based, facility-level per

diem methodology without adjustments for acuity, quality, or outcomes, and home and

community-based services (HCBS) are reimbursed on a fee-for-service basis. LTSS in Missouri

are administered by Missouri Medicaid in conjunction with the Department of Health and Senior

Services (DHSS) and the Department of Mental Health (DMH), each of which administers

HCBS waivers and conducts assessments to determine access to LTSS. The assessment

process currently in use by DHSS uses decades-old standards and, as such, may not

consistently determine institutional level of care accurately. Potential initiatives include

incorporating an acuity adjustment into the nursing home reimbursement methodology,

completing and expanding upon revisions currently underway for the state's assessment

algorithms, more directly applying assessment results in the care planning process, and

improving the consistency of the prior authorization approval process for personal care services.

In total, the gross potential impact across LTSS initiatives ranges from $90 million to $275

million, depending on choices made by the state.

Pharmacy. Missouri spent ~$1.5 billion in SFY2018 on pharmaceutical products. This spending

is inclusive of all participants as the state carves pharmacy benefits out of its MCO

arrangements. The state utilizes a preferred drug list and receives statutory and supplemental

rebates to help control costs. The basis for drug ingredient cost reimbursement was recently

updated, and Missouri is in the process of updating dispensing fees. Missouri rebate

performance is below the average for other states, potentially due to expansive grandfathering.

While the state uses a broad range of approaches to ensure appropriate utilization, there is an

opportunity to expand it to other high-cost drug classes such as oncology, hemophilia, and IVIG.

Potential initiatives include limiting grandfathering, implementing additional utilization

management, joining a purchasing consortium to increase supplemental rebate capture,

requiring NDC submission on claims for non-J-code HCPCS drugs, and applying for a value-

based contracting waiver from CMS. In total, the gross potential impact across Pharmacy

initiatives ranges from $35 million to $60 million, depending on choices made by the state.

Program Integrity. Program integrity functions within the state Medicaid agency center serve to:

prevent fraud, waste, and abuse; ensure proper participant enrollment and identify third-party

resources to pay for medical claims. Numerous divisions within the state help accomplish these

goals; however, the separation of divisions leads to siloed data and communication. Potential

initiatives include expanding adoption of best practices from the National Correct Coding

Initiative, updating certain medical and reimbursement policies to prevent improper payments,

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

implementing claims edits that would prevent improper billing based on Missouri’s current

policies, optimizing the state’s ability to identify and enroll participants who are eligible for

Medicare, and improving third-party liability identification. In total, Program Integrity initiatives

could deliver savings between $65 and $100 million, depending on choices made by the state.

Managed Care. Missouri's Medicaid managed care program covers primarily acute care and

professional services for non-disabled adults and children. Approximately 75% of Medicaid

participants are covered by managed care, whereas approximately 23% of Medicaid

expenditures (~$2.2 billion) flow through managed care contracts. Most pharmacy and

behavioral health services for the managed care population remain "carved out" of managed

care; the Medicaid aged, blind, and disabled (ABD) population is entirely excluded from the

current managed care program. Potential initiatives include incorporating additional adjustments

to managed care rates to remove inefficient utilization (e.g., inpatient stays that could have been

avoided with better outpatient care) from rate calculations, expanding day-one managed care

eligibility to streamline participant transitions and reduce residual fee-for-service payments,

strengthening key contract provisions and the compliance and performance management

relationship between MHD and the MCOs, and altering the scope of the managed care

program—for example, including additional services or Medicaid eligibility groups. In total, the

gross potential impact of all Managed Care-related initiatives ranges from $175 million to $300

million, depending on choices made by the state.

Federal Financing. Federal Financing focuses on identifying opportunities to optimize federal

funding for the state’s Medicaid program. Overall, Missouri has been able to capture a

significant share of the available federal funding opportunities, capturing funds across Medicaid

spending and non-Medicaid spending within DSS, DMH, and DHSS and capturing enhanced

match on select categories. However, there remain several opportunities that the state could

pursue to capture additional federal funding. These opportunities could include new waiver and

grant programs released in the SUPPORT for Patients and Communities Act, enhanced match

on substance use disorder (SUD) focused health homes, among others. In total, potential

impact from these improvements may be $10 million to $20 million, based on choices made by

the state.

Medicaid Management Information System. Missouri’s Medicaid Management Information

System (MMIS) is a set of ~70 components, partially developed within a mainframe-based

system dating from 1979. The system supports a wide range of vital activities within the

Medicaid program, but is not positioned to meet both current and future needs. Its limited

functionalities underlie several of the opportunities for improvement identified in other topical

areas. There is little alignment between program strategy and the MMIS replacement plan, and

the Information Systems (IS) group lacks the wide range of capabilities needed to ensure an

MMIS replacement trajectory that will deliver the future functionality Missouri needs. The

potential initiatives discussed in this section attempt to address these challenges.

Operations. This section analyzes the performance and operational efficiency of three

operational functions: participant managed care enrollment, claims processing, and contact

centers. These functions are currently executed through a mix of state staff and vendor

contracts. In comparison to other states, work processes often appear fragmented, process

steps seem poorly integrated and best-practice management principles are variably applied.

Potential initiatives include organizational process optimization, automation and digitization, and

improved contract management. Adoption of best practices across the different functional areas

could liberate up to 15-20% of operational resources, which could be redeployed to improve

service levels for participants and for other external and internal “customers” of the different

functions.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Implementation Considerations

The potential initiatives described in the following pages are wide-ranging, including operational

improvements to bring the program up to date with common practices among other state

Medicaid programs, as well as best practices and more transformational changes. Some of the

potential initiatives outlined represent alternative ways of achieving similar goals: in some

cases, the initiatives could reinforce one another; in other instances, they could be mutually

exclusive. Such interdependencies will be highlighted throughout.

Broadly, the state could balance two approaches to controlling spending. One approach

commonly adopted by both public programs and managed care would rely primarily on

controlling the unit prices paid for services and seeking to curb utilization through broad-based

utilization management. Such an approach could reduce costs in the short term. However, on its

own such an approach may not provide incentives to improve outcomes. As an alternative

approach, the state could seek to adopt value-based payment and care delivery models that

reward providers for quality and efficiency of the total care delivered to patients. This approach

may support more transformational changes in care delivery, with corresponding improvements

in patient outcomes and experience. However, such an approach is likely to require greater

commitment of resources and will take longer to generate impact given the need for providers to

adopt new capabilities and implement changes in clinical practices.

Aligning the growth of Medicaid expenditures with the state’s economic growth may involve a

combination of these approaches, with targeted use of utilization management and targeted

adjustments in provider rates in the near-term, combined with investments in care management

and value-based payment to support sustainable improvements in quality and efficiency. In

parallel, there may be a series of operational changes that the state could implement to bring

policies and operations up to speed with common practices, such as state-of-the-art program

integrity measures and improvements of internal administrative processes. Such changes could

generate near-term savings to offset investments in transformation changes.

Any substantial portfolio of initiatives would demand careful planning and execution, as well as

investments to support the transformation and build new capabilities. Key requirements for

effective design and implementation of Medicaid transformation include: strong and visible

executive leadership; effective stakeholder engagement; commitment to fact-based decision

making supported by robust data; upskilling of key agency staff; a well-resourced transformation

office; and modernization of the program’s technological infrastructure.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Overview of Medicaid Program

In State Fiscal Year 2018, Missouri Medicaid was a $10.3 billion program, funded by state

general revenue ($2.2 billion), federal funds ($5.5 billion), and other funds ($2.6 billion).

The

“other funds” consisted primarily of revenue from provider taxes ($1.4 billion).

Since 2009, Medicaid spending has grown in proportion to the total state budget, and in

proportion to state general revenues. In 2009, Medicaid spending comprised 17% of state

general revenues; in 2018, it was 24%.

Without significant changes in the Missouri Medicaid

program, spending growth may continue to outpace growth in state general revenues and

could comprise 26-30% of state general revenues by 2023.

The following pages provide a brief introduction of the Missouri Medicaid program and a

summary of key trends in the larger U.S. healthcare context that influence program spending,

as well as state fiscal scenarios that could lead Medicaid spending to represent a greater

share of state general revenues.

THE CURRENT PROGRAM

Enrollment and Spending

Missouri Medicaid is a $10.3 billion program that covers predominantly four types of

participants: low-income children; parents of low-income children; pregnant women; and aged,

blind, or disabled (ABD) individuals.

Children comprise the largest eligibility group in Missouri

Medicaid, representing 63.5% of enrollees; however, persons with disabilities account for the

greatest proportion (46%) of Medicaid spending (see Exhibit 1 and Exhibit 2).

Missouri DSS, “TSM Expenditures History FY05 to FY18,” 2018; Missouri DSS, “Final FY18 Total State Medicaid

Expenditures,” 2018.

Missouri DSS: see note 1.

Analysis based on projections from past trends. Missouri DSS: see note 1.

Missouri DSS, “MO HealthNet enrollees and expenditures,” 2018, see: dss.mo.gov/mhd/general/pdf/mhdollars.pdf.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 1: MEDICAID ENROLLMENT & SPENDING BY ELIGIBILITY CATEGORY, SFY2018

EXHIBIT 2: MEDICAID ENROLLMENT & SPENDING BY TYPE OF COVERAGE, SFY2018

Missouri DSS: see note 5.

63.5%

12.3%

8.2%

16.0%

Persons with

disabilities

Seniors

46.0%

$4.3B

Enrollees by

eligibility category

27.5%

$2.6B

16.9%

$1.6B

9.6%

$0.9B

Expenditures by

eligibility category

100% = $9.4 billion

Children

Pregnant women,

custodial parents

100% = 976,779 enrollees

63.5%

22.9%

12.3%

14.2%

8.2%

62.9%

16.0%

Seniors

Persons with

disabilities

Enrollees by

eligibility category

Expenditures by

type of coverage

Children

Pregnant women,

custodial parents

Fee-for-service

population

FFS spend for

MC population

Managed care

100% = 976,779 enrollees 100% = $9.4 billion

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Services for >155,000 persons with disabilities and >80,000 elderly participants (the aged, blind

and disabled [ABD] population) are provided through Medicaid fee-for-service, comprising

24.2% of enrollees but 62.9% of total program expenditures. The ABD population includes

distinctive subpopulations with different cost patterns: frail elderly, individuals with intellectual

and/or developmental disabilities, individuals with severe behavioral health issues, and others.

Non-disabled children, parents, and pregnant women comprise 75.8% of enrollees (about

650,000 in all) but 37.1% of total program costs, of which 22.9% is covered through managed

care and 14.2% (pharmacy and behavioral health) is paid through Medicaid fee-for-service.

Exhibit 3 shows one possible categorization of diagnostic groups,

including numbers of

participants per group, and average per member per month (PMPM) spending per main

category (institutional long-term services and supports [LTSS], home and community-based

services [HCBS], acute services).

All elderly and 40% of individuals with disabilities are dually eligible for both Medicaid and

Medicare. For these participants, Medicare pays for the acute care costs (e.g., hospitals,

physicians, drugs); Medicaid pays for long-term services and supports (e.g., home care, nursing

homes).

EXHIBIT 3: SUBPOPULATIONS WITHIN THE ABD POPULATION, SFY2018

Medical diagnosis, procedure codes and demographic information from Missouri Medicaid claims for SFY2018

were used to develop sub-segmentations of ABD population by diagnosis category. Diagnosis categories are based

on claims data only, with the exception of the frail elderly category which is based on age and participant chronicity.

Analysis of Missouri Medicaid claims data, SFY2016-18; participants may overlap across categories.

6.8

36.4

38.6

73.4

28.3

20.7

22.2

6.6

Institution

HCBS

LTSS

Non-LTSS

Other

Physical

disability

Sensory Deficit

Conditions

High needs BH

ID/DD

ABI

Neurodegenera-

tive Conditions

SFY 2018 average

participants, K

Average spend, $K

PMPYABD population Description

Annualized

participants, %

total

Frail elderly

35%

70%

27%

37%

20%

21%

298

1,156

2,863

1,264

1,750

229

1,245

1,060

Total spend, $M

▪ Do not meet criteria of any of the above

▪ Traumatic brain injury

▪ Cerebral infarction

▪ Cerebral hemorrhage

▪ Genetic syndromes with intellectual disability

▪ Congenital brain injury

▪ Developmental disability (e.g., Autism)

▪ Presence of a behavioral health diagnosis, or

utilization, with at least one mental health

condition

▪ Dementia, Parkinson’s, Alzheimer’s, ALS,

Multiple Sclerosis

▪ Blindness

▪ Hearing loss

▪ Spinal cord injury

▪ Paralysis

▪ Chronic pain / weakness

▪ Age 65+ with 2 or more chronic conditions

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Financing and Budget

Missouri Medicaid is funded by state general revenue ($2.2 billion), federal funds ($5.5 billion),

and other funds ($2.6 billion).

The “other funds” are primarily provider taxes ($1.1 billion from

hospitals, and $0.3 billion from nursing homes and pharmacies).

As these taxes generate

federal matching funds (nearly two federal dollars for every dollar generated through provider

taxes

), they are significant to the financing of the Missouri Medicaid program. Exhibit 4 shows

the flow of funds, highlighting the hospital tax and its integration in the Medicaid funding flow.

EXHIBIT 4: MISSOURI MEDICAID FUNDS FLOW (SFY2018, USD BILLIONS)

Organization

Three departments jointly manage parts of the Medicaid program. The Department of Social

Services (DSS) operates MO HealthNet (MHD), which is primarily responsible for medical

expenses for eligible individuals. This includes both the fee-for-service and the managed care

populations. The Department of Mental Health (DMH) administers services for populations with

Missouri DSS: see note 1.

Missouri DSS, “Provider taxes overview,” 2018.

Kaiser Family Foundation (KFF), “Federal Medical Assistance Percentage (FMAP) for Medicaid and Multiplier,”

2018, see: www.kff.org/medicaid/state-indicator/federal-matching-rate-and-multiplier.

Office of Administration, “The Missouri budget fiscal year 2018 summary,” 2018, see:

www.oa.mo.gov/sites/default/files/FY_2018_Budget_Summary.pdf; Missouri DSS: see note 1; Missouri DSS: see

note 11; Missouri DSS, “Annual_Table 23 and 24_FY18 by Large Group PMPM,” 2018; Missouri DSS, “Payments

Assessment for SFY 14-18,” 2018.

Federal funds

Other funds

General revenue

Medicaid spend

5.5

2.2

2.6

9.4

8.8

9.4

10.3

1.1

Federal match of provider tax

2.2

State gov’t collects ~5.5%

FRA tax on provider

revenue

Federal gov’t

matches revenue

collected from

provider tax

27.6

Overall MO budget Medicaid Hospitals

IP/OP reimbursement

(Medicaid FFS)

2.8

IP/OP reimbursement

(all other payors)

17.9

IP/OP reimbursement

(Medicaid MC)

1.5

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

developmental disabilities (both intellectual and physical disabilities, as well as certain learning

disabilities), community-based health centers, psychiatric rehabilitation services, comprehensive

substance treatment and rehabilitation (CSTAR) services, and health home programs, amongst

others. The Department of Health and Senior Services (DHSS) operates the Division of Senior

and Disability Services (DSDS), which administers the HCBS benefits for adults 18 and over,

and Special Health Care Needs (SHCN), which administers the Healthy Children and Youth

benefits for persons with special health care needs up to 21 years of age. DSS is responsible

for the largest share of Medicaid spending at approximately $7.9 billion, DMH’s share is $1.5

billion, and DHSS’ share of Medicaid spending is $0.9 billion.

MAJOR INDUSTRY TRENDS AFFECTING THE PROGRAM

Healthcare inflation rising faster than GDP

The United States faces increasing pressure to contain its rising healthcare costs. In 2017, total

U.S. healthcare spending reached $3.5 trillion, marking a 3.9% increase from the previous year,

amounting for almost 18% of gross domestic product (GDP)

. U.S. health spending per person

climbed to over $10,739 in 2017, the third year that the spending has exceeded $10,000.

The

growth in per-person spending, or medical cost inflation, outpaces the general inflation rate (see

Exhibit 5). CMS projects spending to grow 1% faster than GDP to reach $5.7 trillion by 2026.

EXHIBIT 5: MEDICAL INFLATION, GENERAL INFLATION, GDP GROWTH, SFY2001-17, %

Missouri DSS: see note 1.

CMS, “NHE Fact Sheet,” 2017, see: www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-

reports/nationalhealthexpenddata/nhe-fact-sheet.html.

CMS, “The National Health Expenditure Summary including share of GDP, CY 1960-2017,” 2017,

www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-

Reports/NationalHealthExpendData/NationalHealthAccountsHistorical.html.

The Balance, “U.S. Inflation Rate by Year from 1929 to 2020,” 2019, see: www.thebalance.com/u--inflation-rate-

history-by-year-and-forecast-3306093; The Balance, “U.S. GDP by Year Compared to Recessions and Events,”

2019, see: www.thebalance.com/us-gdp-by-year-3305543.

The Balance: see note 17.

03 09052001 1402 04 06 1007 11 20171208 13 15 16

-3

Medical inflation rate

General inflation rate

GDP growth rate

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Costs are rising in Medicare and Medicaid, putting pressure on both federal and state budgets.

In addition, rising health care costs create challenges in the commercial market as well.

Employees’ contribution to health insurance grew almost three times faster than wages between

2010 and 2015,

and middle-class Americans’ healthcare spending increased 60% over the

past 30 years.

The growing financial burden of healthcare has been a significant factor in the

low growth in purchasing power of the middle class in the U.S. over the past two decades.

While rising health expenditures are not unique to the U.S., its spending exceeds that of other

countries even after adjusting for differences in average wealth. While outcomes of U.S.

healthcare exceed those of other countries for some catastrophic illnesses and other complex

conditions (e.g. breast and colorectal cancer), outcomes lag other countries for most chronic

conditions.

Chronic conditions, in particular, account for a significant portion of healthcare

spending growth, underscoring concerns that increased levels of investment in healthcare have

not translated to proportional improvements in life expectancy or quality of life (see Exhibit 6).

EXHIBIT 6: LIFE EXPECTANCY AND HEALTH EXPENDITURE ACROSS COUNTRIES

KFF, “2015 Employer Health Benefits Survey,” 2015, see: www.kff.org/health-costs/report/2015-employer-health-

benefits-survey.

Hamilton Project, “Where Does All the Money Go: Shifts in Household Spending Over the Past 30 Years,” 2016,

see:www.hamiltonproject.org/papers/where_does_all_the_money_go_shifts_in_household_spending_over_the_pa

st_30_y.

Pew Research, “For most U.S. workers, real wages have barely budged in decades,” 2018, see:

www.pewresearch.org/fact-tank/2018/08/07/for-most-us-workers-real-wages-have-barely-budged-for-decades/.

Peterson-Kaiser Health System Tracker, “How does health spending in the U.S. compare to other countries?”

2018, see: www.healthsystemtracker.org/chart-collection/health-spending-u-s-compare-countries;

www.kff.org/slideshow/how-does-the-quality-of-the-u-s-healthcare-system-compare-to-other-countries/.

Our World in Data, “Link between health spending and life expectancy: US is an outlier,” 2017, see:

ourworldindata.org/the-link-between-life-expectancy-and-health-spending-us-focus.

1,000$

Health expenditure

2,000$ 3,000$500$

Life expectancy

0 4,000$ 5,000$ 6,000$ 7,000$ 8,000$ 9,000$

2014

2010

2005

2000

1995

1990

1985

1980

1975

1970

USA

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

CMS taking steps to reduce federal spending on health care

The Centers for Medicare & Medicaid Services (CMS) have undertaken a series of initiatives to

reduce federal healthcare costs for Medicare, ranging from attempts to reduce what it pays for

drugs and outpatient visits in Medicare to limiting cost growth in Medicare Advantage plans.

As the federal government will pay for 50% to 78% of Medicaid costs through federal match in

2019 (65% in Missouri)

, federal spending on Medicaid is similarly assessed for cost reduction

opportunities: CMS has announced its intention to increase the level of scrutiny of Medicaid

waiver expenditures.

In addition, CMS has described its intention to increase audits of state

claims for federal matching funds and beneficiary eligibility determination, among others.

Public and private payors are migrating to value-based payment

There is a broad consensus that one of the key drivers of waste and inefficiency in U.S.

healthcare is the fee-for-service (FFS) payment model that characterizes most provider

payments. FFS stimulates volume rather than coordination of services, and there are no

inherent incentives to achieve optimal outcomes nor to deliver care in an efficient manner.

Both public and private payors are transitioning from FFS to value-based payment (VBP), using

Alternative Payment Models (APMs) to reward providers for delivering high-quality care at lower

cost. Research suggests that well-designed APMs improve the quality of care and can

meaningfully reduce the cost of care if implemented across the full spending base.

PROJECTED SPENDING WITHOUT SIGNIFICANT COURSE CORRECTIONS

Over the last 10 years, Medicaid spending growth has outpaced growth in state general

revenues. For example, in 2009, 17% of state general revenue funds were directed toward

Medicaid; however, in the wake of the Great Recession, Medicaid grew to 22% of state general

revenue by 2012 and, in 2018, reached 24% (see Exhibit 7). Although Medicaid enrolment has

fluctuated over this timeframe, total Medicaid spending growth has outpaced growth in state

general revenues when measured over any five-year timeframe, due to the increase in spending

per participant enrolled in the program. Notwithstanding reductions in Medicaid enrollment

observed since the beginning of SFY2019, program spending is likely to occupy a greater share

of state general revenue over the coming five years, absent changes in program performance.

In this section, we consider three scenarios as a method for gauging the level of fiscal pressure

that may arise from Medicaid spending growth. In Scenario 1, Medicaid spending as a share of

state general revenues continues to grow at a pace similar to the last five years; in Scenario 2,

CMS, “CMS-1695-P: Hospital Outpatient Prospective Payment- Notice of Proposed Rulemaking (NPRM),” 2019,

see: www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/HospitalOutpatientPPS/Hospital-Outpatient-

Regulations-and-Notices-Items/CMS-1695-P.html; CMS, “Prior Authorization and Step Therapy for Part B Drugs in

Medicare Advantage,” 2018, see: www.cms.gov/Medicare/Health-

Plans/HealthPlansGenInfo/Downloads/MA_Step_Therapy_HPMS_Memo_8_7_2018.pdf.

The percentage varies by state; for Missouri the match rate in FY2019 is 65.4%.

CMS, “Budget Neutrality Policies for Section 1115(a) Medicaid Demonstration Projects,” 2018, see:

www.medicaid.gov/Federal-Policy-Guidance/downloads/SMD18009.pdf.

CMS, “CMS announces initiatives to strengthen Medicaid program integrity,” 2018, see:

www.cms.gov/newsroom/press-releases/cms-announces-initiatives-strengthen-medicaid-program-integrity.

Health Care Learning and Action Network, "Alternative payment model (APM) framework. Updated version 2017,"

2017, see: hcp-lan.org/workproducts/apm-refresh-whitepaper-final.pdf.

McKinsey, “The seven characteristics of successful alternative payment models,” 2019, see:

www.mckinsey.com/industries/healthcare-systems-and-services/our-insights/the-seven-characteristics-of-

successful-alternative-payment-models.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

an economic downturn accelerates the growth of Missouri Medicaid spending while reducing the

growth of state general revenue; in Scenario 3, the trends from Scenario 1 are exacerbated by

potential changes in CMS policies that would reduce federal revenue for Missouri.

EXHIBIT 7: MEDICAID SPENDING AS PERCENT OF GENERAL REVENUE, SFY2009-23

Scenario 1—Continuation of Recent Trends: Spending on the program increased from 21%

of state general revenue in 2013 (36% of total state spending) to 24% (40% of total spending) in

2018.

In the first scenario illustrated in Exhibit 7, the assumption is that growth of both

Medicaid spending and total state spending continue at the pace observed over these past five

years. Under this scenario, total Medicaid spending would increase to $12.8 billion by 2023,

while the state’s total spending would grow to $29.2 billion (growing 2.5% annually) by 2023. In

2023, Medicaid spending would comprise 44% of the overall state spending and 26% of state

general revenues.

Scenario 2—Potential Impact of Economic Downturn: In the second scenario, a severe

economic downturn (comparable to the recession that began in 2008) affects both state general

revenues and Medicaid enrollment. Based on analyses by Moody’s, an economic downturn

occurring within the next four years could reduce total general revenue by approximately 12%

over a two-year period. This reduction in general revenue will likely coincide with an increase in

Medicaid enrollment, as individuals lose jobs and incomes fall, resulting in an estimated

increase in Medicaid spending by 1.9% in total over the course of two years.

These changes

could translate to Medicaid spending growing to comprise 51% of total state spending and 30%

of state general revenues in 2023.

Medicaid expenditures have continued to outpace economic growth. Missouri DSS: see note 1.

Missouri DSS: see note 1.

Moody’s, “Stress-Testing States,” 2017, see: www.economy.com/getlocal?q=91a42834-85af-4773-b408-

5da811028c00&app=eccafile.

21%

2009 22

23%

171411

20%

22%

28%

17%

1610

24%

21%

23%

15 2018 19

22%

30%

2023

Status quo

(Scenario 1)

Recession

(Scenario 2)

CMS policy changes

(Scenario 3)

19%

26%

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Scenario 3—Potential Changes in CMS Policies: In the third scenario, no recession occurs,

but CMS policy changes could lead to reductions in federal funds for the state’s Medicaid

program and, consequently, to greater pressure on the state budget. A 3.5% drop in net federal

funds could reduce federal funding by approximately $200 million in SFY2023.

If the Medicaid

program’s total spending were to remain unchanged, the loss of federal funds may need to be

offset by a similar increase in funding from state general revenues. This could increase the

program’s share of state general revenues to 28% in SFY2023.

Savings needed to keep spending growth in line with State General Revenues

Under any of the scenarios described above, significant changes in the structure and

performance of Missouri’s Medicaid program would be necessary to bring Medicaid spending

growth in line with projected economic growth of the state. Following are estimates of the

reductions in Medicaid program spending that would be necessary to maintain spending at 24%

of state general revenues through SFY2023, under each of the three scenarios.

• In Scenario 1 (continuation of recent historical trends), Missouri would need to reduce the

growth rate of Medicaid spending by approximately 2 percentage points to bring it in line

with the growth of state general revenue, to maintain spending at 24% of state general

revenue. In SFY2023, this would equal approximately $735 million savings to total

Medicaid spending, or approximately $260 million savings to state general revenues.

• In Scenario 2 (occurrence of a recession similar in magnitude to that experienced 10 years

ago), it would be necessary to reduce Medicaid spending by nearly $1.7 billion, or $590

million in spending from state general revenues, to maintain spending at 24% of state

general revenues.

• In Scenario 3 (continuation of historical trends, exacerbated by CMS policy changes), it

would be necessary to reduce total program spending by approximately $1.3 billion, or

$460 million in spending from state general revenues to maintain spending at 24% of state

general revenue.

The funding gap implied by the above scenarios is meant to provide context for understanding

the estimated $0.5 billion to $1.0 billion in cost savings associated with the opportunities and

potential initiatives detailed in the pages that follow. Maintaining spending at 24% of state

general revenues may not necessarily represent the state’s policy objective and may not be

feasible in all future scenarios. In all scenarios, however, implementation of initiatives such as

those outlined in the pages that follow could help the Medicaid program to reduce fiscal

pressure on the state budget while maintaining or improving access to high-quality care, without

broad-based cuts in provider rates, or reductions in eligibility or covered services.

As outlined in the section on acute care services, existing risks to the state’s federal match revenue exist (e.g.

inpatient UPL calculations, planned federal reduction of DSH payments, DSH payments-related legal

developments, federal scrutiny of existing provider tax pooling arrangements, federal initiatives to reduce the

provider tax safe harbor, and so forth). $200 million is a low estimate of the impact of any combination of two to

three of these risks becoming reality.

Assumes a corresponding $120 million decrease in provider tax and $80 million decrease in other cuts to federal

funding (e.g., DSH payments). The savings would bring the percentage of Medicaid spending of state general

revenue to SFY2018 levels.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Opportunities and Potential Initiatives

Without significant changes, Medicaid spending may increase from 24% of state general

revenues in SFY2018 to comprise 26% to 30% of state general revenues by SFY2023.

Significant cost savings would be necessary to bring growth of Medicaid spending in line with

the level of economic growth of the state, while preserving access to care for participants.

The Missouri Department of Social Services, MO HealthNet Division (MHD) commissioned a

rapid, in-depth independent assessment of its programs and operations to identify potential

opportunities and strategies to transform the Missouri Medicaid program, including evaluation of

which functions the Department is performing well in, what activities or practices could be

improved, and what priorities could be considered for future investment.

Overall, the Missouri Medicaid program is currently outdated in most aspects compared to other

peer states, and significant opportunities exist relative to industry best practices:

• Dollars spent in the program are not well aligned with value received from delivery system;

• Specifically, methods to pay providers lack incentives to contain costs or enhance quality;

• Approaches to utilization management; eligibility management; fraud, waste, and abuse;

and third-party liability are limited, partially due to the limitations of the MMIS (see below);

• Programs for special needs populations are fragmented;

• There is no substantial measurement nor transparency of outcomes of care; and

• Service levels to consumers and providers could also be improved, including reductions in

average wait times for handling questions, as well as increased service channels.

Leaders and staff in DSS, DMH, and DHSS are aware of these challenges and highly motivated

to modernize the program. However, the foundational operational capabilities to do so are

equally outdated, hampering opportunities for improvement: the existing technology

infrastructure (MMIS) is antiquated; data quality needed for program management is

suboptimal; and access to key management information is absent.

Detailed findings from the assessment conducted over the past several months are outlined in

the pages that follow, organized into eight project areas which collectively address sixteen

performance opportunities prioritized by DSS at the outset of this assessment. For each topical

area, potential opportunities for improvement have been identified and evaluated through

interviews with functional leaders and subject matter experts within the relevant departments,

analysis of claims- and non-claims data, review of activities and operations, assessment of

supporting infrastructure and analytics, and benchmarking against other state Medicaid

programs, Medicaid managed care organizations (MCOs), or other private health insurers.

Based on these opportunities, a wide range of potential initiatives have been outlined for further

consideration by the Department. Depending on the selection of initiatives the state chooses to

pursue, total gross savings to the program (including federal and state share) could total up to

$0.5 billion to $1.0 billion by SFY2023. These estimated savings would be net of reinvestments

in the delivery system (e.g., in primary care, rural health, and the safety net; as well as rewards

for providers that generate savings under value-based payment models) and in the Medicaid

program’s operations to improve service levels to participants and providers.

One-time investments as well as MMIS replacement investments are not included.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Many of the opportunities for improvement could require changes in policies and contracts

requiring cooperation of local providers, support from the state legislature, and in some cases

federal approval. Potential initiatives outlined in this report are not meant to represent advocacy

for specific policies, nor conclusions yet reached by DSS. The state retains sole responsibility

for decision making over which of these potential improvement initiatives (or others) to pursue,

and in what form, in compliance with applicable laws, rules and regulations.

ACUTE CARE SERVICES

Missouri spent ~$4.2 billion in SFY2018 on acute care services, including hospital, clinic,

physician, and diagnostics services, across both the managed care and non-managed care

populations. Including pharmaceutical services, the total is ~$5.7 billion.

Hospitals are paid

through a combination of base rates and “add-on payments”, updated periodically based on

changes in hospital operating expenses. This approach offers minimal incentives for providers

to contain costs, making it an outlier among states. Uniquely, Missouri Medicaid makes add-on

payments to hospitals for services provided to non-Missouri, non-eligible residents.

Physicians and behavioral health providers are also paid per service (fee-for-service [FFS]).

Compared to other states, physicians’ reimbursement is low. Also, total spending on non-

hospital acute care services (physicians, Federally Qualified Health Centers [FQHCs], clinics,

and rural health services) is lower than other, comparable states.

The overall value of care delivered (dollars spending vs outcomes for participants) varies

significantly across counties. In general, >15% of Missouri Medicaid acute care expenditures

may be associated with potentially avoidable exacerbations and complications (PECs), which

includes costs of PECs associated with the prevention and treatment of opioid use disorder

(OUD). In addition, 5% to 10% of expenditures may be associated with inefficiencies, such as a

site of service or choice of therapy that might be more expensive without adding quality.

Potential initiatives to improve incentives and reduce costs range from adjusting rate setting

methodologies, moving to value-based payment models and investing in the rural and safety net

heath care infrastructure, including primary care and behavioral health. In total, the impact of the

acute care initiatives could range from $250 million to $500 million, net of potential

reinvestments in the delivery system, depending on the state’s choices.

To achieve the higher

end of this range, the state may need to pursue a combination of initiatives, striking the balance

between initiatives primarily focused on rates with initiatives focused on value-based payment

(VBP) and investments.

The state could build on its providers’ broad experience with Medicaid, Medicare, and

Commercial Alternative Payment Models (APMs). Missouri has significant experience with

Patient Centered Medical Homes (PCMH) and Health Homes in Medicaid, which aim to

integrate physical, behavioral, and substance use disorder (SUD) care for patients with, or at

risk of, multiple chronic conditions.

Pharmaceutical services are discussed in a separate section. The projected savings are not likely to overlap

between these sections.

Analysis of Missouri Medicaid claims data, SFY2016-18.

Analysis of Missouri Medicaid claims data, SFY2018; Interviews with Medicaid program staff and analysis of state

data, 2018.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

This section will first describe the current reimbursement methodologies and the experience

with and impact (where quantifiable) of VBP models in Missouri. Subsequently, it will highlight

the opportunities to bend the cost curve and improve the value of care for Missourians and

present an associated range of initiatives.

Current situation

This section gives an overview of Missouri’s current methodology of hospital outpatient

reimbursement, inpatient reimbursement, utilization management, out-of-state payments,

hospital tax, physician reimbursement, behavioral health reimbursement, and acute care value-

based payment initiatives.

Population served, and services provided

For Medicaid, the providers discussed in this section serve both the managed care population

(children, parents, and pregnant women) as well as those participants in the disabled population

that are not dually eligible.

Exhibit 8 shows the breakdown of the total Medicaid costs by

service for these populations.

EXHIBIT 8: BREAKDOWN OF ACUTE CARE COSTS BY SERVICE CATEGORY, SFY2018

In the managed care population, costs are driven by mental illness diagnoses (including

substance use disorders) and by perinatal care (pregnancy care, delivery, post-delivery care,

Dually eligible participants receive their acute care services through Medicare.

Analysis of Missouri Medicaid claims data, SFY2016-18; does not include beneficiaries who have no eligibility

during any given month, as well as beneficiaries who are dually eligible or have third party liability; beneficiaries

may overlap across categories.

240

Radiology

Hospital inpatient care

Office and clinic care

ED care

Hospital outpatient care

Specialty pharma

Ambulance and transportation

Lab and pathology

Prescription drugs

550

Ancillary services

DME and supplies

PT/OT/ST

Other locations

Other types of care

1,166

1,488

517

709

484

273

130

Hospital

and ED

Total spend = $ 5,683M

SFY2018, $M

Diagnostics

Other

Total acute care spend,

Share of total

acute care

spend

Average beneficiaries,

000

Share of total

acute care

beneficiaries

Total pop = 1,173K beneficiaries

383

370

139

408

762

419

753

103

262

172

130

Acute care service category

11%

10%

15%

32%

20% 12%

12%

36%

66%

26%

65%

33%

23%

35%

Office

Pharma-

ceuticals

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

and newborn care) (see Exhibit 9). In the non-dual disabled population, mental illness

diagnoses drive more than one-third of the total costs, followed by cardiovascular diagnoses

(see Exhibit 10). In these exhibits, substance use disorders (including opioid use) are included

in the mental illness diagnostic category.

EXHIBIT 9: MEDICAL COSTS BY DIAGNOSIS GROUP FOR MANAGED CARE

POPULATION, SFY2018

Analysis of Missouri Medicaid claims data, SFY2016-18.

Diseases of the musculoskeletal system and connective tissue

Injury and poisoning

Diseases of the nervous system and sense organs

Diseases of the digestive system

Mental Illness

Certain conditions originating in the perinatal period

Complications of pregnancy; childbirth; and the puerperium

Diseases of the respiratory system

Congenital anomalies

Diseases of the circulatory system

Endocrine; nutritional; and metabolic diseases and immunity disorders

Neoplasms

Other

18%

10%

11%

24%

MANAGED CARE POPULATION

Medical spend by diagnosis groups for adult and child claimants, SFY2018

Percent of medical spend

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 10: MEDICAL COSTS BY DIAGNOSIS GROUP FOR NON-DUAL DISABLED

POPULATION, SFY2018

Inpatient reimbursement

For inpatient (IP) care, Missouri uses a hospital-specific per diem, based on historical cost

reports up to two decades old. The base per diem is not differentiated by type of services

provided nor patient characteristics. In SFY2018, $540 million of inpatient payments were paid

to hospitals. In addition, add-on payments are made. $817 million “direct Medicaid” add-ons

compensate providers for differences between the base per diem and trended costs as

determined by more recent cost reports. In addition, direct Medicaid payments help offset

provider tax payments.

Other add-ons include disproportionate share hospital (DSH)

payments ($759 million) and graduate medical education (GME) payments ($139 million) (see

Exhibit 11).

The state uses a vendor to manage utilization of inpatient services. The vendor conducts six

types of reviews: prospective (pre-admission), admission (initial), continued stay review,

retrospective (post-discharge), and ongoing validation reviews. All review determinations are

made using Milliman Care Guidelines

and pertinent medical information received from the

attending physician or hospital regarding the patient's condition and planned services.

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri Foundation for Health, “Briefing Book for Missouri Medicaid,” 2016, see: mffh.org/wordpress/wp-

content/uploads/2016/04/Medicaid-Financing.pdf.

Medical spend by diagnosis groups for non-dual disabled claimants, SFY2018

Percent of medical spend

Diseases of the circulatory system

Diseases of the respiratory system

Diseases of the nervous system and sense organs

34%

13%

Mental Illness

Diseases of the musculoskeletal system and connective tissue

Injury and poisoning

Neoplasms

Endocrine; nutritional; and metabolic diseases and immunity disorders

Diseases of the digestive system

Congenital anomalies

Complications of pregnancy; childbirth; and the puerperium

Certain conditions originating in the perinatal period

Other

13%

<1%

NON-DUAL DISABLED

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

MHD covers up to 24 hours of observation services, ordered for patients who require

significant periods of treatment or monitoring before a decision on admission is made. Only

one observation code per stay can be billed, capping the reimbursable portion to 24 hours.

Outpatient reimbursement

For outpatient (OP) FFS reimbursement, Missouri pays a percentage of charges for individual

services. The percentage is based on analysis of historical cost reports trended to the current

state fiscal year. Currently, the state is transitioning towards a Medicare-based outpatient fee

schedule model. In SFY2019, hospitals received $319 million in outpatient base rate payments.

In addition, add-on payments are made to further help offset provider taxes (these are included

in the “direct Medicaid” add-ons, see Exhibit 11).

The state uses a vendor to conduct prior authorizations for advanced imaging (CT/CTA,

MRI/MRA, and PET) and select cardiac procedures (cardiac nuclear medicine and cardiac

catheterization).

EXHIBIT 11: HOSPITAL REIMBURSEMENT BREAKDOWN, MEDICAID FFS, SFY2019

Provider tax

Like other states, Missouri taxes hospitals and uses these revenues to fund Medicaid and draw

down federal funds at the Missouri federal match rate of 65.4% (see Exhibit 4). Missouri’s

hospital tax rate is higher than most other states (greater than 5.5%).

Missouri compensates

Missouri DSS, “FRA 19-3 – 10 24 18 – FINAL,” 2019; UPL analysis, Missouri DSS, 2018-19.

Missouri DSS, see note 44; uses FY18 data for outpatient base rate payments.

KFF: see note 12.

Description and methodology

Add-on payment stream

Base reimbursement stream

540

319

817

759

OP base

IP base

139

Total FFS Medicaid hospital spend

(SFY 19), USD, millions

Graduate

Medical

Education

(GME)

payments

Direct

Medicaid

payments

Direct

Medicaid

payments

Inpatient

base rate

payments

Outpatient

base rate

payments

Dispropor-

tionate

Hospital

payments

Graduate

Medical

Education

payments

▪ Per diem payments for inpatient services

▪ For a given hospital, per diems are the same regardless of diagnosis or type of care

(“single per diem”)

▪ Per diems built off cost reports from the time the hospital enrolled. They are not trended

over time, regardless of changes to patient or service mix

▪ Payments for outpatient services

▪ Rates for each hospital are calculated as a percentage of billed charges

▪ The payment percentage is based off historical cost-to-charge ratios, trended forward to

current year

▪ Outpatient rates are trended over time based on updated cost reports

▪ Payments to compensate for costs not covered by per diem (see below), including

– Compensation for FRA payments (i.e., provider tax payments)

– Payment for difference between per diem and trended costs (based on cost-reports)

▪ Decreases in inpatient base rates will increase Direct Medicaid payments to compensate

for the increased difference between per diem and trended costs

▪ Payments to compensate for costs of care to uninsured individuals, distributed proportionally

to hospitals based on total uncompensated care

▪ Payments to compensate for costs associated with offering medical education through

residency programs

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

hospitals for these Federal Reimbursement Allowance (FRA) payments through the FFS add-on

payments. To attempt to make the tax closer to budget neutral for hospitals, the Missouri

Hospital Association (MHA) operates a pooling mechanism (see Exhibit 12).

Since 2017, when managed care was implemented statewide, the state includes a portion of

add-ons (primarily to offset provider taxes) in the managed care capitation rate. Through a

Memorandum of Understanding, the MCOs and the MHA have agreed to uphold efforts to

compensate hospitals for their costs attributable to the FRA assessment.

EXHIBIT 12: MISSOURI HOSPITAL ASSOCIATION FRA FUNDING POOL

Physician reimbursement

Physicians are reimbursed through a fee schedule based on a percentage of what Medicare

pays for the same services. Once fees are set (e.g., when fees are initially calculated as a

percentage of Medicare’s rates for a certain procedure code), they are static until the state

legislature changes them. Physicians who are organized in clinics can bill the services provided

Missouri Hospital Association, “How the FRA funds are used,” 2014, see: web.mhanet.com/FRA%20Tutorial.pdf.

MHA Management Services Corporation, “MSC Health Plan MOU and Amendment,” 2017.

Missouri Hospital Association, “Missouri’s Hospital Provider Tax Pooling Arrangement,” 2016, see:

web.mhanet.com/article/4387/Missouri8217sHospital-Provider-Tax-Pooling-Arrangement.aspx?articlegroup=2663.

Missouri DSS: see note 1. KFF: see note 12.

FRA assessed at

~5.50% of each

provider’s adjusted

net revenue (both

IP and OP, Medicaid

and non-Medicaid)

Tax collection

and matching

Reimbursement to

hospitals

Hospitals are

compensated for FRA

payments primarily

through Direct

Medicaid add-on

payments

State paid federal

matching funds

at 65.40%

Provider

participates

in a MHA

pool?

Provider

keeps full

add-on

payment

Provider

receives more in

payments from FRA than

it pays in FRA (pool

contributor)?

Provider

transfers FRA-

funded

payments to

pool operated

by MHA

To net out the FRA paid with the payments received,

MHA reimburses providers that receive less in FRA-

funded payments than they pay (pool recipients)

After payment to pool recipients, MHA returns remaining

funds in central account to pool contributors

Pooling of FRA funds

Yes

Provider waits

for pooling

and

redistribution

of payments

Yes

MOU b/t MHA and MCOs: agreed to

uphold attempt to compensate individual

hospitals FRA-related gains and losses

for payments within Managed Care

Intention of

voluntary MHA

pooling arrangement

is participation of all

FRA paying

hospitals

FRA funds approximately 33% of the Missouri Medicaid program:

$1.2B collected from providers with additional $2.2B in federal match

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

through their clinics, for which reimbursement rates are generally higher than they are for

physicians.

Behavioral health reimbursement

Behavioral health services are covered by both DMH and MO HealthNet. DMH covers

Community Psychiatric Rehabilitation services, which include intake and annual evaluations,

behavioral health assessment, psychosocial rehabilitation, and day treatment for youth.

DMH

also covers Comprehensive Substance Treatment and Rehabilitation (CSTAR) services. MO

HealthNet covers other behavioral health services, such as various psychotherapy services

(e.g., family and group therapy, individual psychotherapy), services in a school setting, applied

behavioral analysis, and selected telehealth services, among other covered services.

Behavioral health services not covered by Medicaid include housing supports, drug screens,

transportation, and occupational therapy for adults. Behavioral health services are reimbursed

on a fee-for-service basis, determined by relevant information (e.g., charge information from

providers across the state, recommendations from the State Medical Consultant) and current

appropriated funds.

Value-based payment (VBP)

There is significant experience with VBP in the state, both within Medicaid and across other

payors, although many of these models are not yet fully mature or at scale. The initiatives in

Medicaid are primarily focused on PCMHs and health homes (both within the FFS and through

the managed care local community care coordination program [LCCCP]), through which

providers may receive additional payments to improve the value of the care delivered. The

impact of some initiatives can be quantified.

• Local community care coordination program (LCCCP). Missouri contracts require MCOs to

develop a LCCCP to be approved by the state in which MCOs are to develop VBP

contracts (such as ACOs, PCMHs, primary care case management programs [PCCM])

with providers. Provider participation should have reached 10% in June 2018, with 20% of

participants enrolled in the LCCCP by the end of the contract period. The program has

recently started; no results are yet available.

• Accountable care organizations (ACOs). In addition to the LCCCP initiatives, there are at

least 13 ACOs in Missouri, concentrated in St. Louis, Kansas City and Springfield, of which

11 are Medicare ACOs, and two are commercial ACOs. The six Medicare ACOs for which

the number of participants has been published jointly serve >184,000 Missourians. At least

three of these ACOs have risk-based contracts; one reported $8.9 million in earned

savings in 2017.

• Patient-centered medical home (PCMH). There are currently 419 NCQA accredited

PCMHs in Missouri,

contracting with MCOs through the LCCCP program and with

commercial plans. Fifty-three practices participate in CMS’ Medicare Comprehensive

Primary Care Plus Initiative (CPC+), in which Blue Cross Blue Shield of Kansas City and

MO HealthNet Community Psych Rehab Program Manual.

http://manuals.momed.com/collections/collection_cpr/print.pdf.

MO HealthNet Behavioral Health Services Manual. http://manuals.momed.com/collections/collection_psy/print.pdf.

CMS, “Shared Savings Program Accountable Care Organizations (ACO) Public-Use Files,” 2017, see:

www.cms.gov/Research-Statistics-Data-and-Systems/Downloadable-Public-Use-Files/SSPACO/index.html.

National Committee for Quality Assurance, “Practices,” see: reportcards.ncqa.org/#/practices/list?state=Missouri.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

UnitedHealth also participate; these practices receive care management fees and quality-

and efficiency bonus payments. No Missouri-specific results have been published.

• Health homes. Missouri was one of the first states to create health homes. Health homes

must meet specific quality criteria and receive a per member per month (PMPM) payment

for care management and other dedicated health home services. Primary care health

homes (PCHH) focus on patients with at least two physical chronic conditions, such as

diabetes, cardiovascular disease, or substance use disorder. As of 2017, PCHHs received

$63.72 PMPM for health home services; of the 38 PCHHs, 25 are federally qualified health

centers (FQHCs), 11 are hospital affiliated providers, and two are clinics. In 2018, 24,580

Medicaid participants were receiving care from PCHHs. According to evaluations published

by the state, PCHHs saved $98.35 PMPM, compared to baseline in 2016 (see Exhibit

13).

Lower actual and risk-adjusted PMPM costs for the PCHH population are partially

driven by lower inpatient costs (see Exhibit 14).

EXHIBIT 13: IMPACT OF PRIMARY CARE HEALTH HOMES: HOSPITAL USE, SFY2012-18

Missouri DSS, “Paving the way,” 2017, see: dss.mo.gov/mhd/cs/health-homes/pdf/pchh-paving-the-way.pptx;

Missouri DSS, “MO HealthNet PCHH Progress Report 2014-2017 FINAL 07192018,” 2018, analysis of Missouri

Medicaid claims data, SFY2016-18.

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri DSS: see note 54.

Percentage of PCHH enrolees who had a hospitalization, %Percentage of PCHH enrollees who had an ED visit, %

Yr 5Yr 2 Yr 3BL Yr 6Yr 4Yr 1

Yr 3Yr 1 Yr 4BL Yr 2 Yr 5 Yr 6

4.7

0.5

BL Yr 1 Yr 2 Yr 6Yr 3 Yr 4 Yr 5

1.5

0.8

0.5

0.4

0.3

0.2

Yr 3BL Yr 1 Yr 2 Yr 5Yr 4 Yr 6

In total, the percentage

of high utilizers who are

admitted to the hospital

has been reduced by

86%.

The average number of

hospitalizations has

decreased by 87 % from

baseline to year six.

The impact of PCHH on ED and hospital use has been especially effective among high utilizers

There has been an 86%

decrease in ED visits for

individuals who are

considered to have high ED

or hospital utilization.

The average number of ED

visits decreased from 4.7

visits per person to less than

one visit/person by year six,

an 89% decrease.

% of high utilizers with ED visits % of high utilizers with hospital admissions

Average # of ED visits for high utilizers Average # of hospitalizations for high utilizers

Yr 4Yr 1BL Yr 3Yr 2 Yr 5 Yr 6

Yr 4Yr 3BL Yr 6Yr 1 Yr 2 Yr 5

There has been a 35%

decrease in ED use for all

PCHH enrollees from

baseline, through year 6 of

the PCHH program

There has been a 20%

decrease in hospital use

from the baseline,

through year six of the

program.

Utilization results across all PCHH enrollees

Utilization results across high utilizers

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 14: COMPARISON OF EXPENDITURES IN PCHH PROGRAM WITH NON-PCHH

PARTICIPANTS WITH COMPARABLE RISK-PROFILE (MANAGED CARE), SFY2018

Community mental health center health homes (CMHC HH) focus on patients with (serious)

mental illness and/or substance use disorder. CMHCs receive $85.23 PMPM to support the

infrastructure needed to deliver CMHC HH services; of the 28 CMHCs, 22 are clinics and six

are hospital affiliated providers (15 of these CMHCs have become certified community

behavioral health clinics [CCBHCs; see below]). As of January 2017, 24,844 participants

were enrolled in CMHC HH. An evaluation by the state concluded that in 2016, CMHC HHs

saved $284.94 PMPM compared to baseline (see Exhibit 15 for additional results).

compare participants served by CMHC HHs with participants with comparable conditions

and co-morbidities, individuals with high behavioral health needs were identified within the

CMHC HH population as well as in the non-HH population.

In this comparison, participants

in the non-dual disabled population show similar nominal PMPM costs but lower risk-

adjusted PMPM costs for the CMHC population. As in the PCHH analyses, these results

were driven partially by higher pharmacy costs and lower inpatient costs in the CMHC

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri DMH, “DMH CMHC Healthcare Homes progress report,” 2016, see:

dmh.mo.gov/mentalillness/provider/docs/cmhchchprogreport16.pdf.

In these analyses, participants were flagged as having high behavioral health needs if they either (1) have

diagnoses of schizophrenia, bipolar disorder with psychosis, major depression w/ psychosis, attempted suicide or

self-injury, homicidal ideation, or substance use with pregnancy or one year postpartum OR (2) have one or more

behavioral health-related utilization of inpatient hospital visit, crisis unit visit, residential facility visit, rehab facility

visit, medication-assisted treatment, ED visit, or injection antipsychotics AND presence of bipolar disorder without

psychosis, major depression without psychosis, other depression, PTSD, substance use, conduct disorder,

personality disorder, psychosis, ODD, or eating disorders.

SFY2018

10%

13%

20%

16%

13%

34%

23%

5% 75M

1,447M

In PCHH

program

Not in PCHH

program

880

1,194

452

618

100,951

7,065

Ambulance and transportation Home and community-based

Other types of care

Ancillary services

DME and supplies

ED care

Hospital inpatient care

Hospital outpatient care

Specialty pharmaInstitutional care Treatment and evaluation at other sites

Lab and pathology

Office and clinic care

Physical, occupational, speech therapy

Radiology

Retail prescriptions

MANAGED CARE POPULATION

Spend by service category by HH enrollees, vs. comparable population

Total

members

included Breakdown of total spend

Risk-

adjusted

PMPM, $

Program

type

Actual

PMPM, $

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

population. This comparison could indicate that the CMHC is cost-effective, improving the

care for these patients (including utilization of needed drugs) and reducing potentially

avoidable hospital admissions (see Exhibit 16). (The results for the managed care

population, mostly children, did not show a comparable difference.)

EXHIBIT 15: IMPACT OF COMMUNITY MENTAL HEALTH CENTER HEALTH HOMES ON

HOSPITAL USE, SFY2011-2015

Missouri DMH: see note 58.

34%

28%

26%

22%

25%

45%

40%

34%

31%

Yr 4BL Yr 1 Yr 2 Yr 3

Hospital ER

2.32

2.09

1.90

0.73

0.56

0.51

0.42

0.33

2011 14

12 13 2015

0.5

2.5

1.0

1.5

2.0

Avg # of ER visits/Enrollee/Year

Avg # of Hospitalizations/Enrollee/Year

Percent of enrollees with one or more hospitalization or

ER visit

Average ER and hospitalizations per enrollee per

calendar year

Average number of hospitalizations has been reduced 14%, and average emergency room visits decreased 19%

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 16: COMPARISON OF PARTICIPANTS WITH SEVERE BEHAVIORAL HEALTH

NEEDS IN CMHC PROGRAM WITH NON-CMHC PARTICIPANTS WITH SEVERAL

BEHAVIORAL HEALTH NEEDS (NON-DUAL DISABLED POPULATION), SFY2018

Certified Community Behavioral Health Clinics (CCBHC), a new initiative started in 2017,

focus on a similar patient population as CMHCs, and can also provide Health Home

services. Of the 28 CMHCs mentioned above, 14 have become CCBHCs (and one new

CCBHC has been created). As of November 2018, CCBHCs HHs served ~16,650 Medicaid

participants (largely participants who were enrolled in CMHC HHs before).

CCBHCs are

reimbursed using a prospective payment system, in which health home payments are

included. No results have yet been published.

– Bundled Payments. There are currently seven participating healthcare facilities in CMS’

Medicare Bundled Payment for Care Improvement (BPCI) Advanced program, which have

selected between one and 19 episodes, including sepsis, hip/knee replacement, and spinal

fusion surgery.

Additionally, 36 hospitals participated in CMS’ Medicare Comprehensive

Care for Joint Replacement (CJR) program in 2016-2017, earning savings of on average

~$2 million.

Excludes health home PMPM payments; analysis of Missouri Medicaid claims data, SFY2016-18.

Interviews with Medicaid program staff.

CMS, “BPCI Advanced,” 2018, see: innovation.cms.gov/initiatives/bpci-advanced/.

CMS, “Comprehensive Care for Joint Replacement Model,” 2018, see: innovation.cms.gov/initiatives/cjr.

SFY2018

17%

25%

14%

22%

14%

30%

23%

591M

4% 65M5%

High acuity

population in

CMHC

program

High acuity

population

not in CMHC

program

3,449

3,379

816

1,034

1,582

14,579

Ambulance and transportation

Hospital outpatient careDME and supplies

Treatment and evaluation at other sitesED care

Ancillary services

Physical, occupational, speech therapy

Office and clinic care

Home and community-based

Hospital inpatient care

Institutional care

Lab and pathology

Other types of care

Radiology

Retail prescriptions

Specialty pharma

Spend by service category by HH enrollees, vs. comparable population

Total

members

included Total spend

Risk-

adjusted

PMPM, $

Program

type

Actual

PMPM, $

NON-DUAL DISABLED

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Durable Medical Equipment

• The state follows a CMS fee schedule for most DME products (with exceptions such as

speech generation software and accessories, and certain types of hospital beds, which use

negotiated rates on a case-by-case basis). Total SFY2018 costs were $52 million, of which

27% were for respiratory DME, 27% were for bulky DME (such as wheelchairs and lift

chairs), 14% were for orthotics and prosthetics, and the remainder were for other DME

categories such as incontinence products and infusions.

• Utilization management techniques are in place for DME products. Prior authorization,

precertification, or meeting medical necessity criteria is required for most high-cost/high-

utilization products such as power wheelchairs and other bulky DME.

Potential opportunities for improvement

This section identifies potential opportunities to improve Missouri’s current approach to hospital

and physician reimbursement, as well as regulatory and stakeholder risks. The opportunities are

not intended to be mutually exclusive: opportunities for savings or improved outcomes may

overlap.

• The cost-based, single per diem payment method for hospital inpatient care

provides limited incentives to contain costs and improve quality. An outlier among

states, Missouri’s use of a single per diem lacks a direct connection between payments,

actual care provided, and types of patients served. This lack of incentive for efficiency is

exacerbated by the tight coupling between the reported cost of delivery and the level of the

per diem: increased costs lead to higher per diems. Most state Medicaid programs (similar

to Medicare and most commercial health insurers) currently use Diagnosis-Related Groups

(DRGs), which make a fixed payment for the entire stay in the hospital, creating cost-

containment incentives while also accounting for patient mix and severity.

• There is no inpatient readmissions policy. While Missouri spent $160 million on hospital

readmissions in SFY2018, it has no policies in place to address potentially avoidable

readmissions. CMS, MCOs, and other state Medicaid agencies have extensive experience

with such policies.

• The cost-based, outpatient payment method also contains limited incentives for

cost containment. Proactive outreach to avoid exacerbations of depression or

inefficiencies in diabetes care is not part of the standard fee schedule; in fact, reducing ER

visits or hospital (re-)admissions reduces revenue for providers in a FFS payment

system.

In addition, providers tend to have no access to data about the overall costs and

outcomes of the care they provide, making it difficult to fully mobilize to prevent potentially

avoidable complications and inefficiencies.

Most states use outpatient fee schedules that are indexed to Medicare’s fee schedule or

ambulatory payment group models.

Missouri has started to move towards a comparable,

Analysis of Missouri Medicaid claims data, SFY2016-18.

MACPAC, “State Medicaid Payment Policies for Inpatient Hospital Services,” 2018, see:

www.macpac.gov/publication/macpac-inpatient-hospital-ayment-landscapes.

Krupka, DC et al., “The Impact On Hospitals Of Reducing Surgical Complications Suggests Many Will Need

Shared Savings Programs With Payers,” 2012, Health Affairs, 31: 2571-78.

Such as Enhanced Ambulatory Patient Groups (EAPG) or Ambulatory Payment Classification (APC)

methodologies.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Medicare-based outpatient fee schedule model, but has not yet completed that transition.

For SFY2020, the state currently estimates this transition to be budget-neutral.

Expansion of this approach could generate savings over time as hospitals improve

operational efficiencies. Savings could be realized more quickly depending on the

approach taken to setting prices under the new fee schedule.

• Several categories of high-cost outpatient services do not require prior

authorization. Prior authorization (PA) is limited to advanced imaging and select cardiac

procedures. Other states and MCOs incorporate measures to ensure appropriate utilization

on other OP procedures such as sleep studies, radiation therapy, and arthroscopies.

• Providing add-on payments to hospitals for non-Missouri residents served is a

unique feature of the Missouri Medicaid program. Throughout the U.S., hospitals

serving out-of-state Medicaid patients will be paid by the patient’s home state according to

that state’s Medicaid regulations. In Missouri, the state provides additional add-on

payments (estimated at approximately $177 million in SFY2019) to its hospitals for

services provided to persons eligible for Medicaid from Kansas, Illinois or elsewhere.

• Managed care payments to hospitals are set at a higher rate than FFS payments.

Excluding compensation for provider taxes, current MCO inpatient base payments are

approximately 30% higher than FFS per diem payments.

An estimated >$100 million of

MCO payments to hospitals are at rates above 120-130% of FFS payments.

• Variability in reimbursement levels between hospitals is significant. Excluding the

Medicaid portion of each hospital’s provider tax assessment, the difference between the

Medicaid payments hospitals received and their individual UPLs varied between <50% and

>150% of their hospital-specific UPLs in SFY2016.

The variation in outpatient procedure

fees is currently being reduced through the introduction of the Medicare-based outpatient

fee schedule.

• Physician reimbursement is lower than in most other states. Physicians are paid

based on a fee schedule that is historically linked to Medicare but is not regularly updated.

Reimbursement rates are less than in other states: Missouri Medicaid pays 79% of the

national average (ranked 46th); for primary care, the state pays 81% (ranked 42th).

Compared to Medicare fees, Medicaid pays 60% for overall physician services (ranked

44th) and 55% for primary care services (ranked 41th).

Spending on non-hospital

physician services, including Federally Qualified Health Centers (FQHCs), clinics, and rural

health services, is lower than other comparable states: In SFY2016, Missouri spent 5% of

Missouri DSS, “MHD – FY 19 Core exercise,” 2018.

Third-party analysis provided by MHD, 2018.

Missouri DSS, “20181119 Medicaid Update,” 2018.

Missouri DSS, FY2016 UPL analyses; AHA Hospital Cost Report Files (HCRIS): CMS.Gov Case Mix Index Data;

Missouri DSS, FY2016 FRA schedule; FRA share of Medicaid revenue calculated by multiplying each hospital’s

FY2016 Medicaid revenue, the FRA tax rate (5.95% in FY2016), and the percentage of FRA payments from FFS

(91%, per the FRA schedule).

KFF, “Medicaid physician fees,” 2016, see: www.kff.org/state-category/medicaid-chip/medicaid-physician-fees/.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

total expenses on non-hospital physician services, as opposed to 9% in comparable

states.

• >15% of Missouri Medicaid acute care expenditures may be associated with

potentially avoidable exacerbations and complications (PECs). A PEC is any event

that negatively affects the patient and is potentially controllable by the health care delivery

system: an ER visit for an asthma exacerbation, a hospital readmission for a post-surgical

wound infection, or an emergency admission for a patient with a depression. PECs are an

inherent part of health care: a patient with bronchitis can develop a pneumonia, and post-

surgical complications will likely never be completely eradicated. But improving the

coordination and quality of care can significantly reduce the volume and costs of PECs.

As mentioned earlier, the current FFS reimbursement method does not reward

coordination of care or adequate care management. Likewise, preventing PECs tends to

negatively impact provider economics. Reducing such events, however, is an important

source of value for payors and patients alike: addressing PECs means reducing total costs

of care through improving outcomes for patients.

In Exhibit 17, the risk-adjusted total costs of care for the Medicaid managed care

population

are shown per county and mapped against the percentage of total costs that

are associated with PECs. The percentage of costs associated with PECs per county is

highly variable, ranging from <10% to >23%. (For persons with disabilities, the variability is

comparable; percentages range between 6% and 14%.)

Reducing PECs by 20% would

amount to ~$170 million in savings or opportunities for reinvestment.

FMR, 2016; comparable states are other states with enrollment in managed care of 30% or less (Alaska,

Arkansas, Alabama, Colorado, Connecticut, Idaho, Massachusetts, Montana, North Carolina, North Dakota,

Oklahoma, South Dakota, Vermont, and Wyoming).

De Brantes, F, A Rastogi, and M Painter, “Reducing potentially avoidable complications in patients with chronic

diseases,” 2010, Health Serv Res, 45: 1854-71.

Excluding dually eligible beneficiaries.

PECs percentages are lower for this population as a larger proportion of the spending is LTSS spending, which is

not included in PECs.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 17: POTENTIALLY AVOIDABLE EXACERBATIONS AND COMPLICATIONS AND

RISK ADJUSTED SPENDING BY COUNTY, MANAGED CARE POPULATION

There are several ways value-based payment could support reducing PECs: incentivizing high

quality, integrated primary care; rewarding a focus on high-cost patients who cycle in and out

of ERs and hospitals (for persons with disabilities, 4% of participants account for nearly one-

third of all ER visits) (see Exhibit 18); and strengthening the role of behavioral health care

throughout the care cycle (mental health and substance use are the main reasons for hospital

admissions amongst the individuals with disabilities, and – after maternal and newborn care –

the second main reason in the children and adults population). See Exhibit 19 for the

admissions for mental health diagnoses in the non-dual disabled population.

Analysis of Missouri Medicaid claims data, SFY2016-18; excludes counties <15,000 member-months. “Managed

Care Population” refers to those individuals eligible for managed care: children, parents, and pregnant women.

Analysis of Missouri Medicaid claims data, SFY2016-18.

7% 8% 9% 10% 11% 12% 13% 14% 15% 16% 17% 18% 19% 20% 21% 22% 23% 24% 25% 26% 27%

% Total cost associated with PECs

Better

performance

260

280

300

320

340

360

380

400

420

440

460

480

500

520

540

560

580

600

620

640

Risk adjusted

PMPM

County, bubble size

represents total $ paid

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 18: 4% OF MANAGED CARE PARTICIPANTS ACCOUNT FOR ALMOST ONE-

THIRD OF ED VISITS, OF WHICH >50% ARE POTENTIALLY AVOIDABLE

EXHIBIT 19: HOSPITAL ADMISSIONS IN THE NON-DUAL DISABLED POPULATION

Analysis of Missouri Medicaid claims data, SFY2016-18.

Analysis of Missouri Medicaid claims data, SFY2016-18; excludes top and bottom 2.5% of episodes by cost.

5 visits

4 visits

0 visit 561

1 visit

2 visits

3 visits

6+ visits

148

Members

(K)

Grand

total:

841 K 621 K

8973

8060

ED visits

(K)

SFY2018

1,150

156

256

661

406

106

285

$2,860 M

171

372

521

694

849

1,065

1,783

288

456

545

618

664

730

810

Spend PMPM ($)

ED visit

count

Risk adj spend

PMPM ($)

Total medical

spend

(M)

$299 $369

Potentially avoidable

MANAGED CARE POPULATION

58%

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

• 5-10% of Missouri Medicaid acute care expenditures may be associated with the

location where services are provided and the choice of diagnostics and

interventions. FFS does not incentivize efficiency considerations in making diagnostic or

therapeutic decisions, nor does it stimulate providers to select the most cost-effective

location to perform these services. Serving people in the ER is costly compared to serving

them in a doctor’s office; opting for an MRI scan where a CT scan could suffice is similarly

inefficient. The risk-adjusted variation in spending per county is ~100%, driven partially by

differences in hospital admissions that do not appear to be due to differences in overall

differences in risk score per county (see Exhibit 20 and Exhibit 21). In general, across all

payors, Missouri’s hospital utilization is high compared to that of other states (see Exhibit

22). Analyzing spending per episode of care shows similar variability in expenditures:

Exhibit 23 illustrates that perinatal care costs also vary ~100% for perinatal care between

high-volume zip codes.

As with PECs, some efficiencies may be unavoidable: the MRI

may simply be available faster, or the primary care practice – as an alternative to the ER –

might be closed. Yet reducing these inefficiencies tends to be feasible. Reducing these

inefficiencies by 20%, without negatively impacting the quality of care, would amount to

$55 million to $110 million in savings or opportunities for reinvestment.

EXHIBIT 20: RISK ADJUSTED SPENDING BY COUNTY, MANAGED CARE, SFY2018

Analysis of Missouri Medicaid claims data, SFY2016-18. Some of these admissions and costs may be due to

PECs.

Calculation: 20% of 5-10% of $5.7 billion (total acute care spending) = $55-110 million (rounded).

Analysis of Missouri Medicaid claims data, SFY2016-18; not calculated for counties with fewer than 1,000

managed care-eligible Medicaid-enrolled residents (“N/A”).

MANAGED CARE POPULATION

Total cost of care PMPM by county, SFY2018

Lowest quartile (< $326)

2nd quartile ($326 – 347)

3rd quartile ($347 – 376)

Highest quartile (> $376)

N/A

Cape

Girard-

eau

Audrain

Barry

Bates

Benton

Bolli-

nger

Boone

Butler

Callaway

Camden

Carter

Cass

Cedar

Chariton

Christian

Clay

Cole

Cooper

Craw

ford

Dade

Dallas

Dent

Franklin

Gas-

conade

Greene

Henry

Hickory

Howard

Howell

Iron

Jackson

Jasper

Jefferson

Johnson

Laclede

Lafayette

Lawr-

ence

Lincoln

Mcdonald

Madison

Maries

Miller

Mississippi

Moniteau

Mont

gomery

Morgan

New Madrid

Newton

Oregon

Osage

Ozark

Perry

Pettis

Phelps

Pike

Platte

Polk

Pulaski

Ralls

Ray

Reynolds

Ripley

St. Charles

St. Clair

Ste. Genevieve

St.

Francois

St. Louis

Saline

Scott

Shannon

Stoddard

Stone

Taney

Texas

Vernon

Warren

Washin-

gton

Wright

St. Louis City

Adair

Andrew

Atchison

Caldwell

Clark

Daviess

Kalb

Grundy

Harrison

Lewis

Linn

Livingston

Macon

Marion

Mercer

Nodaway

Scotland

Shelby

Sullivan

Buchanan

Web-

ster

Carroll

Douglas

Wayne

Barton

Monroe

Ran-

dolph

Gentry

Worth

Clinton

Holt

Knox

Putnam

Schuyler

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 21: INPATIENT ADMISSIONS BY COUNTY, MANAGED CARE, SFY2018

EXHIBIT 22. HOSPITAL USE COMPARED TO OTHER STATES (ALL-PAYORS)

Analysis of Missouri Medicaid claims data, SFY2016-18; not calculated for counties with fewer than 1,000

managed care-eligible Medicaid-enrolled residents (“N/A”).

KFF, “Hospital Outpatient Visits per 1,000 Population by Ownership Type,” 2016, see: www.kff.org/other/state-

indicator/outpatient-visits-by-ownership.

MANAGED CARE POPULATION

Inpatient admissions by county, SFY2018

Lowest quartile (<102 per 1,000)

2nd quartile (102 – 112 per 1,000)

3rd quartile (113 – 120 per 1,000)

Highest quartile (> 120 per 1,000)

N/A

Audrain

Barry

Bates

Benton

Bolli-

nger

Boone

Butler

Callaway

Camden

Cape

Girard-

eau

Carter

Cass

Cedar

Chariton

Christian

Clay

Cole

Cooper

Craw

ford

Dade

Dallas

Dent

Franklin

Gasconade

Greene

Henry

Hickory

Howard

Howell

Iron

Jackson

Jasper

Jefferson

Johnson

Laclede

Lafayette

Lawr-

ence

Lincoln

Mcdonald

Madison

Maries

Miller

Moniteau

Mont

gomery

Morgan

Newton

Oregon

Osage

Ozark

Pemiscot

Perry

Pettis

Phelps

Pike

Platte

Polk

Pulaski

Ralls

Ray

Reynolds

Ripley

St. Charles

St. Clair

St.

Francois

St. Louis

Saline

Scott

Shannon

Stoddard

Stone

Taney

Texas

Vernon

Warren

Washin-

gton

Wright

St. Louis City

Adair

Andrew

Atchison

Caldwell

Clark

Daviess

De Kalb

Grundy

Harrison

Holt

Lewis

Linn

Livingston

Macon

Marion

Mercer

Nodaway

Putnam

Schuyler

Scotland

Shelby

Sullivan

Buchanan

Web-

ster

Ran-

dolph

Ste. Genevieve

Mississippi

New Madrid

Dunklin

Carroll

Clinton

Gentry

Worth

Knox

Monroe

Barton

Douglas

Wayne

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 23: VARIATION IN HIGH-SPEND EPISODES: PERINATAL, SFY2018

• There is little to no transparency of outcomes of care in Medicaid. Available data

(e.g., external quality reviews of the MHD managed care program)

covers a limited range

of performance measures. There is no readily publicly accessible information about the

outcomes of care delivered per (sub)population or condition and per (groups of) provider.

This limits consumer choice, accountability, and the opportunity and incentive for provider

self-improvement.

• The incentives embedded in several existing programs can be made stronger and

aimed more explicitly at the outcomes of care that matter most to participants. There

are opportunities to build upon the success of the primary care initiatives, Accountable

Care Organizations (ACOs), bundled payments, and health homes: increasingly link

upside incentives to the outcomes of care, and tie the rewards received to the amount of

savings realized. By reducing PECs by 1 percentage point in a Missouri managed care

population, a PCMH, health home or ACO with 50,000 attributed lives could receive $1.5

million

in savings shared to further invest in improving their care (assuming 50% shared

savings). Sharing in the savings could also help these providers to focus even more on the

social determinants that may drive up PECs.

Analysis of Missouri Medicaid claims data, SFY2016-18; figures are based on total claims-based expenditure

associated with a perinatal episode triggered by live birth diagnosis and delivery procedure code. The top and

bottom 2.5% of episodes by cost were excluded in analyses of variation.

Behavioral Health Concepts, “2016 MO HealthNet Managed Care Program External Quality Review: Report of

Findings,” 2016, see: dss.mo.gov/mhd/mc/pdf/2016-external-quality-review-report-mohealthnet-managed-care.pdf.

Estimate based on a 500 PMPM average spending of which 18% is associated with PECs.

10,000

1,000

14,000

2,000

3,000

6,000

4,000

5,000

7,000

8,000

9,000

11,000

12,000

13,000

15,000

16,000

Each vertical bar represents a zip code

High

avg.

episode

cost ($)

Low avg.

episode

cost ($)

MANAGED CARE POPULATION

Average episode cost by member zip code for all zip codes

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

• There are physician shortages in many parts of the state, particularly rural areas.

80% of Missouri counties are considered physician shortage areas, and only 10% of new

physicians enter rural primary care. At 267, the state falls below the national average

physician to patient ratio of 272 physicians per 100,000 people; for primary care, the state

has 87 PCPs per 100,000 people, compared to 91.7 nationally. In rural areas, these issues

are particularly challenging: of the 101 rural counties, 99 are Primary Medical Health

Professional Shortage Area (HPSAs), 97 are Mental Health HPSAs, and 95 are Dental

HPSAs.

This may contribute to system inefficiencies and the incidence of PECs: these

services tend to be key to avoiding PECs and can lead to institutional care when

community care might have been preferable. Creating innovative delivery or

reimbursement models fitting the challenges of rural healthcare is difficult within the limits

of the FFS fee schedules.

• The financial viability of many rural and safety net providers is precarious. Some

rural and safety net providers are financially frail, with year over year negative results.

Approximately 90% of safety net hospitals and ~60% of rural hospitals had negative

margins in SFY2016.

Without a rural health care infrastructure that is viable and meets

local community needs, access to care for rural Missourians could be threatened. This

could in turn lead to higher downstream costs due to missed (secondary) prevention

opportunities.

In addition to the previously outlined potential opportunities, there are regulatory and

stakeholder challenges which may impact current reimbursement approaches:

• Recent changes in CMS IP Upper Payment Limit (UPL) calculations may result in inpatient

payments exceeding the UPL and thus a corresponding loss of federal funds. UPLs limit

state Medicaid FFS spending on specific provider classes (e.g., hospital inpatient) to what

Medicare would have paid for these services. For IP, both FFS base payments, out-of-

state payments, and add-on payments count against the IP UPL. FFS payments that

exceed the UPL are not eligible for federal match. CMS has recently introduced a new

template to calculate the inpatient UPL, leading to Missouri’s inpatient payments to

possibly exceed the UPL by $16 million.

• Missouri’s tax rate is currently within the federal safe harbor limit (6%), but regulatory

scrutiny of exact mechanisms used to compensate hospitals for tax payments could

increase. Alternatively, CMS may reduce the current 6% provider tax limit below which it

has so far allowed comparable payment arrangements in several states to e.g. 5% or even

3%.

• Provider tax compensation arrangements in FFS and managed care are under pressure as

hospitals that are net contributors to pooling mechanisms may opt out.

Under the

existing MHA pooling mechanism, participating providers who receive more in estimated

Association of American Medical Colleges, “2017 State Physician Workforce Data Report,” 2017, see:

members.aamc.org/eweb/upload/2017%20State%20Physician%20Workforce%20Data%20Report.pdf.

Missouri DHSS, “Health in Rural Missouri Biennial Report 2016-2017,” 2017, see:

health.mo.gov/living/families/ruralhealth/pdf/biennial2017.pdf.

AHA Hospital Cost Report Files (HCRIS); CMS.Gov Case Mix Index Data.

UPL calculation, Missouri DSS, January 2019.

KFF: see note 12.

MHA: see note 49.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

FRA-related add-on payments than their provider tax payments contribute to the pool so

that other hospitals can be compensated. If pool contributors withdraw from the voluntary

transfers, the pool would become less able to compensate pool recipients, thus

exacerbating the concern of some hospitals with the provider tax.

• In SFY2016, Medicaid made $19.7 billion in DSH payments nationally ($8.5 billion in state

funds and $11.2 billion in federal funds). A reduction of $4 billion is planned for 2020, with

the reduction increasing to $8 billion for each year from 2021 to 2025.

This could lead to

a substantial reduction in available federal DSH funding for the state.

• CMS may recoup parts of DSH payments made to hospitals from previous years’

allotments. Many states and hospitals have operated under the assumption that third-party

payments did not have to be included in Medicaid DSH payment calculations and audits. In

2017, however, CMS issued a final rule stating that inclusion was needed, which has since

been contested in several courts. If CMS prevails, this could lower hospital-specific DSH

limits, creating, for example, a potential risk for Missouri hospitals of $96 million from 2011

and 2012 allotments alone.

• Missouri’s DME rates are higher than CMS’ DME fee schedule across most product

categories. Commercial payors, MCOs, and some state Medicaid agencies set DME rates

significantly below CMS’ DME rates (between approximately 65% to 75%); Missouri’s DME

fee schedule is currently priced at over 100% of CMS’ DME rates. The variance between

current pricing and pricing at 70% of CMS’ DME rates is >$10 million.

Potential initiatives

The following is a wide range of potential initiatives that Missouri Medicaid may consider, either

in combination or as alternatives for improving the financial sustainability of the program. In

total, the gross impact of the hospital, physician and behavioral health reimbursement initiatives

outlined below could range from $250 million to $500 million, depending on choices made by

the state. This excludes the impact on provider tax revenues (see the section on federal

financing).

Potential initiatives to improve incentives and reduce costs include adjusting the inpatient (IP)

and outpatient (OP) base rate methodologies and the add-on payments for out-of-state patients.

Following other states, Medicare and commercial plans, Missouri could also consider

transitioning further to value-based payment models and transparency of care costs and

outcomes, which would maximize incentives for providers to deliver high-quality care while

lowering costs. Through the latter, the state could work to address the significant costs

associated with potentially avoidable exacerbations and complications (PECs) as well as other

inefficiencies. In addition, value-based payment models could facilitate investments in the rural

and safety net heath care infrastructure, including primary and behavioral health.

To achieve the higher end of the estimated impact range, the state would likely need to combine

a focus on adjusting hospital reimbursement rates and utilization management with a broader

MACPAC, “Disproportionate share hospital payments,” see: www.macpac.gov/subtopic/disproportionate-share-

hospital-payments.

Modern Healthcare, “CMS appeals Missouri court decision stopping DSH clawback,” 2018, see:

www.modernhealthcare.com/article/20180416/NEWS/180419941.

Analysis of Missouri Medicaid claims data, SFY2016-18; CMS, “DME18-A,” 2018, see:

www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/DMEPOSFeeSched/DMEPOS-Fee-Schedule-

Items/DME18-A.htm.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

value-based payment program in which providers could share in the savings realized across the

total cost of care or in episodes of care. The state could choose its preferred balance between

these approaches, which would imply choosing between those initiatives that address similar

opportunities in different ways. The list below is intended as an outline of potential options for

the state, providing the state with the opportunity to decide on both sizing and timing of the

potential implementation of a selection of initiatives.

1. Implement an inpatient hospital readmissions policy. Inpatient hospital readmission

policies are used by commercial payors, MCOs, other state Medicaid programs, and CMS

to not only ensure appropriate utilization of services but also to improve quality. This policy

could be modeled after policies that MCOs have today and further refined by the state.

This could help ensure safe and appropriate discharge of participants and would also

provide important feedback to hospitals. Operationally, this initiative would require modest

policy and MMIS changes.

2. Expand prior authorization (PA) to additional outpatient procedures. PA policies are

likewise used by commercial and other payors. This initiative could add select categories

to the current PA list, and it could potentially make changes to the approach used in the

existing outpatient PA process. This may require additional system edits and updates to

current vendor contracts.

3. Adjust outpatient base rate methodology. Missouri could consider further anchoring

outpatient base rate payments to a percentage of Medicare fee schedule rather than a

percentage of charges across all outpatient services. This could allow Missouri to improve

alignment between payments and services provided, increase predictability of outpatient

expenditures, and be better able to compare rates both within the state and with other

states. In addition, the Medicare fee schedule evolves with changes in the science and

practice of medicine, thus ensuring the appropriateness of the payment methodology over

time. As this transition has already been set in motion, the implementation complexity of this

initiative would be limited.

4. Adjust inpatient base rate per diem methodology. To increase provider incentive for cost

containment, Missouri could adopt a stratified per diem for inpatient services, offering

different per diem rates for different types of patients (e.g., medical, surgical, maternity,

neonatal, psych). Rates could be set in one of two ways: 1) based on current payment levels

using a state-set trend factor, which would build off current price-setting methodology; or 2)

based on regional average costs for each per diem category. Both approaches would likely

improve alignment between payments and services offered, but they would not maximize

cost containment incentives given the pay-per-service setup. The second approach, basing

rates on regional average costs, may better improve alignment between payments and

services as it eliminates link to historical costs. But additional risk adjustment would likely be

necessary to capture within-region variations across hospital types (e.g., safety net hospitals

may not incur the same costs as non-safety net). While it would not be challenging to

implement the change from a regulatory respective, redesigning the per diem methodology

– including ensuring a smooth transition without disruptive impact on reimbursements of

individual hospitals – is not a well-standardized approach and is likely to be complex from a

technical and operational standpoint.

5. Consider case rate methodology for inpatient and/or outpatient services. Missouri

could move away from per diems and payments for individual outpatient services toward a

case rate-based reimbursement model. Such models employ a grouping mechanism that

varies for inpatient and outpatient services and are in use in many other states. For

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

outpatient, widely used grouper options are Enhanced Ambulatory Patient Groups (EAPG)

and Ambulatory Payment Classification (APC). For inpatient, the standard is Diagnosis-

Related Groups (DRG). Like the stratified per diem method above, pricing could be based

on regional average costs or historical pricing with forward-looking trend factors set by the

Medicaid program. Although payments are no longer determined at the individual service

level, this payment methodology would still be volume-focused and hence would still limit

cost containment incentives. The implementation complexity will likely be significant: in

particular, the change from single inpatient per diem payments will require a thorough

rebasing effort so that the transition is within the planned inpatient expenditures, remains

predictable, includes the needed add-ons, and does not create financial disruptions for

individual providers. In addition, the current MMIS is not currently equipped to handle case

rate-based reimbursement models. Workarounds through additional DRG grouping

applications exist, but these would have to handle all payments to hospitals.

6. Reevaluate add-on payments for out-of-state (non-MO) residents. Missouri could

reduce or eliminate the reimbursements it makes to hospitals for treating out-of-state

patients. Out-of-state payments are concentrated in a limited set of hospitals. The technical

implementation complexity of this initiative is likely to be low, but the impact on affected

providers may be significant.

7. Modify Direct Medicaid payments methodology. The Direct Medicaid payments (one

component of the add-on payments) attempt to bridge the gap between base rate payments

and the hospitals’ costs to serve the Medicaid population. The state could consider limiting

the reliance on cost reports so that reduced utilization or reductions in payments due to

other initiatives are not compensated by increased Direct Medicaid payments.

8. Apply UPL caps to individual hospitals. The state could consider applying hospital-

specific outpatient and inpatient UPL caps. Currently, consistent with federal regulations, the

state applies UPL caps to the total of payments made within the applicable service category,

but it does not apply individual hospital’s UPL caps. As the UPL in Missouri is significantly

impacted by the OOS payments, reducing them would affect the UPLs of the recipients of

OOS payments.

9. Adjust MCO hospital payments. The state could cap MCO hospital payments at a fixed

percentage of Medicaid FFS payments.

100

This initiative would require a modification of

MCO contracts.

10. Improve physician and behavioral health reimbursement. For physicians, not only has

the methodology for establishing rates (e.g., as a percentage of Medicare) not been

updated, but once set, the rates do not change. As a result, physician reimbursement is low.

It is likely that increasing reimbursement could help reduce provider shortage. Likewise,

there is a shortage of behavioral health providers. The state could consider integrating this

initiative in an overall VBP program.

11. Re-examine payment levels for financially vulnerable rural and safety net providers.

To the extent that other initiatives are undertaken that could reduce revenue to hospitals

generally, the state could consider re-examining the effects of the initiatives on financially

vulnerable rural and safety net providers in particular to determine whether adjustments in

payment levels, value-based payment structures, or other changes are necessary to

mitigate the potential for erosion of access to care.

100

Missouri DSS: see note 72.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

12. Transition to value-based payments. In line with the healthcare industry trend led by other

states, Medicare, and commercial plans, Missouri could consider moving from its current

FFS payment methods to value-based payment (Alternative Payment Models, or APMs). In

VBP, providers do not miss or lose revenue by increasing efficiencies and reducing

potentially avoidable care services, as they tend to do in FFS. Rather, APMs allow providers

to be rewarded if they reduce hospital (re-)admissions and nursing home admissions. This

creates a strong business case for providers to invest in primary care, care coordination,

integration of physical and behavioral health care, and home health care. In addition, if

investing in social determinants of health creates net savings in Medicaid expenditures,

shared savings or other VBP payments can be leveraged to fund those services.

Some forms of value-based payment could be implemented without changing the

underlying architecture of the current FFS payment methods by overlaying rewards (and in

some cases penalties) on top of FFS payment streams. This would facilitate

implementation, as current administrative and billing processes, including the role of the

MMIS, would require little change. The additional analytical capabilities required for VBP

tend to be delivered by vendors, requiring limited interaction with the MMIS. With a

combination of models, building upon current, successful initiatives, Missouri could include

most Medicaid participants in VBP initiatives:

– Population-based models. Advanced Primary Care models (APCs) build on the PCMH

model but increase accountability for improved outcomes of the total attributed

population and total cost of care. Improvement in these parameters leads to higher

bonus payments and a share in the savings realized. Accountable care organizations

(ACOs) take this responsibility for the total costs and outcomes of care for a population

one step further. If the ACO manages to reduce total costs of care below a target

budget (usually based on the historical costs of care of the ACO’s population), they

receive up to 50% of the savings. ACOs can be upside-only (e.g., only savings are

shared), or they can include up- and downside risk (e.g., both potential savings and

losses are shared). As risk-based APMs reduce the payor’s risk for losses, they can

share significantly higher percentages of the realized savings (up to 100%) with the

providers. ACOs can be led by primary care organizations, hospitals, and (virtually)

integrated groups of providers, amongst others.

For specific high-need subpopulations (e.g. individuals with co-morbidity, severe mental

illness, and/or substance use disorder), the existing health home model(s) could be

leveraged to further improve outcomes.

Population-based models stimulate a focus on prevention and the management of

chronic disease and individuals with severe comorbidities: avoiding the need for

(institutional) care (including many PECs) is the most cost-efficient way to realize

savings.

– Bundled payments or episode-based models. In episode-based models, providers

assume responsibility for the costs and outcomes of a set of services to treat a certain

clinical condition or conduct a certain procedure. Like an ACO, episodes have target

budgets, and they can be upside-only or risk-based. Episode-based models stimulate

the creation of patient-centered care pathways across organizational boundaries. PECs

and the inefficient utilization of care services are addressed through care coordination.

Episodes tend to achieve impact faster as population-based models, as the

opportunities for improvement tend to be clear and specific. Several private and public

payors have combined population- and episode-based payments to create a “best-of-

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

both-worlds” mix of incentives for population health and high-value condition-specific

care.

– Global budgets for rural hospitals. In rural areas, the state could consider global hospital

budgets. Sixty-seven percent of Missouri counties have less than 5,000 managed care

Medicaid participants. Access to primary care physicians and hospital facilities can be

difficult, and the financial viability of many rural hospitals is under pressure. For such

hospitals, global budgets (based on the expected cost of the hospital services for an

attributed population) could create financial stability and facilitate transformation to a

care delivery model aligned with local community needs. The establishment of regional

ACOs or advanced primary care models with, for example, performance-dependent

PMPM care management fees could further help to mature these geographies’ regional

care infrastructure.

EXHIBIT 24: MEDICAID ENROLLMENT PER COUNTY, DISTRIBUTION OF BEDS,

SFY2016

101

13. Create transparency for outcomes of care. Providing transparency of outcomes for

(sub)populations and key conditions/procedures is a prerequisite of any health care

system oriented towards value. Juxtaposing these outcomes to the risk-adjusted costs of

care shines light on the performance of the healthcare delivery system and provides the

information providers, payors, participants, and policymakers require to make informed

choices and focus improvement efforts. As the collection point of all Medicaid claims

101

Analysis of Missouri Medicaid claims data, SFY2016-18.

100

5K or less 5 - 10K 10 – 15K 15K or more

Total members distribution by county, SFY2018

Audrain

Barry

Barton

Bates

Benton

Bolli-

nger

Boone

Butler

Callaway

Camden

Cape

Girard-

eau

Carroll

Carter

Cass

Cedar

Chariton

Christian

Clay

Cole

Cooper

Craw

ford

Dade

Dallas

Dent

Douglas

Dunklin

Franklin

Gasconade

Greene

Henry

Hickory

Howard

Howell

Iron

Jackson

Jasper

Jefferson

Johnson

Laclede

Lafayette

Lawr-

ence

Lincoln

Mcdonald

Madison

Maries

Miller

Mississippi

Moniteau

Monroe

Mont

gomery

Morgan

New Madrid

Newton

Oregon

Osage

Ozark

Pemiscot

Perry

Pettis

Phelps

Pike

Platte

Polk

Pulaski

Ralls

Randolph

Ray

Reynolds

Ripley

St. Charles

St. Clair

Ste. Genevieve

St Francois

St. Louis

Saline

Scott

Shannon

Stoddard

Stone

Taney

Texas

Vernon

Warren

Washington

Wayne

Webster

Wright

St. Louis City

Adair

Andrew

Atchison

Caldwell

Clark

Clinton

Daviess

De Kalb

Gentry

Grundy

Harrison

Holt

Knox

Lewis

Linn

Livingston

Macon

Marion

Mercer

Nodaway

Putnam

Schuyler

Scotland

Shelby

Sullivan

Worth

Buchanan

Size: Avg. # of

beds in FY16

≤100

100-200

200-400

>400

Critical access hospitals

Private hospital systems

Rural

Safety net

Academic hospitals

Other

Annualized members K

Cum.

Share of total members

(%)

St. Charles

Greene

St. Francois

Boone

Jasper

Jackson

140

St. Louis

Pemiscot

Clay

Jefferson

Buchanan

Callaway

Christian

Franklin

Butler

Pettis

Cass

Washington

Cape Girardeau

Barry

12Scott

Platte

Lincoln

Other

139

McDonald

355

114

St. Louis city

Grand total 1,156 K

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

data, the state could publish such information on the total costs and outcomes of care

per county or (group of) provider(s).

102

14. Include MCOs in a VBP program to maximize impact and align incentives for

providers across the total Medicaid population. The majority of Medicaid program

participants are enrolled in MCOs.

103

To create the volume for providers to be

sufficiently incentivized to participate, both FFS and managed care participants may

need to be included, and the APMs across MCOs and Medicaid FFS may need to be

adequately aligned. If some MCOs implement bundles and others carve out ACO

subpopulations in different ways, providers cannot (and will not be motivated to) make

the investments to change their business models. In addition, without alignment between

APM definitions, the measurement of outcomes and financial performance will likely not

be statistically feasible. Following the example of an increasing number of states,

Missouri could consider working with its MCOs to facilitate this alignment and change

MCO contracts accordingly.

15. Explore multi-payor VBP alignment. To further increase the potential impact of value-

based payment, the state could consider collaborating with non-Medicaid payors in the

state to align APMs and set collective goals. To significantly increase impact (see

previous initiative), multi-payor models are becoming increasingly widespread: CMS’

Comprehensive Primary Care models (CPC and CPC+) are an example. Two options

the state could explore are, first, alignment with the other main state government payor,

the Missouri Consolidated Health Care Plan (MCHCP).

104

Second, the state could

consider engaging with CMS to facilitate mutual alignment between the existing and

forthcoming Medicare APMs and the Missouri VBP strategy.

105

16. Update the DME fee schedule. Missouri could update its DME rates to match those of

other state Medicaid agencies and MCOs, which could potentially be supported by

competitive procurements in specific categories. Operationally, this would require a

change in the fee schedule, potential procurements, and efforts to ensure access.

102

Transparency of costs and outcomes requires a minimum number of attributed participants to allow comparisons;

individual professionals, for example, may not see sufficient participants to be meaningfully compared to others.

103

To optimally align incentives between the state, the MCOs and providers in APMs, Missouri could consider

carving in pharmacy and behavioral health for the MCO population.

104

Including MCHCP would add approximately 100,000 lives. See: http://www.mchcp.org.

105

This could be relevant for both duals (who make up a disproportionately large share of both Medicaid’s as well as

Medicare’s total spending) as well as for non-duals (where alignment between APMs would increase impact in the

same way as alignment with other payors would).

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

LONG-TERM SERVICES AND SUPPORTS

Certain elderly populations and others with disabilities are eligible to receive long-term services

and supports (LTSS) to assist with activities of daily living and otherwise support greater

independence. Before receiving LTSS, Medicaid-eligible individuals undergo an assessment

process, which determines eligibility but does not impact placement, type, or intensity of

services to be provided. Once individuals are deemed eligible, the services they may receive fall

into two categories: 1) institutional services and 2) home and community-based services

(HCBS, which can be split into residential and non-residential services). Nursing facilities are

reimbursed using a cost-based payment methodology without adjustments for acuity, quality of

care, or outcomes. As a result, there are minimal incentives for these facilities to provide

differentiated care to high-needs patients, or to transition lower-needs participants back to their

homes or the community. HCBS are provided through a combination of State Plan and waiver

programs; HCBS providers are not held accountable for nursing home (re-)admission rates.

Opportunities to improve quality and control costs of LTSS are primarily to be realized from

increasing the proportion of LTSS recipients that receive services at home or in the community

rather than in more costly institutional settings, and improving care planning and care

management of members regardless of their setting of care.

106

Potential initiatives include

improving utilization management, adjusting the nursing facility reimbursement methodology to

an acuity-based system, completing and expanding upon planned improvements to the

assessment algorithm and process, expanding current grant- or waiver-funded programs to

cover services that support individuals in the home or community, and shifting to value-based

payment. In total, the gross impact of the LTSS initiatives could range from $90 million to $275

million, net of potential reinvestments in the delivery system, depending on choices that the

state may make in the selection, design, and implementation of initiatives.

107

Current situation

This section gives an overview of the population receiving LTSS in Missouri, the assessment

and service authorization process, institutional services, and HCBS.

LTSS population and services

In SFY2018, approximately 106,000 individuals received LTSS in Missouri, representing 39% of

the state’s total aged, blind, and disabled (ABD) population.

108

However, spending for recipients

of LTSS, which was approximately $2.9 billion in SFY2018, represented 71% of the state’s total

spending on the ABD population. 53% of Medicaid elderly and 33% of persons with disabilities

receive LTSS (see Exhibit 25).

109

106

IBM Watson Health, “Medicaid Expenditures for Long-Term Services and Supports in FY 2016,” 2016, see:

www.medicaid.gov/medicaid/ltss/downloads/reports-and-evaluations/ltssexpenditures2016.pdf.

107

AARP, Commonwealth Fund, SCAN Foundation, “Long-Term Services & Supports State Scorecard,” 2018, see:

longtermscorecard.org/databystate/state?state=MO; IBM Watson Health: see note 106.

108

Analysis of Missouri Medicaid claims data, SFY2016-18.

109

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 25: LTSS PARTICIPANTS ~40% OF ABD POP’N, 70% OF COSTS, SFY2018

110

There exist several publicly available reports and datasets that compare performance of the

LTSS system across states. These sources reveal several insights about the current

performance of Missouri’s system. For example, Missouri is ninth in the country on the

performance of its No Wrong Door system, which is a national program to streamline access

to new LTSS options, improving the patient experience and potentially reducing cost of

care.

111

In addition, the state is near the top quartile of states when ranked by the share of

LTSS expenditures that goes towards HCBS. In SFY2018, approximately 61% were for home

and community-based services (see Exhibit 25). The national average of the HCBS proportion

of total LTSS spending was 57% in SFY2016.

In other areas, Missouri performs below the national average. For example, the state ranks

in the country in the percentage of nursing home residents that have low care needs (24%

vs. the national average of 11%), suggesting opportunities for a greater share of LTSS

recipients to be supported within the home and/or community.

112

Furthermore, Missouri ranks

in the country in the employment rate (19% vs. the national average of 22%) for adults

with Activities of Daily Living (ADL) disabilities, relative to those without them.

110

Analysis of Missouri Medicaid claims data, SFY2016-18; Validation checks performed against Table 23 suggests

a total of 1.04 million beneficiaries after inclusion of Women's Health Services (977k not including Women’s Health

Services). There is ~12% gap in enrollment due to differences in participant accounting (e.g., exclusion of non-

claimants).

111

AARP, Commonwealth Fund, SCAN Foundation: see note 107.

112

AARP, Commonwealth Fund, SCAN Foundation: see note 107.

27%

(1.5B)

100% =

61%

(165K)

39%

(106K)

5.5B

Members

52%

(2.9B)

20%

(1.1B)

Non-LTSS

spend for

LTSS

beneficiaries

Spend

LTSS spend

Spend for

non-LTSS

beneficiaries

270K

SFY2018 LTSS beneficiaries, K, and spend, $

Breakdown of SFY2018 Medicaid participants by

eligibility type, K

Elderly (81)

Total Medicaid

members

Managed care

population

(903)

Non-ABD

population

Disabled

(189)

ABD population

1,173

270

▪ LTSS analysis is focused on 270K participants in ABD population that make up 23% of total Medicaid

enrollment

▪ Across the elderly and disabled beneficiaries, we found that individuals receiving LTSS services make up

~40% in enrollment but ~70% of total spend

Individuals

receiving

LTSS

Individuals

not

receiving

LTSS

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Non-institutional services with the highest overall spending are residential services and

personal care.

113

Residential services are covered exclusively by DMH’s Comprehensive

Waiver (see “HCBS” section for additional details on this waiver). Personal care is covered by

both the State Plan and almost all the waivers. The State Plan pays for the personal care for

the majority (58%) of LTSS participants.

See Exhibit 26 for a breakdown of the LTSS spending by service category and the number of

participants receiving services in each category.

114

EXHIBIT 26: BREAKDOWN OF LTSS SPENDING BY SERVICE CATEGORY, SFY2018

115

Assessment and service authorization

Individuals can receive LTSS through either DHSS or DMH. DHSS follows different

authorization procedures for adults and for children. For adults, personal care services can be

either agency-directed (e.g., where a state agency is responsible for managing participants’

personal care, including selection and supervision of personal care assistants) or consumer-

directed (e.g., where the participants manage their own services by selecting, hiring, and

supervising their own personal care assistants). For adults, LTSS can also include institutional

care (nursing homes). DHSS uses the interRAI HC assessment (commonly used nationally) to

113

Analysis of Missouri Medicaid claims data, SFY2016-18.

114

Analysis of Missouri Medicaid claims data, SFY2016-18.

115

Analysis of Missouri Medicaid claims data, SFY2016-18; participants may overlap across categories.

Residential services 661

Targeted case management

Community services

701

Skilled nursing facility

Intermediate care

Counseling and therapy

Adult day care / day habilitation

LTSS Other

Career and financing

Personal care

Private duty nursing

Respite care

Targeted case management

Private duty nursing

1,041

182

Institutional

Total spend = $ 2,886M

SFY2018

Waiver

HCBS

State plan

HCBS

Total LTSS spend, $M

Share of total

LTSS spend

Average participants, K

Share of total

LTSS pop.

Total pop =

105.2K participants

0.4

1.2

29.8

0.3

1.5

6.1

5.8

15.0

6.8

3.4

0.2

0.4

13.3

61.7

6.0

LTSS service category

24%

58%

13%

14%

36% 28%

23%

3% 0%

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

determine need for institutional level of care. For children, personal care services – which must

be agency-directed – can be authorized through the Bureau of Special Health Care Needs and

are renewed every six months. For these children, the criterion for care is medical necessity

rather than institutional level of care needs. Meanwhile, DMH has multiple assessments and

determines which to use primarily based on the age of the individual (e.g., the MOCABI

116

for

adults; the Vineland assessment or another age-appropriate

117

substitute for children).

The department conducting the assessment then processes the results of the assessment to

determine whether the individual is eligible to receive LTSS. DHSS uses a points-based system:

individuals who receive a score of 24 points or above are eligible for institutional level of care,

which makes them eligible for LTSS offered through DHSS. On the other hand, if DMH

determines that the individual has two or more (three or more for waivers) functional limitations,

the assessor completes a Level of Care form to demonstrate the need for intermediate care

facility for individuals with intellectual disabilities (ICF/IID) level of care.

Once an individual is deemed eligible for LTSS, the departments then engage in person-

centered care planning, in which case managers work directly with individuals to determine

setting, level, and type of care to be provided. For services provided through DHSS (except for

those provided through the Independent Living Waiver), person-centered care planning is

performed by regional assessors. For the Independent Living Waiver, person-centered care

planning is provided through targeted case management (TCM) providers covered under the

waiver. For services provided through DMH, care planning is performed by TCM providers

covered by the Medicaid State Plan. The care planning process does not consistently use the

results of the assessment to inform the setting, level, or type of care authorized. Rather,

assessors are trained to use an HCBS manual (for DHSS services) or an Individual Support

Plan (ISP) guide (for DMH services) to inform what, where, and how much of each service can

be authorized, but unlike many other states, Missouri does not require that the assessor follow

these guidelines.

Institutional services

Nursing facilities are reimbursed using a cost-based per diem methodology at the facility level.

While not uncommon among states, this methodology does not take into account patient

acuity, intensity of service, quality, or outcomes in determining nursing facility payment levels.

As a result, nursing facilities are not necessarily incentivized to provide cost-efficient or

appropriate levels of care. Furthermore, the per diem rates are based on historical cost reports

that can date back over two decades; yearly incremental adjustments are determined by the

state legislature. In total, Missouri’s nursing facility payments fall well below the nursing facility

Upper Payment Limit.

118

Approximately 40% of the funds used to pay nursing facility reimbursement is derived from the

Nursing Facility Federal Reimbursement Allowance (NFFRA).

119

Like other states, Missouri

taxes nursing facilities and uses these revenues to fund Medicaid and draw down federal funds

at the Missouri federal match rate of 65.4%.

120

116

Missouri Critical Adaptive Behaviors Inventory.

117

Interviews with Medicaid program staff members.

118

UPL analysis, Missouri DSS, 2018.

119

Missouri DSS, “SFY 2014-2019 Rate by Funding Source & Cost Component – 9-28-18,” 2018.

120

KFF: see note 12.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

For individuals with intellectual and/or developmental disabilities, the state operates four public

ICFs/IID known as habilitation centers, which collectively house 315 participants, and contracts

with a number of private ICFs/IID, which house another 82 individuals, totaling 397 participants

in 2018 (down from 435 in 2017).

121

This number will likely continue to trend downward, as the

state plans to reduce admissions in Missouri ICFs/IID further.

HCBS

Missouri covers HCBS through a combination of State Plan and waiver programs. State Plan

services include targeted case management, personal care, and private duty nursing. For these

services, individuals who exhaust the maximum amount allowed by the State Plan may then

access additional allotment of these services through waivers, which cover care beyond what

the State Plan is able to fund. Waivers can include a broad range of services, such as personal

care and residential services. These waivers do not qualify for enhanced federal match, and

each waiver has an expiration date, at which point the state can elect to renew the waiver or

allow it to expire.

There are nine HCBS waivers, four of which are administered in coordination with DMH and five

of which are administered in coordination with DHSS. While these waivers use various rate-

setting mechanisms, they are fundamentally cost-based. Most individuals can only be on one

waiver at any given time. HCBS are split between residential and non-residential services.

Residential services, which include shared living and group home services, serve over 6,800

individuals, primarily through the DMH Comprehensive Waiver.

The set of waivers

122

includes the following (see Exhibit 27 for a summary of trends in costs

and participant count in these waivers):

• Aged and Disabled Waiver (DHSS through the Department of Senior and Disability

Services / DSDS, served 15,200 individuals in Waiver Year [WY] 2016): For individuals

age 65 years and older (or 63 and older if they have disabilities) that have impairment and

unmet needs. Services covered include homemaker and chore services, home-delivered

meals, respite, and adult day care.

• Adult Day Care Waiver (DHSS through DSDS, served 1,588 individuals in WY16): For

individuals age 18 to 63 years with impairments and unmet needs. This waiver exclusively

covers adult day care services.

• Independent Living Waiver (DHSS through DSDS, served 190 individuals in WY16): For

individuals age 18 to 64 years with cognitive and/or physical disabilities but also the ability to

self-direct. This is the only one of the nine waivers that covers targeted case management;

part of the waiver’s purpose is to serve as a continuation of State Plan targeted case

management.

123

It is also the only one of the five DHSS waivers that explicitly covers self-

directed personal care.

• Comprehensive Waiver (DMH, served 8,882 individuals in WY15): For individuals with

intellectual and/or developmental disabilities. This is the only one of the nine waivers that

121

Missouri DSS, “State Operated Habilitation Centers,” 2018.

122

DHSS waivers with < than 400 individuals served (number served in WY16): AIDS Waiver (90), Medically Fragile

Adult Waiver (166). DMH waivers with <400 individuals served (number served in WY15): MO Children with

Developmental Disabilities Waiver (320).

123

Interviews with Medicaid program staff members.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

covers residential services (e.g., group home, shared living, individualized supported living;

see Exhibit 28), but it also covers a range of other services, including personal care.

• Community Support Waiver (DMH, served 1,886 individuals in WY15): For individuals

with intellectual and/or developmental disabilities who already have a place to live in the

community. Given that requirement, residential services are not covered by this waiver, but it

otherwise covers the same range of services as the Comprehensive Waiver. It has an annual

per capita cost cap of $28,000.

• Partnership for Hope Waiver (DMH, served 2,614 individuals in WY15): For individuals

with intellectual and/or developmental disabilities who reside in one of 104 Missouri

counties plus St. Louis City. It covers the same set of services as the Community Support

Waiver – in addition to others, such as dental services – and it has an annual per capita

cost cap of $12,362.

EXHIBIT 27: COST PATTERNS FOR THE LARGEST LTSS WAIVERS

124

Missouri DHSS, “HCBS Waivers DHSS,” 2018; Missouri DMH, “HCBS Waivers DHS,” 2018.

17,378

17,067

16,343

15,280

15,200

-3% p.a.

95.5

99.9

64.8

75.4

94.2

-10% p.a.

5,749

5,518

5,846

4,936

4,266

-7% p.a.

15.4

14.7

14.4

8.9

+20% p.a.

268

187

135

152

190

14132012 201615

-8% p.a.

201615142012 13

1.7

1.8

1.7

0.5

1.6

-2% p.a.

6,772

9,147

3,839

10,716

8,772

2012 13 14 15 2016

+7% p.a.

1,333

1,543

1,522

1,588

+6% p.a.

6,673

9,338

9,640

9,703

+13% p.a.

567.2

621.6

630.8

512.7

+7% p.a.

8,126

8,442

8,461

8,503

+2% p.a.

63,091

67,193

73,461

74,190

+6% p.a.

19.2

27.0

23.1

15.6

+20% p.a.

1,406

1,504

1,560

1,886

+10% p.a.

11,087

12,741

14,817

14,338

+9% p.a.

5.8

14.0

10.1

13.3

+34% p.a.

1,314

2,009

2,373

2,614

+26% p.a.

4,445

5,023

5,589

5,372

+7% p.a.

Costs by waiver year

USD, millions

Number of individuals served by

waiver year

PMPY, USDName of the waiver

Adult Day Care

Waiver

Comprehensive

Waiver

Community Support

Waiver

Partnership for Hope

Waiver

Aged and Disabled

Waiver

Independent Living

Waiver

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 28: SERVICES COVERED ACROSS LTSS WAIVERS

125

While Missouri Medicaid covers a range of HCBS, the plurality of spending is for personal

care, covered by both the State Plan and every waiver except for the Adult Day Care Waiver.

Potential opportunities for improvement

This section identifies potential opportunities to improve Missouri’s current approach to LTSS.

The opportunities are not intended to be mutually exclusive: cost savings opportunities

identified in individual opportunities may overlap with those identified in others. When

compared to experiences and practices in other states, the following observations can be

made:

• The patient journey to get access to LTSS can be complex. Three state agencies

(DSS, DHSS, and DMH) play a role in the process of determining eligibility for LTSS and

planning care for LTSS recipients. As a result, while Missouri has adopted the principle of

“no wrong door” for eligibility and access to LTSS, the participant journey (see Exhibit 29)

can be complex and can vary widely depending on the participant’s condition and entry

point into the system.

125

Missouri DMH and DHSS: see note 124.

Operated by DHSS Operated by DMH

Aged and

Disabled

Independent

Living

AIDS

Medically

Fragile Adult

Adult Day Care

Comprehensive

waiver

Community

Support

MOCDD

Partnership for

Hope

Waiver

Service category

Community

services

Counseling

and therapy

Residential

services Other

Career and

financing

Adult day

care/day hab

Personal

care

Private duty

nursing Respite

Supplies and

technology

Targeted case

management

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 29: LTSS PATIENT-CENTERED PROCESS FLOW

126

• The DHSS assessment process to determine need for institutional level of care

uses decades-old standards and, as such, may not consistently determine

institutional level of care needs. DHSS is currently considering changes to the

algorithm it has used to determine nursing facility level of care. Although the state has

changed the threshold scores for determining LTSS eligibility, the algorithm has not

meaningfully changed since 1982.

127

• Assessment results are not consistently used to inform setting of care, type, or

intensity of services authorized. The care planning process currently does not

consistently use the results of the level of care assessment to inform the plan of care. As

a result, the setting of care, services, and service levels participants are authorized to

receive may not be consistent across programs or care planners, and the care provided

may not match participants’ needs.

• Personal care services are administered inconsistently depending on the channel

through which they are received. For example, utilization of consumer-directed

126

Interviews with Medicaid program staff members.

127

Missouri DSS, “Rules of Department of Health and Senior Services, Division 30-Division of Regulation and

Licensure, Chapter 81-Certification,” 2018, see: www.sos.mo.gov/cmsimages/adrules/csr/current/19csr/19c30-

81.pdf.

Patient receives

personal care

funded by State

Plan

Assess-

ment to

determine

and

Medicaid

eligibility NF

level of care

Case manager

develops person

centered care plan to

determine service

types and levels to

authorize

Institutional

/ residential

Personal

care

Other State

Plan

service

Non-PC

DMH

waiver

service

Patient receives

personal care

funded by

waiver(s)

▪ Patient is deemed not eligible

for LTSS services if score is

below 24 points on

assessment (i.e., below NF

level of care)

▪ If receiving DMH waiver,

patient must receive case

management; otherwise

targeted case management

provided only if necessary

▪ State Plan services that are

also covered by waivers (e.g.,

dental, PT/OT/ST for children,

applied behavioral analysis for

children with autism, DME)

require that patients exhaust

coverage from State Plan first

before seeking waiver

coverage

▪ Patients seeking just teaching

and cueing services can

receive services through DMH

waiver directly

Non-PC

DHSS

waiver

service

Service authorization Service delivery

Assessment

of patient

Targeted case

management to

determine service

types and levels to

authorize

Administered by DHSS

Administered by DMH

Administered by either DHSS or DMH

Potential service options

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

personal care services is almost two times higher than the agency-directed model.

128

Currently, so long as a given participant is eligible to receive consumer-directed care, she

may elect to choose it (e.g., participants are not allocated to one or the other). The

difference in utilization does not appear to be correlated with participant mix or participant

risk. In addition, average reimbursement rates vary depending on whether they are

provided through DMH or through DHSS; while these rates have converged in recent

years, there remain differences in rates, primarily due to funds available for each

department’s waivers.

129

• Nursing facility rates are based on historical costs, and they do not reimburse

based on patient acuity or create incentives for quality or outcomes. While there are

yearly adjustments to the per diem rates, the rates are based on cost reports from

SFY2001 (trended to SFY2005).

130

Although these per diems are intended to cover

nursing facilities’ costs, the reimbursement methodology does not necessarily reflect their

current costs. Additionally, there is little correlation between nursing facility per diem rates

and either patient acuity or facility quality (see Exhibit 30: darker bubbles represent

facilities that experience higher patient acuity on average, while bubbles on the right

represent facilities with higher Star ratings).

131

Currently, per diem rates vary from

$135.08 to $175.41 by facility, meaning the facility with the highest per diem rate receives

approximately 30% more than the facility with the lowest per diem rate.

132

Finally, per

diem rates do not incentivize facilities to discharge residents or attempt to avoid

admissions where feasible.

128

Analysis of Missouri Medicaid claims data, SFY2016-18.

129

Missouri DSS, “Adult Day Care Waiver,” 2018; Missouri DSS, “Aged and Disabled Waiver,” 2018; Missouri DSS,

“Aids Waiver,” 2018; Missouri DSS, “Community Support Waiver,” 2018; Missouri DSS, “Comprehensive Waiver,”

2018; Missouri DSS, “Independent Living Waiver,” 2018; Missouri DSS, “Medically Fragile Adult Waiver,” 2018;

Missouri DSS, “MOCDD Waiver,” 2018; Missouri DSS, “Partnership for Hope Waiver,” 2018; Missouri DSS, “Private

Duty Nursing,” 2018; Missouri DSS, “State Plan Personal Care,” 2018; Missouri DSS, “Targeted Case

Management,” 2018; interviews with Medicaid program staff members.

130

Missouri DSS, “Rules of Department of Social Services, Division 70-MO HealthNet Division, Chapter 10-Nursing

Home Program,” 2018, see: www.sos.mo.gov/CMSImages/AdRules/csr/current/13csr/13c70-10.pdf.

131

Analysis of Missouri Medicaid claims data and MDS data, 2016-18.

132

Missouri DSS, “Nursing Facility Rate List,” 2018 (updated 7/1/2018), see:

dss.mo.gov/mhd/providers/pages/nfrates.htm.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 30: SKILLED NURSING FACILITIES BY PER DIEM, STAR RATING, SFY2018

133

• Nursing homes have a relatively high number of low-acuity Medicaid residents. In

Missouri, 23.7% of nursing home residents have low care needs (e.g., could potentially be

adequately served through HCBS services) compared to the national median of

11.2%.

134

Diverting participants with low care needs to HCBS to reach the level of

median state performance could yield a reduction of spending of up to $90 million.

135

• Occupancy rates in nursing facilities are relatively low. With an average nursing

facility occupancy rate of 72%, Missouri ranks 43rd amongst other states, with the top 12

at occupancy rates of 88% or higher.

136

With further reductions likely, the inefficiencies

inherent to low occupancy rates will increase, and some nursing homes may not be able

to maintain their current business model.

• Additional waivers or grants could provide key services to certain subpopulations.

For example, the Money Follows the Person (MFP) program is set to expire. Extending it

or substituting it with a waiver may help ensure that individuals transferring from nursing

133

Acuity measured as average score on 10 MDS functional status questions (e.g., Section G0110) on most recent

25 days of information retained by the state. Providers with fewer than 15 data points excluded. High complexity

refers to SNFs with average ADL score of 4+; low with an average score of 2 or less; analysis of Missouri Medicaid

claims data and MDS data, 2016-18.

134

AARP, Commonwealth Fund, SCAN Foundation: see note 107.

135

IBM Watson Health: see note 106.

136

KFF, “Certified Nursing Facility Occupancy Rate,” 2016, see: www.kff.org/other/state-indicator/nursing-facility-

occupancy-rates.

146

148

140

168

158

164

176

166

154

156

170

150

162

160

152

144

136

138

142

High complexity Medium complexity Low complexity

Bubble size indicates total spend

Per Diem rate,

Star rating, number of stars

1 2 3 4 50

Some SNFs are

paid higher than

average despite

low STAR ratings,

suggesting

potential

opportunity to

redirect care to

higher quality

providers

Providers with

high quality of

care and low per

diem rates

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

facilities or habilitation centers have access to the resources they need to facilitate their

transitions. By shifting more people from institutional settings back to the community, this

change could result in savings of $12.5 million to $14 million.

137

Additionally, the state

could consider following through on discussions to implement a waiver that covers

children with developmental disabilities who do not require habilitative services, which

would cover the cost of care for children who do not qualify for Medicaid because of their

parents’ income. Currently, children need to be hospitalized for a certain period before

they can be considered eligible for Medicaid regardless of parental income; this may

result in children being hospitalized even if it does not suit the level of care they require.

• There may be additional opportunity to provide care for participants in less

intensive and restrictive settings even across the continuum of HCBS services.

Though a substantial amount of rebalancing from institutional to residential and other

HCBS (waiver and State Plan) services has taken place, there may be opportunity to

transition members receiving residential services in congregate care settings away from

their homes to less intensive and restrictive settings within the continuum of HCBS

services. See Exhibit 31 for a breakdown of LTSS spending based on utilization levels of

different services.

138

137

Interviews with Medicaid program staff members and analysis of state data, 2018.

138

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 31: LTSS UTILIZATION FOR LTSS ADMINISTERED BY DMH, SFY2018

139

• There are limited incentives connecting reimbursement of HCBS providers and

outcomes of care. Reimbursement of HCBS providers is not tied to their success in

keeping their clients out of nursing homes (or other forms of residential care). Likewise, as

payment is based on units of care delivered, there is no economic incentive to stimulate

participants’ independence from care. Payments to provider groups that aim to relocate

participants from nursing homes or residential care facilities to their homes could be tied

to their success rate, for example. Sharing in the savings could also help these providers

to focus even more on the social determinants that often stand in the way of successful

transition.

• There is little to no transparency of outcomes of care in LTSS. While available data

on the performance of LTSS in Missouri show mixed results, there is little or no publicly

accessible information about the outcomes of care delivered per (sub)population or

condition and per provider (or group thereof). This limits consumer choice, provider

accountability, and the information necessary for provider self-improvement.

139

Only considers population receiving services from administered by DMH; analysis of Missouri Medicaid claims

data, SFY2016-18; residential services include individual supported living, group home, and shared living services;

CRG stands for Clinical Risk Group.

4.6

<0.1

0.4

<0.1

0.1

2.2

0.4

6.8

0.4

0.6

4.4

<0.1

0.1

2.9

6.0

3.0

5.5

5.6

5.5

6.0

5.8

6.0

6.6

7.0

6.3

6.1

137

149

213

212

Waiver PCAICF

State Plan PCAA

Residential Other HCBS

Average spend by service

category, $K PMPY

Utilization

CRG risk score

Average annualized

beneficiaries, 000

ICF / IIDs

HCBS

(excl.

residen-

tial

services)

Residen-

tial

services

0-30 days

31-60 days

90+ days

61-90 days

0-30 days

Overall

11-20 hr / wk

21-40 hr / wk

41-80 hr / wk

0-10 hr / wk

Overall

81-120 hr / wk

160+ hr / wk

121-160 hr / wk

90+ days

Overall

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Potential initiatives

Based on a review of Missouri’s current approach, interviews with functional leaders and subject

matter experts within the relevant departments, and analysis of other states’ activities, this

section discusses potential initiatives Missouri could consider to improve the value of LTSS in

the state, which include reducing costs and, through increasing the number of participants that

would be able to remain in their own homes and/or in the community, possibly improving

participant experience, quality, and outcomes. In total, the gross financial impact of LTSS

initiatives ranges from $90 million to $275 million, depending on choices made by the state.

1. Include an acuity adjustment in the nursing home reimbursement methodology.

Missouri could consider adding an acuity adjustment to the current per diem methodology.

By using an acuity adjustment such as a resource utilization group (RUG)-based grouper,

Missouri could categorize patients based on need and reimburse nursing facilities

accordingly, using a stratified set of per diem rates. This shift would enable allocation of

resources based on need. Additionally, it may encourage further rebalancing from

institutional care to HCBS.

2. Rationalize rates for similar HCBS services provided through different programs and

funding authorities. For services provided through multiple waivers or through a

combination of State Plan and one or more waivers (e.g., personal care services), Missouri

could consider standardizing rates independent of the funding source for the service.

Without standardization, providers may be reimbursed different amounts for care provided

to patients with similar needs and acuity levels, which may encourage them to participate

selectively in certain programs while not participating in others. This may result in access

issues in certain programs and/or geographic areas, eroding patient experience and

outcomes. DHSS has engaged an external vendor to conduct a rate study to determine the

validity of the reimbursement rates for services covered in their waivers, which may reveal

further opportunities to rationalize rates.

3. Complete and expand upon revisions currently underway to assessment algorithm

and process. The state recently announced changes to DHSS’ algorithm to assign points

using the interRAI HC assessment instrument, which represent the first major changes

since 1982. These revisions could improve the accuracy of the level of care assessment

process. The state could also consider further streamlining and strengthening the

assessment process across populations, programs, and departments (e.g., improving

capture of personal care data with review on a per-reviewer and per-physician basis,

especially in the consumer-directed program).

4. More directly employ assessment results in care planning process. In addition to

improving the assessment process as is currently planned, Missouri could consider

incorporating additional functionality into the assessment instrument. First, it could be used

to determine eligibility for services. Second, it could more closely tie results of assessment

to the care planning process. For example, DHSS has previously considered using a case

rate-based system, using a RUG-based grouper mechanism layered on top of the current

interRAI HC assessment. This could include more consistently using assessment results as

a standardized basis for setting of care determinations and the types and intensity of

services to be provided. Third, the assessment instrument could be used to determine

payment levels for care. Fourth, the assessment results could serve as an auditing

mechanism: care planners and/or providers could be flagged if they are providing a level of

care that is inconsistent with the results of the assessment.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

5. Improve the consistency of the approval process for personal care services. The

state could better capture personal care (PCA) PA data digitally and review it on a per

reviewer and physician level to ensure consistency in implementing assessment tools and

appeals processes. This would be especially important in the consumer-directed program,

as different PA approvers may be inconsistent in the type and degree of services they

authorize for different individuals with similar care needs.

6. Extend Money Follows the Person (MFP) through a new grant or waiver. On average,

MFP in Missouri has helped 206 individuals each year to transition back to their

communities.

140

The quality of life of individuals living at home may be much higher than it

may be for those living in an institution; in addition, the cost of a year of nursing home care

is $45,000, versus ~$8,300 for home-based care. According to experts interviewed, if the

state includes a rent subsidy for those in the MFP program, it could double the number of

transitions per year, to approximately 400 per year.

7. Implement additional waivers (e.g., waiver for children with developmental

disabilities who do not require habilitative services) or expand current waivers.

Implementing such a waiver would allow children who are ineligible for Medicaid because

of their parents’ income to receive Medicaid services without hospitalization. This would not

only allow children to receive care from the comfort of their homes, if they do not require

more intensive care, but would also potentially reduce the cost of care.

8. Missouri could consider introducing Alternative Payment Models (APMs) for LTSS

services. The main value opportunity for LTSS services is moving care from a nursing

home or residential services to care in the participant’s home where possible. The costs of

this care are generally less than half the cost of intuitional care and living at home tends to

be highly preferable.

141

Improving care planning and management for this population can

also be a significant source of value. An Accountable Care Organization model, specifically

designed for LTSS, may be one option to incentivize providers to create this value. Yet for

those providers most likely to do so – home care providers – taking on the financial

responsibility for nursing home costs is a large risk and is likely not feasible for many

smaller providers. Alternatively, such providers could be incentivized by tying a part of their

reimbursement to the key outcomes that matter to participants, such as the extent to which

they can be successful in delaying or avoiding nursing home admissions, improving self-

determination, encouraging independence at home, etc.

9. Create transparency of the outcomes of care. Providing transparency of outcomes for

(sub)populations is a prerequisite of any healthcare system oriented towards value.

Juxtaposing these outcomes to the risk-adjusted costs of care shines light on the

performance of the healthcare delivery system and provides the information providers,

payors, participants, and policymakers require to make informed choices and focused

improvement efforts. As the collection point of all Medicaid claims and assessment data,

the state could publish such information on the total costs and outcomes of care per county

per provider, or per group of providers.

140

Missouri DSS, “Money follows the person,” 2018.

141

Missouri DSS: see note 140.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

PHARMACY

In SFY2018, Missouri Medicaid spent ~$1.5 billion on pharmaceutical products.

142

Missouri is

one of four states that carves pharmacy out of its managed care arrangements.

143

This carve-

out gives the state complete responsibility for paying for and managing the utilization of drugs

for all participants. To ensure appropriate utilization and control spending, the state has

established a preferred drug list (PDL), which requires prior authorizations, step therapy, and

quantity limits for select drugs. Through its process of “grandfathering” treatment, Missouri does

not require participants that are established on a non-preferred drug to switch to a preferred

drug. In addition to the PDL, Missouri receives statutory and supplemental rebates from

pharmaceutical manufacturers as means of cost containment. The state uses a vendor to help it

maintain its PDL and to assist in supplemental rebate negotiations.

The state pays for retail drugs in two ways: an ingredient cost and a dispensing fee. In terms of

reimbursement for the ingredient cost, Missouri, like other state Medicaid agencies, has

converted to an average actual cost methodology. The state is in the process of revising its

dispensing fee.

Potential initiatives for Pharmacy include the elimination of grandfathering of drug selection,

implementing additional utilization management, joining a purchasing consortium to increase

supplemental rebate capture, requiring NDC submission on claims for non-J-code HCPCS

drugs, establishing a preferred specialty pharmacy, and applying for a value-based contracting

waiver from CMS. When combined, the potential impact of Pharmacy initiatives could range

from $35 million to $60 million, net of ongoing operational costs.

144

This savings opportunity is

variable and dependent on decisions that are made with respect to initiatives discussed in the

managed care and acute care services sections (e.g., including pharmacy as an MCO-covered

benefit).

Current situation

This section gives an overview of Missouri’s current pharmacy in terms of spending and

structure, reimbursement methodology, utilization management (UM) practices, clinical

guidelines and (for pharmacy) rebate capture.

Program spending and structure

In SFY2018, 25 drugs accounted for ~25% of Missouri’s $1.5 billion pharmacy spending, while

4141 drugs accounted for the other 75%. Total pharmacy costs have grown 5% over the last

three years. Treatment for attention-deficit/hyperactivity disorder (ADHD), hepatitis C,

behavioral health conditions, hemophilia, rheumatologic conditions, diabetes, asthma, growth

deficiency syndromes, and pain are the main drivers of pharmacy spending and growth (see

Exhibit 32).

145

142

Analysis of Missouri Medicaid claims data, SFY2016-18; only includes pharmaceutical products billed as

separate pharmacy claims.

143

KFF, “States Focus on Quality and Outcomes Amid Waiver Changes: Results from a 50-State Medicaid Budget

Survey for State Fiscal Years 2018 and 2019,” 2018, see: www.kff.org/report-section/states-focus-on-quality-and-

outcomes-amid-waiver-changes-pharmacy-and-opioid-strategies.

144

Analysis of Missouri Medicaid claims data, SFY2016-18; interviews with Medicaid program staff members.

145

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 32. 25 DRUGS ACCOUNT FOR ~25% OF PHARMACY SPENDING, SFY2018

146

Missouri carves pharmacy benefits out of its managed care program. Missouri’s SFY2018

spending of $1.5 billion was paid on a fee-for-service basis. All pharmacy program operations,

including utilization management, are the responsibility of MHD. Missouri also utilizes a

preferred drug list vendor. This vendor assists the state with supplemental rebate negotiation

and updating/reviewing the state’s PDL. Finally, the state has an open pharmacy network,

including an open specialty network. An open network allows participants to use any pharmacy

of their choice.

Reimbursement

Missouri uses a recently modified hierarchy method to determine reimbursement for drug

ingredient costs. Missouri reimburses covered drugs by applying a hierarchy method that

starts with National Average Drug Acquisition Cost (NADAC), followed by Missouri Maximum

Allowed Cost (MAC), and Wholesale Acquisition Cost (WAC). Missouri uses the usual and

customary (U&C) charge submitted by the provider if it is lower than the chosen price.

Reimbursement for covered drugs for 340B providers who carve-in for Medicaid was modified

by applying the following method: WAC-25% or the U&C charge submitted by the provider if it

is lower.

147

146

Analysis of Missouri Medicaid claims data, SFY2016-18.

147

Missouri DSS, “State of Missouri Pharmacy Manual,” 2019, see:

manuals.momed.com/collections/collection_pha/print.pdf.

Drugs

LATUDA

MAVYRET

INVEGA SUSTENNA

HUMIRA PEN

METHYLPHENIDATE ER

VYVANSE

PROAIR HFA

LYRICA

ADVAIR DISKUS

LANTUS SOLOSTAR

NORDITROPIN FLEXPRO

SYMBICORT

SPIRIVA

ADDERALL XR

ARIPIPRAZOLE

FLOVENT HFA

NOVOLOG FLEXPEN

SYNAGIS

OXYCONTIN

KOGENATE FS

METFORMIN HCL ER

PALIPERIDONE ER

ONFI

SUBOXONE

GENVOYA

36.0

28.9

28.8

20.7

20.6

18.9

18.4

16.6

16.5

15.9

15.0

12.8

12.6

12.2

11.9

11.7

10.8

10.4

10.1

9.7

9.1

8.8

8.5

SFY2018 Spend, ($M) SFY2017 to SFY2018 spend trend

22%

16%

13%

-1%

-8%

-6%

18%

-7%

-10%

-22%

16%

12%

-7%

-25%

110%

55%

13%

17%

39%

31%

-5%

65%

-5%

-1%

-6%

-19%

-21%

14%

-1%

-24%

-2%

-6%

16%

-3%

-7%

-28%

52%

-13%

13%

37%

26%

SFY2017 to SFY2018 utilization trend

New to the market in SFY2018

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Missouri also has structured fees for reimbursement rates for pharmacy dispensing fees. The

state currently pays $9.55 in base dispensing fee to all pharmacy providers, and $4.82 in

enhanced dispensing fee to in-state pharmacy providers. In addition, $5.00 in preferred

generic product incentive fee is paid for each multi-source product to in-state pharmacy

providers. In addition to the retail fees, an additional $0.50 in long-term care dispensing fee is

paid per claim under specific circumstances. Outpatient physician-administered drugs are

reimbursed as a percentage of billed charges for hospital providers. These fees are under

active review with CMS.

148

Utilization management practices and clinical guidelines

Utilization management protocols are in place for a range of drug classes but lacking in some.

Prior authorization (PA), step therapy, and quantity limits are used across the PDL. However,

these UM techniques are lacking in certain drug classes (e.g., hemophilia, oncology). Newly

approved drugs are automatically placed on the PA list for the first six months after launch.

Additionally, Missouri uses an automated PA system for first-level clinical reviews. The system

can match participant diagnosis codes to approval criteria to generate automated

approvals/denials.

Rebate capture

The state collects both federal and supplemental rebates. Missouri’s SFY2018 federal and

supplement rebate capture rates were 52.4% and 3.0% of total pharmacy spending,

respectively.

149

All claims for physician administered drugs with “J” prefixed HCPCS codes are

required to be submitted with an NDC so that rebates can be captured.

The state’s PDL vendor negotiates supplemental rebates on its behalf. The state collects

supplemental rebates in various therapeutic categories such as: growth hormones, anti-virals

to treat hepatitis C, ADHD therapies, and drugs used to treat rheumatologic conditions.

The process for rebate invoicing to manufacturers is highly manual. This process involves

using different computer systems to compare claims and invoices. Additionally, it takes the

effort of multiple FTEs to convert data from one system to another, do quality checks, send

invoices to manufacturers, and review any appeals that come back from the manufacturers.

In Missouri, providers may choose to either carve-in to or carve-out of 340B. The state follows

the guidelines set forth by the Health Resources and Services Administration (HSRA). All

covered entity providers are identified as such by the state and rebates are not collected on

these drugs.

Potential opportunities for improvement

This section identifies potential opportunities to improve Missouri’s pharmacy program. When

compared to the practices of other states, the following observations can be made:

• Missouri’s expansive grandfathering practice limits the state’s ability to shift

utilization to the lowest net cost drug. While some states allow grandfathering for

specific drug classes, most require participants to follow changes to the PDL.

• For certain high-cost drug classes (such as oncology, hemophilia, and IVIG), there

are no medical necessity policies. MCOs and some state Medicaid agencies have

148

Missouri DSS, “Missouri MoHealthNet Provider Bulletin Volume 39 Number 52,” 2017, see:

dss.mo.gov/mhd/providers/pdf/bulletin39-52_2017april14.pdf.

149

Analysis of Missouri Medicaid claims data, SFY2016-18.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

medical policies and often use utilization management levers such as PA, step therapy,

and quantity limits to ensure appropriate utilization in these high-cost drug classes.

150

• Missouri’s rebate capture rates are below the national average. While federal rebate

capture has improved from 43.4% to 52.4% between SFY2016 and SFY2018, Missouri is

still below the weighted national average of 55.5% (see Exhibit 33) and further below the

highest-performing quartile. Additionally, the state’s SFY2018 supplemental rebate capture

rate of 3.0% also falls below the weighted national average of 3.8%.

151

These deviations

from the mean may be in part due to grandfathering practices or PDL design.

EXHIBIT 33: STATE-BY-STATE, FEDERAL REBATE CAPTURE IN SFY2016

152

• Missouri does not currently participate in any value-based contracts with

pharmaceutical manufactures. Value-based contracting is becoming more popular with

commercial and MCO players. Recently, CMS approved waivers for Oklahoma and

Michigan to negotiate value-based contracts with pharmacy manufacturers.

150

Missouri DSS: see note 147.

151

Medicaid.gov, “Expenditure Reports from MBES/CBES,” see: www.medicaid.gov/medicaid/finance/state-

expenditure-reporting/expenditure-reports/index.html; analysis of state data, 2018.

152

Medicaid.gov: see note 151.

FFS national rebate benchmarks, SFY2016

Alabama

Arkansas

Arizona

California

Colorado

Connecticut

Florida

Georgia

Iowa

Idaho

Illinois

Indiana

Kansas

Kentucky

Louisiana

Massachusetts

Maine

Michigan

Minnesota

Missouri

Mississippi

Montana

North Carolina

North

Dakota

Nebraska

New Hampshire

New Mexico

Nevada

New York

Ohio

Oklahoma

Oregon

Pennsylvania

South

Carolina

South

Dakota

Tennessee

Texas

Utah

Virginia

Vermont

Washington

Wisconsin

West

Virginia

Wyoming

Washington, DC

Delaware

Maryland

New Jersey

Rhode Island

Hawaii

Alaska

Data not available

Lowest Quartile 2nd Quartile 3rd Quartile Highest Quartile

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Potential initiatives

Based on review of Missouri’s current approach, interviews with functional leaders and subject

matter experts, and analysis of other states’ activities, this section has identified six potential

initiatives Missouri could consider for improvements to its pharmacy program. These initiatives

build on the existing progress made by the state and could result in a reduction of total Medicaid

expenditures from $35 million to $60 million, depending on state choices.

1. Implement medical necessity guidelines and prior authorizations in drug classes that

do not have such policies. The state could implement new medical necessity policies for

oncology, hemophilia, IVIG, and other select high-cost physician-administered therapies.

This could not only bring Missouri in line with other states and MCOs but could also require

that participants are receiving care based on accepted clinical guidelines in the proper

clinical sequence. A vendor could be utilized to handle this process, or the process could be

done in-house.

2. Reduce grandfathering. Missouri could consider only targeted use of grandfathering for

specific drug classes (e.g., antipsychotics) based on an review of clinical need.

Operationally, some requirements would include proper notification to participants and

providers to ensure that all stakeholders are aware of pending changes and to avoid any

impact on access.

3. Join a purchasing consortium to increase supplemental rebate capture. There are

three supplemental rebate consortiums that state Medicaid programs utilize today: the

National Medicaid Pooling Initiative (NMPI), the Optimal PDL Solution (TOPS) and the

Sovereign States Drug Consortium (SSDC). Missouri would need to consider how these

consortia fit with their current approach and PDL vendor. Additionally, the state would need

to submit a State Plan Amendment to CMS.

4. Require NDC submission on claims for non-J-code HCPCS drugs. This initiative could

ensure that rebates are captured on all physician administered drugs. Operationally, some

requirements would include provider notification and modest MMIS system edits.

5. Consider whether to contract with a specialty pharmacy. The state could establish a

preferred specialty pharmacy which may provide lower prices for certain specialty drugs,

and potentially better care management and improved clinical outcomes for participants.

Before doing this, the state would need to determine whether such an approach would be

consistent with any willing provider regulations. Additionally, the state would likely have to

go through the required procurement process.

6. Apply for a value-based contracting waiver from CMS. The state could apply for a value-

based contracting waiver from CMS, which would allow the state to negotiate drug prices

with manufacturers based on clinical outcomes. CMS approval of a State Plan Amendment

would be required, as would negotiation with manufacturers to determine the optimal

drug(s), outcome(s), and pricing.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

MANAGED CARE

In 2017, Missouri’s managed care program for children and families was expanded statewide

under three capitated managed care organizations (MCOs).

153

The state has taken several

steps to improve the performance of the managed care program and ensure its value, and the

current managed care contracts attempt to create an environment that fosters innovation

through incentive programs and specialized care coordination programs. Nevertheless, both the

managed care contracts and rates can be improved to further increase efficiency, eliminate

ambiguity in contract language, and lay the foundation for improved MCO performance and

state-of-the-art performance management. Finally, the state could consider increasing the

scope of managed care and carving in pharmacy and behavioral health services for the current

managed care populations. The state could also consider introducing managed care for (parts

of) the ABD population or continuing to improve management of those populations outside of

the managed care program.

The total potential impact across these initiatives ranges from $175 million to $300 million, net of

recurring investments.

154

While there are opportunities to improve the performance and

efficiency of the current managed care program, the largest component of this potential impact

could be achieved through the inclusion of additional services (e.g., behavioral health,

pharmacy) and populations (e.g., ABD) in managed care. If managed care were expanded to

the ABD population, MCOs could realize savings partially through implementing similar

initiatives as described in the LTSS section above. As a result, there is natural overlap in the

potential impact of these areas; if services for the ABD population – including LTSS – are fully

carved into managed care, then the aforementioned total potential impact would overlap with

the $90 million to $275 million from the LTSS section (and eliminating any incremental savings

from it).

Current situation

Scope of managed care

The managed care program encompasses children, parents, and pregnant women, and it

excludes most pharmacy and behavioral health services

155

(see Exhibit 34). Medicaid ABD

populations are entirely excluded from the managed care program. Children in foster care or in

subsidized, post-adoption or guardianship programs are included on an opt-out basis. Total

managed care spending is ~$2.2 billion (see Exhibit 35; children, parents, and pregnant

women comprise ~67% of Medicaid enrollees but drive only 23% of the spending, excluding

FFS spending for that same population).

156

153

Analysis of state data, 2018; Missouri Foundation for Health, “Missouri Medicaid Basics,” 2017, see:

mffh.org/wordpress/wp-content/uploads/2017/03/MedicaidBasics2017.pdf.

154

Missouri DSS, “Annual Table 23 and 24 for FY18,” 2018; Analysis of state data, 2018.

155

Missouri Foundation for Health: see note 153.

156

Missouri DSS: see note 5.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 34: CURRENT BEHAVIORAL HEALTH COVERAGE

EXHIBIT 35: MHD ENROLLMENT, MANAGED CARE AND FEE-FOR-SERVICE SPENDING

BY ELIGIBILITY GROUP AND CATEGORY OF SERVICE, SFY2018

157

Missouri DSS: see note 154; “children” excludes eligibility groups associated with foster care; “foster children”

includes foster care, child welfare; estimated share of Title XIX HDN population attributable to subsidized child

FFS MC

Foster Children (Ages 0-20);

Independent Former Foster

Adolescents (Ages 21-25)

All other MC eligibility groups:

Adults (21+), Pregnant Women,

Children COA 1, CHIP Exp, CHIP

Separate

FFS

FFS (under 21)

FFS

MO HealthNet Covered Benefits

Outpatient “clinic option” services

Inpatient Behavioral Health Admissions

Inpatient Admission with both Behavioral and

Physical Diagnoses

Applied Behavior Analysis for Autism Spectrum Disorder

Tobacco Cessation Counseling - psychologists,

LPCs, LCSWs

CPR (Comprehensive Psychiatric Rehab - DMH)

CSTAR (Comprehensive Substance Treatment and

Rehab - DMH)

TCM (Targeted Case Management - DMH)

Waivers

FFS FFS

Managed care Fee-for-service

156Disabled

751M 1226M 866M 139M 32M 138M1183M

46M 114M 1076M 106M <1M 114M141M

Seniors

Behavioral

Health LTSS

Rehab &

specialty EPSDT OtherMedical

Children

368M 146M <1M 7M 62M <1M1556M

162M

Members

Thousands

590

Custodial

parents

155M 26M 6M 2M <1M <1M401M

72M

Pregnant

women

23M 5M 1M 1M 1M <1M131M

42M

Foster

children

52M 131M <1M 2M 28M <1M71M

46M

<$100M $100-500M >$500M

Expenditures, $ Millions

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Managed care rate setting

Missouri’s MCO rate-setting methodology encourages efficiency, adjusts payments based on

risk, and manages non-benefit expenses. Efficiency adjustments have been implemented to

avoid payments for some avoidable emergency department (ED) and inpatient (IP) services

(e.g., low-acuity non-emergency adjustment for ED utilization that could have been diverted to

other settings, potentially preventable hospital admissions adjustment for inpatient utilization).

A risk-adjusted efficiency adjustment process is also used to address differences in claim

levels among MCOs within a region after adjusting for the underlying risk profile of each

MCO’s population. Furthermore, a general ledger review of MCO administrative costs has

been performed recently, and target MCO profit margins (i.e., underwriting gains) were

adjusted to account for lower corporate taxes in Calendar Year 2018.

Contracting, compliance, and performance management

Current managed care contracts establish minimum standards for MCO performance and

attempt to create an environment that fosters innovation. Contract provisions cover areas

including care management, utilization management, provider payment, program integrity,

provider network, grievances and appeals, among others. For care management, Missouri

requires initial screening within 90 days of enrollment, with shorter timelines for pregnant

women, children with elevated blood lead levels, and members with diseases. The current

contracts contain provisions to stimulate innovation and value in the managed care program,

through the Local Community Care Coordination Program (LCCCP) as well as member and

provider incentive programs.

The performance management regime established through current contracts relies primarily

on performance withholds, liquidated damages and sanctions. The performance withhold

program is under revision based on negotiation between MHD and the MCOs, with an

intention to use predominantly HEDIS measures going forward.

158

Liquidated damages for

contract compliance infractions cover a broad set of potential operational issues, with penalties

ranging from $100 per day for failure to submit a report to $10,000 per month for failure to

adhere to claims processing standards.

The performance management relationship between MHD and MCOs centers on ensuring

basic contract compliance and rectifying performance issues. MHD requires 24 distinct reports

from MCOs in addition to submission of encounter data and other information. To date, MHD

activities have focused on improving and validating the quality of the information submitted by

MCOs.

Potential opportunities for improvement

• The rate-setting methodology could be further strengthened. While the current

methodology employs several strong elements to ensure managed care rates account for

all reasonable, appropriate, and attainable costs, opportunities remain to further enhance

the rates. Additional efficiency adjustments are available for each of the major categories

of expenditures to remove inefficient utilization (e.g., inpatient stays that could have been

avoided with better outpatient care) from rate calculations.

159

Steps could also be taken to

simplify the rate cell structure by combining small, high-cost rate cells to reduce potential

welfare programs, and independent foster children ages 18-26; “custodial parents” excludes independent foster

children ages 18-26.

158

Interviews with Medicaid program staff members.

159

Interviews with Medicaid program staff members; analysis of state data, 2018.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

volatility in capitation payments. Finally, as discussed in the acute care section, the state

could consider capping MCO hospital payments at a fixed percentage of Medicaid FFS

payments, while simultaneously adapting overall MCO capitation rates accordingly.

• Day one MCO eligibility and/or passive MCO enrollment could be implemented for

additional populations. Except for foster children, new participants eligible for managed

care will remain in fee-for-service for some time before either choosing or being

automatically assigned to an MCO. States such as Ohio have adopted day one eligibility

with passive enrollment for individuals eligible for Medicaid managed care, while still

preserving a participant’s ability to actively choose or switch MCOs for a period of time.

160

In such states, individuals may be enrolled in an MCO retroactively to the first day of the

month in which Medicaid eligibility is determined. In such states, there is no fee-for-service

period before MCO enrollment occurs. This can reduce the administrative burden and

financial risk to the state and accelerate the process of availing participants to care

management and coordination.

• Operational contract provisions could be further strengthened to improve program

performance, increase efficiency and improve member and provider experience.

Timeliness standards for key processes (e.g., provider payment, prior authorization,

grievances and appeals) can be further specified. Program integrity requirements

(including fraud, waste, and abuse) can be further elaborated to define overpayments to be

investigated and clarify roles (e.g., between the state and MCOs) in preventing, detecting,

recovering and retaining overpayments. The state may also have an opportunity to revisit

provider network and network adequacy requirements, especially considering CMS’

November 8, 2018 notice of proposed rulemaking on Medicaid and CHIP managed

care.

161

• Care management requirements can be further elaborated and appropriately

enforced. The contracts do not clearly specify standards for risk stratification and

identification of participants for care management, the proportion of participants the state

expects to receive care management, case load standards for care managers, or care

management activity requirements once participants are enrolled in care management

programs. While the contracts allow MCOs to coordinate care management activities with

providers including health homes, it does not set forth a clear expectation or requirement

for them to do so. In addition, initial steps to increase healthcare value (e.g., member

incentive programs, provider incentive programs and LCCCPs) have not seen broad

uptake and MCO progress in implementing these programs and realizing their potential for

impact has been uneven.

162

• For a subset of health home enrollees, the state pays both MCOs and health homes

for care management services. Approximately 6,500 MCO members are enrolled in a

Primary Care Health Home (PCHH), and 5,500 members are enrolled in a Community

Mental Health Center Healthcare Home (CMHC HH) or Certified Community Behavioral

Health Clinic Health Home (CCBHC HH).

163

In addition to their regular payments from

160

Ohio Medical Assistance provider agreement for managed care plan. Ohio Department of Medicaid, 2018.

161

CMS, “CMS Proposes Changes to Streamline and Strengthen Medicaid and CHIP Managed Care Regulations”,

2018, see: www.cms.gov/newsroom/press-releases/cms-proposes-changes-streamline-and-strengthen-medicaid-

and-chip-managed-care-regulations.

162

Interviews with Medicaid program staff members.

163

Interviews with Medicaid program staff members.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

MCOs, Missouri health homes receive additional care management payments directly from

the state. Some behavioral health services provided by health home providers are carved

out of managed care, but the responsibility for care management and coordination with

other services is attributed to both the MCO and the health home.

• The state has not fully leveraged the available levers for incentivizing MCO

performance or disincentivizing MCO underperformance. Additional levers are

available to the state to create positive incentives for MCO performance on, for instance,

operational or quality metrics. At present, the auto-assignment algorithm used to assign

participants to MCOs only takes into consideration the level of MCO enrollment in each

region (subject to minimum and maximum enrollment levels for each MCO). Other states

have incorporated operational or quality metrics (e.g., encounter data submission or

provider payment operational measures; HEDIS quality measures) into the auto-

assignment algorithm to reward better performing MCOs with additional participants.

164

Pooled rewards, bonuses, or public report cards could also be considered as additional

performance management levers. Furthermore, while the MCO contracts specify a broad

set of liquidated damages or sanctions for performance infractions, the state could revisit

the structure and magnitude of these penalties to ensure their efficacy, and clearly

communicate to MCOs which areas of performance will be most closely monitored. Exhibit

36 shows the incentive and disincentive levers currently used in Missouri against a broader

set of levers observed in other states, highlighting several opportunities for new levers to

encourage MCO compliance and performance.

164

KFF, “Medicaid Reforms to Expand Coverage, Control Costs and Improve Care: Results from a 50-State

Medicaid Budget Survey for State Fiscal Years 2015 and 2016,” 2015, see: www.kff.org/medicaid/report/medicaid-

reforms-to-expand-coverage-control-costs-and-improve-care-results-from-a-50-state-medicaid-budget-survey-for-

state-fiscal-years-2015-and-2016.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 36: MCO COMPLIANCE AND PERFORMANCE MANAGEMENT LEVERS

165

• Reporting requirements for MCOs can be improved, reducing administrative burden

and improving the value of the information received. Current required reporting

includes seven financial data reports (e.g., unaudited and audited financial statements,

copies of administrative services contracts and management agreements), and 17

operational data reports (e.g., contact center reports, provider network reports, care

management logs).

166

The state does not appear to be fully processing the volume of

detailed data contained in these reports and providing concise, aggregated analysis and

feedback that can drive MCO performance improvement. In addition, in several cases,

ambiguity or disagreement over the type of information required, granularity, or frequency

at which it must be reported have made it difficult to yield reliable data and produce

meaningful insights. These issues have led to challenges in establishing the preconditions

for optimal performance dialogues between the state and MCOs.

• The poor quality of the encounter data limits adequate performance management.

The state’s MMIS system is not able to take in some encounters or encounter data

variables. The quality of MCO encounter data submissions is variable, in part because

encounters are being held back due to issues the MMIS system has in processing

encounters. The result is that the state has neither a complete, accurate set of encounter

data, nor a full understanding of which encounters are not being submitted. Consequently,

the state does not appear to be performing certain analyses on spending or spending

165

Based on review of approximately 15 publicly available managed care contracts across states from 2013-2018.

166

Missouri DSS, “Managed Care Contracts,” 2018, see: dss.mo.gov/business-processes/managed-care/.

Cost/

efficiency Quality Network Pharmacy

Contract

perform-

ance

Data sub-

mission

Program

integrity

APM /

innovation

Member

experienceLever

Contract dimensions

Used by MO

Use pooled rewards

Freeze enrollment

Report health

grades

Increase auto-

enrollment priority

Increase

transparency

Decrease auto-

enrollment priority

Enforce

fines/penalties

Require corrective

action plans

Offer bonuses

Use withholds

Termination

Incentives

Dis-

incentives

Build encounter

data quality and

timeliness into

auto-enrollment

algorithm

Report publicly on

how MCOs are doing

at participation in

their member and

provider incentive

programs

Offer

bonuses to

drive APM

and

innovation

Observed in other states

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

trends, for example, or adequately comparing performance between plans, in ways that

could be beneficial to the management of the Medicaid program, as a whole, and the

managed care program specifically.

• The performance dialogues between the state and the MCOs could be more focused

on the value of the care delivered. The amount of and reliance on detailed process

measures and the poor quality of the encounter data lead to a lack of focus on key

outcomes in the performance dialogues between the state and the MCOs. Recent

interactions between the state and MCOs have focused on improving the timeliness and

validity of information reported, rather than MCO performance on improving quality,

outcomes, and experience for the population. Performance dialogues could be advanced

to cover more substantive, outcome-, and improvement-oriented conversations. This could

be consistent with and supportive of the implementation of value-based payment programs

and reimbursement models that reward quality and outcomes as discussed elsewhere in

this document (e.g., in the Acute Care Services and LTSS sections). The state also has an

opportunity to codify the cadence and approach to performance dialogues with MCOs,

establishing its agenda and priorities for these conversations rather than reacting to MCO

priorities. Exhibit 37 provides a conceptual illustration of the evolution of the relationship

between the state and its MCOs, highlighting potential priorities for more sophisticated

levels of state/MCO collaboration.

EXHIBIT 37: EVOLUTION TOWARD MORE ADVANCED COLLABORATION WITH MCOS

Description

Maturing managed care

Strategic Sourcing

Advanced managed care

Supplier Collaboration

Shift to managed care

Transactional Sourcing

Characteristics

▪ RFPs scored on a set rubric

and awarded to MCOs with

most points

▪ Meets mandatory

requirements

▪ Ad-hoc improvement

initiatives

▪ Compliance-based

performance conversations

▪ Greater focus on quality,

outcomes and member

engagement

▪ Improved integration of care

across behavioral and

physical health

▪ Comprehensive and

deliberate sourcing strategy

▪ Fact-based, holistic

performance conversations

▪ Openness to joint innovation and

collaboration

▪ Partnership to improve care

coordination and integration

▪ Cost and investment transparency

to support shared prioritization

▪ Attention to MCO capability

development

▪ Payment for quality, value and

outcomes

▪ Performance-based partnerships

▪ Advanced analytics to improve

quality and efficiency

▪ Active management across MCO

portfolio

▪ Partner to address non-core

Medicaid policy goals

▪ Structured sourcing and

contract negotiations

▪ Effective program integrity/

performance management

▪ Management of complex

categories

▪ Established path to program-

wide payment innovation

▪ Focused on monitoring

contractual compliance

▪ Dominated by “firefighting”

on unexpected issues

▪ Basic KPIs/performance

management processes

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

• The scope of services covered under managed care for children, parents, and

pregnant women is narrower than that in many other states. For the managed care

population, most prescription drugs and certain behavioral health services are carved

out.

167

A significant majority of managed care states include pharmacy benefits in their

managed care contracts (though different approaches exist to managing rebates,

formularies, and preferred drug lists), and a growing number include a comprehensive set

of behavioral health services. In light of the increasing emphasis on the need to integrate

physical and behavioral health services (including substance use), many states have

decided that a coordination barrier between physical and behavioral health may hamper

the realization of optimal outcomes for patients. Carving in these services can create

additional value (in efficiency, quality, and experience) through integrated care

management across a more comprehensive continuum of services for covered

participants. In addition, moving to value-based payment may also be facilitated by carving

in these services as MCOs would otherwise have different incentives than VBP providers

in making drug or behavioral health treatment choices.

• The state could consider including (a portion of) the ABD population in managed

care. In Missouri, the Medicaid ABD populations remain in traditional (FFS) Medicaid.

While the multiple improvements to efficiency, quality of care and outcomes discussed in

the preceding sections (e.g., care management, rebalancing of the LTSS system,

reimbursement based on quality and outcomes) could be achieved through multiple

models, managed care represents one potential approach to support these efforts. A

transition to managed care could be accomplished through a Medicaid managed care

model that includes only the Medicaid benefits for dual or non-dual eligible ABD

beneficiaries and/or through one of the several available models for integrating Medicare

and Medicaid benefits. Recent guidance from CMS has signaled a renewed focus on

programs that integrate Medicare and Medicaid benefits.

168

As shown in Exhibit 38, a

majority of states now include at least part of this population in managed care.

169

This may

be due to a belief that managed care models present opportunities to improve care

management and thus improve quality, outcomes and experience for this population, while

increasing the efficiency of the program by better managing medical cost trends over time.

While the body of empirical evidence across states to support these claims remains

nascent, several studies that have focused on specific subsegments of the ABD population

(e.g., LTSS recipients or participants with high behavioral health needs) have shown

evidence of the potential for well-designed and implemented managed care programs to

improve program performance.

170

167

Missouri DSS: see note 166.

168

Medicaid.gov, “Ten Opportunities to Better Serve Individuals Dually Eligible for Medicaid and Medicare,” 2018,

see: www.medicaid.gov/federal-policy-guidance/downloads/smd18012.pdf.

169

McKinsey, “Next-generation contracting: Managed Medicaid for individuals with special or supportive care

needs,” 2016, see: healthcare.mckinsey.com/next-generation-contracting-managed-medicaid-individuals-special-or-

supportive-care-needs.

170

McKinsey: see note 169; McKinsey, “Improving care delivery to individuals with special or supportive care

needs,” 2016, see: healthcare.mckinsey.com/improving-care-delivery-individuals-special-or-supportive-care-needs.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 38: MANAGED CARE FOR SPECIAL OR SUPPORTIVE CARE NEEDS

171

Potential initiatives

Based on a review of Missouri’s current approach, interviews with functional leaders and subject

matter experts within the relevant departments, and analysis of other states’ activities, this

section has identified 12 potential initiatives Missouri could consider to improve managed care

in the state. The total potential impact across these initiatives ranges from $175 million to $300

million, which may include the savings from the LTSS section depending on whether the state

fully carves in services for the ABD population

1. Incorporate additional efficiency measures into the managed care rate-setting

process. Three efficiency adjustments have been put into place in the current managed

care rate-setting methodology: 1) removing claims for potentially preventable inpatient

admissions, 2) removing emergency department claims that could have been avoided, and

3) conducting an overall adjustment for risk-adjusted efficiency. These efficiency

adjustments can be continued. In addition, there are several other efficiency adjustments

available that have not yet been employed, covering spending areas such as short-stay

admissions, readmissions and maternity care (e.g., inpatient stays that could have been

avoided with better outpatient care). These additional adjustments would need to be

examined for potential overlap with the adjustments current in place (e.g., risk-adjusted

efficiency, a more broad-based adjustment, may already capture some of the value that

could be captured through new adjustments), but they have the potential to create additional

cost savings for the program.

2. Implement stop-loss provision and combine small rate cells. The current rate structure

contains several small but high-cost, potentially volatile rate cells (e.g., a rate cell for

participants in neonatal intensive care units). The state could consider implementing a stop-

loss provision and combining smaller, more volatile rate cells with larger, more stable ones.

171

McKinsey: see note 169.

LTSS

BH and LTSS

BH, LTSS, and I/DD BH and I/DD I/DD and LTSS

2005

D.C.

2016

D.C.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

This could increase the predictability of state outlays for managed care capitation payments

and simplify administration of the rate structure.

3. Expand day one managed care eligibility and passive enrollment to additional

populations. Day one MCO eligibility and passive enrollment could be expanded beyond

foster children to additional populations. Passive enrollment, while still allowing participants

to choose or switch MCOs as required by federal regulations, can streamline participant

transitions, increase access to care management for participants by requiring it at the time of

Medicaid enrollment, and reduce the burden on the FFS system.

4. Further specify contract provisions regarding key operational processes and

timelines. Contract requirements laying out the process and required decision timelines for

prior authorization, provider payment, and resolution of grievances and appeals could be

clarified and strengthened. In addition, program integrity language can be further elaborated

to set expectations and clarify roles between the state and MCOs for eliminating

overpayments due to fraud, waste, and abuse. These improvements to the MCO contract

could remove ambiguity and improve MCO performance and the state’s ability to monitor

and manage MCO performance against these requirements.

5. Clarify and strengthen care management requirements. The state could enhance care

management requirements by adding specificity around risk stratification and participant

identification, the proportion of participants to receive care management, case load

standards for care managers, and/or care management activity requirements for MCOs. The

state could consider further clarifying expectations for MCOs to collaborate and/or formally

delegate care management requirements to health homes or other care management

entities. This could ensure clarity of roles and prevent against payment by the state for

duplicative care management efforts by multiple parties (e.g., MCOs and health homes).

6. Clarify and strengthen incentive programs and programs intended to encourage

adoption of value-based payment. The state could engage in a focused effort to

collaborate with and manage MCOs in designing and rolling out member incentives,

provider incentives, and LCCCP programs. Depending on the choices the state makes in its

approach to value-based payment, it could incentivize or require MCOs to align or integrate

their efforts with the state’s strategy and include definitions for Alternative Payment Models

in MCO contracts and/or performance management.

7. Deploy additional levers to incentivize MCO performance on key metrics. In addition to

the revisions to the withhold program currently underway, the state can consider additional

levers such as MCO prioritization in the auto-assignment algorithm based on performance,

pooled rewards, bonuses, or public report cards. Expanding the levers in use can enable the

state to incentivize performance across a broader set of metrics covering operational

performance, quality, and healthcare value (e.g., encounter data submission,

member/provider incentive program participation, LCCP or VBP program participation, care

management). If the state were to prioritize improving data submission, it would need to

ensure that remaining obstacles in the state’s encounter data intake process are resolved.

8. Optimize financial penalties to better regulate MCO performance on key metrics. The

state could revisit the structure and magnitude of the sanctions and liquidated damages set

forth in the contract to ensure their efficacy. The state could also more clearly communicate

to MCOs which areas of performance will be most closely monitored in a given time period.

9. Streamline MCO reporting requirements and improve accuracy and timeliness of

information reported by MCOs; establish cadence for performance management

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

dialogues. Accuracy and timeliness of information reported by MCOs could be improved to

enable more informed, focused performance management discussions. This could include

further streamlining of MCO reporting requirements, shifting from a focus on processes to

outcomes based on collaboration between MHD and each of the MCOs. A cadence for

performance management dialogues between the state and MCOs could be established

along with clear priorities and expectations for the topics to be covered in each discussion.

10. Carve in additional services to managed care for the current managed care

population. The scope of services covered under managed care for the current managed

care population could be broadened to include pharmacy benefits and additional behavioral

health services (e.g., those under DMH-administered programs). Including these services

could enhance the MCOs’ ability to manage the overall health and total cost of care for the

managed care population as well as VBP programs, which could help improve quality,

outcomes, and participant experience while increasing program efficiency.

11. Transition to a single-MCO model with specialized capabilities for the foster care

population. The structure of the managed care program for children in foster care or in

subsidized, post-adoption, or guardianship programs could be modified to place this

population into a single MCO offering specialized capabilities, experience and expertise with

this population, potentially procured through a more tailored procurement process. This

could avoid the sometimes-fragmented nature of current services for this vulnerable

population, ensure the application of focused expertise and experience within one MCO and

optimally leverage its infrastructure to meet this population’s needs. Relying on the expertise

of one MCO may also improve the ability of the state to conform to the regulatory

requirements associated with serving this population (e.g. the management of

psychotropics).

12. Expand the scope of the managed care program to include the ABD population (in

whole, in part, or on a phase-in basis). Expanding managed care to portions of the ABD

population represents one potential approach to achieving the improvements to efficiency,

quality of care, and outcomes discussed in the preceding sections, among alternatives such

as improved state-led care management programs or meaningful adoption of alternative

payment models. Expanding managed care to this population would likely require statutory

change and could take many forms given the heterogeneity of the ABD population and the

services required by its various subpopulations. In general, MCO capabilities in serving the

ABD population – and state experience in operating managed care programs for this

population – vary widely by subsegment of the population and associated services.

Managed care programs covering the core medical, behavioral, LTSS, and pharmacy

benefits of non-dual eligible ABD participants are becoming increasingly common, as are

managed care programs focused on covering the LTSS services for dual-eligible

beneficiaries. Meanwhile, managed care programs for persons with intellectual and/or

developmental disabilities (whether residing in an institutional setting or on an HCBS waiver)

remain relatively rare. Any potential consideration of managed care for the ABD population

may take into consideration the diverse and nuanced characteristics and needs of the

various subsegments of this population. Finally, through enrolling elderly and/or dually

eligible participants with disabilities in Medicaid managed care plans, the state could take

advantage of the increased opportunities recently provided by CMS to improve integrated

care for dually eligible populations through, for example, Dual Eligible Special Needs Plans

(D-SNPs) or Medicare Advantage Medicare-Medicaid Plans (MMPs).

172

Medicaid.gov: see note 168.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

PROGRAM INTEGRITY

To maintain the functional integrity of the state’s Medicaid organization, Missouri has divisions

that prevent fraud, waste, and abuse (FWA) and ensure proper payments. Fraud, waste, and

abuse detection and prevention are largely the responsibility of Missouri Medicaid Audit and

Compliance (MMAC) and the Welfare Investigations Unit (WIU), but multiple other agencies

within DSS conduct or enable investigations or enforcement. The WIU is responsible for

preventing participant fraud, while MMAC is responsible for enrolling, auditing, investigating,

and sanctioning providers.

The Cost Recovery Unit administers cost avoidance and a recovery program to offset

expenditures for the state Medicaid agency. This unit ensures that appropriate third-party

resources (including but not limited to Medicare, commercial insurers, workers’ compensation,

probate-estate recoveries, and others) are utilized as the primary source of payment prior to the

state paying for services. Enrollment of eligible participants into Medicare is especially important

for the state as this population typically has more limitations of average daily living, poorer

health, and higher medical expenditures.

When combined, the potential initiatives could save $65 million to $100 million or more, net of

recurring investments, depending on decisions made by the state.

173

Potential initiatives include

enhancing the quality and quantity of FWA claims-based analytics, increasing coordination

between MMAC and relevant internal and external stakeholders, optimizing the identification

and enrollment of Medicare-eligible participants, and improving the implementation of certain

pre-payment edits.

Current situation

This section provides an overview of Missouri’s current FWA organizational structure and

functionality, third-party liability (TPL) identification, Medicare Buy-In, and estate recoveries.

Organizational structure and functionality

There are multiple divisions responsible for conducting investigations or performing

compliance duties within the state. These divisions include Missouri Medicaid Audit and

Compliance (MMAC), the Division of Legal Services (which includes the Welfare Investigations

Unit [WIU] and the General Assignment Unit), Family Support Division, Division of Youth

Services, Children’s Division, and the Division of Finance & Administrative Services.

Missouri handles Medicaid participant fraud through the WIU. WIU deters participant fraud,

prosecutes offenders, and collects money lost to the state because of fraud. The WIU currently

has 18 investigators.

Missouri handles provider fraud and abuse through MMAC. In SFY2018, MMAC produced

about $40 million in savings for the state.

174

MMAC is responsible for enrolling, auditing,

investigating, and sanctioning providers. MMAC is currently appropriated 76.5 FTEs, including

twenty-nine FTEs dedicated to provider audits and participant lock-in and eight investigators.

MMAC works with an analytics vendor and the CMS Unified Program Integrity Contractor to

identify opportunities to improve program integrity. Through its investigations, MMAC provides

feedback to the policy teams within MHD, another unit within DSS.

TPL identification

173

Analysis of Missouri Medicaid claims data, SFY2016-18; interviews with Medicaid program staff members.

174

Analysis of state data, 2018; interviews with Medicaid program staff members.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

TPL information is obtained at the time a participant is enrolled. Family Support Division (FSD)

specialists obtain TPL information primarily during the MHD eligibility determination process.

Supplementing this initial contact are data matches with both private and public entities, edits

within the claims processing system, direct inquiries to participants, non-custodial parents and

other potential liable parties. MHD uses a vendor to perform data matches between participant

claims’ data and external sources of third-party coverage.

TPL identification for participants enrolled into MCOs is the responsibility of the MCO. CMS

recommends states use one of four options to ensure that they meet the coordination/TPL

requirements: (1) exclude individuals with known sources of TPL from enrollment in MCOs; (2)

enroll individuals with known sources of TPL in MCOs, with the state retaining responsibility for

COB/TPL; (3) enroll individuals with known sources of TPL in MCOs and contractually require

that the MCO assume responsibility for COB/TPL; or (4) exclude individuals with commercial

managed care coverage from enrollment in MCOs but enroll individuals with other types of

third party coverage in the MCOs.

175

Missouri uses the third option, and as such, MCOs act as

agents for the state for coordination of benefits and third-party reimbursement in the following

circumstances: workers’ compensation, tortfeasors, motorist insurance, and liability/casualty

insurance. The state’s MCOs are required to report their identified savings and the future

capitation payments are adjusted accordingly.

Dual enrollment and Medicare Buy-In

Missouri is one of nine 209(b) states. At least one of Missouri’s income eligibility criterion is

more restrictive than the SSI program, thus making it one of nine states that are considered

209(b) states (see Exhibit 39).

176

States that elected this option may not use more restrictive

standards than those in effect in January 1, 1972, and must provide for deductions of incurred

medical expenses from income through Medicaid spenddown so that individuals may reduce

their income to the income eligibility level. As a result of being a 209(b), the participant

enrollment process is separate from the SSD/I determination.

175

Medicaid.gov, “Medicaid Third Party Liability & Coordination of Benefits,” see:

www.medicaid.gov/medicaid/eligibility/tpl-cob/index.html.

176

Social Security Administration, “SI 01715.020 List of State Medicaid Programs for the Aged, Blind and Disabled,”

2016, see: secure.ssa.gov/poms.nsf/lnx/0501715020.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 39: MEDICAID-SOCIAL SECURITY ELIGIBILITY AND ENROLLMENT

177

Missouri identifies Medicare leads through three main sources. As a cost-saving measure,

Medicare premiums are paid for participants of Old Age Assistance, Permanently and Totally

Disabled, Aid to the Blind, Temporary Assistance for Needy Families, Specified Low Income

Medicare Beneficiary, and Qualified Medicare Beneficiary programs who meet the criteria for

Medicare coverage. Staff verifies Medicare leads through reports produced from files sent by

CMS, the Social Security Administration, or the TPL/Medicare contractor through a data

match.

178

In addition, the state has a policy that mandates participants apply for Medicaid, and

they must also apply for Medicare.

Estate recoveries

Missouri identifies estate recoveries through data matches from various organizations. The

state uses data from the DHSS’ Vital Statistics, FSD county office staff, and cooperation of

other public and private groups. When cases are established, TPL staff verifies expenditure

documentation and assembles data for evidence. The TPL staff appears in court to testify on

behalf of the state and to explain MHD policies and procedures.

Potential opportunities for improvement

This section identifies potential opportunities to strengthen the state Medicaid agency’s program

integrity. When compared to common practices in other states, the following observations can

be made:

• Improved coordination across multiple agencies could help improve fraud, waste,

and abuse (FWA) efforts. Each division may have its own computer system, eligibility

criteria, provider and participant enrollment service authorizations, service delivery,

payments, audits, investigations, and compliance functions. In addition, divisions with

177

Social Security Administration: see note 176.

178

Missouri DSS, “Third Party Liability,” see: dss.mo.gov/mhd/general/pages/about.htm#tpl.

1634 209(b) SSI

Description

209(b)

states

1634 states

SSI States

There are three classifications for states’

eligibility and Medicaid enrollment

State by state eligibility determination classification

▪ Have at least one eligibility

criterion more restrictive than SSI

▪ Must allow for Medicaid

spenddown

▪ Eligibility determinations are the

responsibility of the state

▪ Have the same eligibility criteria as

SSI

▪ Eligibility determinations are made

by SSA

▪ Have the same eligibility criteria as

SSI

▪ Eligibility determinations can be

made by SSA or the state

▪ States may require separate

enrollment process

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

primary fraud and abuse functions are dependent on staff within the other divisions to

detect potential fraud or abuse situations and make a referral in an acceptable format with

supporting documentation (see Exhibit 40). However, MHD has recently established an

interdisciplinary taskforce to improve coordination of FWA activities.

EXHIBIT 40: DIVISIONS INVOLVED IN FWA EFFORTS AND ASSOCIATED CHALLENGES

• MMAC could increase collaboration with relevant clinical policy teams. A closed-loop

communication system between FWA and clinical policy teams – meaning whenever FWA

is identified, the situation is communicated to the clinical staff – could help shape

corresponding policy changes in a timely fashion. This might also require additional

capacity within the clinical policy teams.

• The analytical capacity and capability, and the range of FWA concepts tested in

Missouri could be increased in line with other Medicaid programs and commercial

plans. The internal analytics function – currently two FTEs – could benefit from additional

capacity. MMAC’s analytics vendor currently tests for between 25 to 40 program integrity

opportunity areas; this funnel could be greatly expanded using a prioritized subset of

opportunity concepts adopted by other programs.

• Ensuring access to a larger set of higher quality data could improve Program

Integrity (PI) performance. MHD current faces challenges in the quality of MCO

encounter data as well as some aspects of FFS data. Approaches to improve this data

quality are described in other parts of this document. In addition, MHD could work with

CMS to access other data sources directly (e.g., Medicare claims) that might be helpful in

PI opportunity identification.

• The state’s enrollment of dual-eligible participants into Medicare is lower than

historic state and national averages. In SFY2013, 16% of Missouri’s participants were

Children’s

Division

Division of

Finance &

Administrative

Services

MMAC

Division of

Legal

Services

Family

Support

Division

Division of

Youth

Services

DSS

Numerous divisions within DSS either conduct FWA

tasks or enable other divisions to do so

Challenges with this model

▪ Each division may use unique software and

databases for determining:

– Eligibility

– Provider and participant enrollment

– Service authorizations

– Service delivery

– Payments, audits

– Investigations

– Compliance functions

▪ Divisions with primary fraud and abuse

functions are dependent upon staff within the

other divisions to recognize a situation could

be fraud or abuse and make a referral in an

acceptable format with their supporting

documentation

▪ Ability to work with data across divisions

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

dually enrolled in Medicare and Medicaid, which was consistent with the national average

at the time. In SFY2018, the state’s dual-eligible enrollment as a percentage of total

participants was 14.5% (see Exhibit 41).

179

The decrease in dual enrollment appears to be

more prominent in the disabled population that is less than 65 years of age.

EXHIBIT 41: CHANGES IN DUAL ENROLLMENT AS PERCENTAGE OF TOTAL MEDICAID

ENROLLMENT

• Missouri could increase the number of sources it currently uses for TPL

identification. Other state Medicaid agencies and CMS have pharmacy claims databases

to identify primary payors. Using pharmacy claims databases typically allows states to

identify an additional cohort of participants who have TPL at a faster rate because of the

faster typical timing of pharmacy claims.

Potential Initiatives

Based on review of Missouri’s current approach, interviews with functional leaders and subject

matter experts, and analysis of other states’ activities, this section includes five potential

initiatives Missouri could consider improving program integrity. In total, the financial impact of

Program Integrity initiatives could range from $65 million to $100 million, depending on state

choices.

1. Expanding the national correct coding initiatives (NCCI) coding edits that the state

has in place. CMS developed the National Correct Coding Initiative to promote national

correct coding methodologies and to control improper coding leading to inappropriate

payment in claims. There are two broad classifications of NCCI edits: Procedure-to-

Procedure edits, which prevent improper payment when incorrect code combinations are

179

KFF, “Dual Eligibles as a Percent of Total Medicaid Beneficiaries,” 2013, see: www.kff.org/medicaid/state-

indicator/duals-as-a-of-medicaid-beneficiaries; analysis of Missouri Medicaid claims data, SFY2016-18; interviews

with Medicaid program staff members.

15.1

14.7

14.0

14.5

SFY2016SFY2015 SFY2017 SFY2018

16%

Lowest Quartile 2nd Quartile 3rd Quartile Highest Quartile

Duals as a percent of total enrollment in SFY2013

Since SFY2013, dual-enrollment has fallen

to 14.5% of the total population

US average: 16%

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

reported, and Medically Unlikely edits, which prevent improper payments when services are

reported with incorrect units of service. Missouri has implemented some of these edits but

has not implemented the full suite of edits. This initiative would require changes to the MMIS

system to implement the full suite of codes, among other requirements.

2. Create an experimental, investigation, and unproven (EIU) medical procedure policy

to prevent improper payments. This agency policy would identify devices or procedures

that have not been proven to be medically effective. This initiative would require the state’s

clinical staff to identify these procedures and review the procedure on annual basis.

Additionally, the initiative would require feasible changes to the MMIS system.

3. Expand the analytical funnel to identify additional improper payments that can be

prevented using claims edits and pre-pay changes or can result in recoveries. As an

example, for given procedure codes, Missouri has set billing limits that the state only

reimburses hospital observation stays for up to 24 hours. The state could ensure that the

claims system is preventing payment for procedures after the allotted 24-hour period.

Opportunities that take the form of edits would require feasible changes to the MMIS

system.

4. Optimize the state’s ability to identify and enroll participants who are currently and

may become Medicare eligible. The state could implement (either internally or through a

vendor) new claims-based technology that would allow the state to identify participants who

are currently Medicare-eligible or may become eligible. Missouri staff could then help notify

participants about this benefit. This would require medium-complexity changes to the MMIS

system and potentially a new vendor.

5. Improve TPL identification. Missouri could begin to utilize additional sources (e.g.,

pharmacy claims data) to increase TPL identification rate. To do this, the state could

contract with a vendor that would add additional sources of data.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

FEDERAL FINANCING

Missouri has captured a significant share of the federal funding it is eligible for, but there may be

additional opportunities to capture federal revenue through new federal programs, both through

grants and enhanced match. The state also could consider evaluating the use of inter-

governmental transfers (IGT) as an alternative or supplemental financing approach. The total

federal financing opportunity is expected to be $10 million to $20 million in grant funding and

additional enhanced match.

180

This section will describe the current state of federal financing in Missouri, observed

opportunities for improvement, and potential initiatives for the state to consider.

Current situation

Overall Medicaid spending across departments

Medicaid spend represents over 80% of the budget for DSS and approximately two-thirds of the

budget for DHSS and DMH (see Exhibit 42).

181

The largest areas of Medicaid spending include

managed care, pharmacy reimbursement, hospital and nursing facility reimbursement, physician

reimbursement, and community programs. Nearly all these funds receive some form of federal

match based on the category of spending (see Exhibit 43). Federal funds represent

approximately 65% of the total spending across top Medicaid spending categories.

EXHIBIT 42: MEDICAID SPENDING BY DEPARTMENT

180

U.S. Congress, “H.R.6- SUPPORT for Patients and Communities Act,” 2018, see: www.congress.gov/bill/115th-

congress/house-bill/6; CMS, “Maternal Opioid Misuse (MOM) Model,” 2018, see:

www.cms.gov/sites/drupal/files/2018-10/10-23-

2018%20Fact%20Sheet%20Maternal%20Opioid%20Misuse%20%28MOM%29%20Model%20%28FINAL%29.pdf;

CMS, “Integrated Care for Kids (InCK) Model,” 2018, see: www.cms.gov/newsroom/fact-sheets/integrated-care-

kids-inck-model.

181

SFY2018 available appropriations: Office of Administration, “Fiscal Year 2019 Budget Priorities,” 2019, see:

oa.mo.gov/sites/default/files/FY_2019_Budget_Summary.pdf; Medicaid spend: Missouri DSS, see note 1.

32%68%DMH 2,158

DSS 9,32485% 15%

DHSS 1,41366% 34%

Non-MedicaidMedicaid

Total Medicaid spend and department budget appropriation, SFY2018, USD, millions

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 43: DSS MEDICAID SPENDING

182

Unmatched spending across departments

In DSS, only a handful of narrow categories do not receive federal funds. Some examples of

these categories are Medicare buy-in, through which the state helps pay Medicare premiums

for Medicare Part A and Part B for qualified individuals; state-only assistance, which includes

social services block grants; Temporary Assistance for Needy Families (TANF) grants; and

State General Fund. In addition, DHSS and DMH have more categories of unmatched

spending, although the value of unmatched dollars is a small percentage of the total spending

across the departments. Within DHSS, there could be potential to receive additional match for

Alzheimer’s services and communicable diseases, and within DMH, opportunity exists in

autism spending, crisis intervention services (24-hour hotline and mobile outreach for

psychiatric patients, although only outreach could be eligible for match), and some emergency

room enhancements (ERE).

Other state funds

Provider taxes contribute $1.4 billion to the state Medicaid program, of which $1.1 billion is

derived from hospital taxes and $0.3 billion from nursing homes and pharmacies. The hospital

tax (>5.5%) and nursing home tax ($13.40 per patient day) rates are both high compared to

other states.

183

The use of intergovernmental transfers (IGTs) is limited.

182

Office of Administration and Missouri DSS: see note 181; Missouri DSS, “MHD-FY 18 MMIS Expenditures Final

8-13-18,” 2018; interviews with Medicaid program staff members.

183

Missouri DSS, see note 11.

Federal matching, %

Medicaid spend by department – DSS , USD, millions

108

Managed Care

Pharmacy – Medicare Part D

Pharmacy

449186

1,256

168

130

Nursing Facilities

Physician

Rehab & Specialty

Hospital Care

MMIS

Other

1,961

178

1,274

298

704

564

506

835

212

481

331

425

Premium Payments

364

328

259

280

254

State - GR FederalState - Other

64%

66%

64%

65%

60%

66%

69%

54%

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Potential opportunities for improvement

This section identifies potential opportunities to improve Missouri’s current approach to federal

financing. The opportunities are not intended to be mutually exclusive: potential savings

identified in individual opportunities may overlap with those identified in others.

• Missouri could consider leveraging new federal programs that provide federal

funding for innovative Substance Use Disorder/Opioid Use Disorder (SUD/OUD) and

behavioral health models. A variety of new funding opportunities have recently been

made available to states, including CMMI grants for the design of alternative payment

models, guidance from CMS on additional demonstration opportunities that grant

increased flexibility in how Medicaid funds are used, and the wide-ranging funds made

available to a variety of agencies through the SUPPORT for Patients and Communities Act

(SUPPORT Act). Exhibit 44 contains a breakdown of different sources of non-federal funds

for Medicaid payments.

EXHIBIT 44: SHARE OF NON-FEDERAL FUNDS FROM DIFFERENT SOURCES

184

Potential initiatives

The state could consider several potential initiatives to improve its federal financing. The total

opportunity could be $10 million to $20 million in grant funding and additional enhanced match,

depending on decisions made by the state.

185

These initiatives address overlapping populations

and provide different types of funding (grants, enhanced match, regular match for new sets of

services). The state could consider strategically combining initiatives to maximize efficiency and

184

MACPAC, “The Impact of State Approaches to Medicaid Financing on Federal Medicaid Spending,” 2017, see:

www.macpac.gov/wp-content/uploads/2017/07/The-Impact-of-State-Approaches-to-Medicaid-Financing-on-Federal-

Medicaid-Spending.pdf.

185

U.S. Congress, CMS: see note 180.

70%

72%

80%

36%

22%

10%

19%

32%

16%

14%

44%

42%

DSH payments

Managed care

payments

Fee for service

Payments

Total

Non-DSH

supplemental

payments

State funds

Other sourcesProvider taxes and donations

Funds from local governments

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

generate funding to support the design, development, and implementation of the models as well

as the associated care delivery costs.

1. Access enhanced match by strengthening SUD focus in health homes. While Missouri

has exhausted the eight quarters of enhanced match for the health home program, the

SUPPORT Act allows for the creation of a new SUD-focused SPA that would cover 10

quarters of enhanced match for individuals with SUD not previously covered under a health

home. There are three groups of individuals whom the state could potentially consider as

part of a new SUD-focused SPA: (a) participants with SUD who meet the existing health

home criteria but were never successfully engaged (e.g., no payment occurred for those

participants); (b) participants with SUD who are newly eligible and meet the existing health

home criteria; and (c) participants who are not eligible under the current criteria but would be

eligible if the state created additional eligibility pathways for the SUD population (e.g.,

making receipt of MAT a qualifying factor, creating an eligibility pathway for pregnant women

with OUD). The state would need to meet reporting requirements outlined in the SUPPORT

Act (e.g., quality of care reporting, reporting of costs of individuals in health homes).

186

2. Pursue a State Plan Amendment to access federal funds for SUD services provided in

IMDs. Missouri may be able to leverage the Amendment to the IMD Exclusion to use federal

funds to pay for treatment services in residential settings that qualify as IMDs. To access the

funds, Missouri would need to design a program emphasizing quality and value. Missouri

could consider working with CMS to develop a State Plan Amendment (SPA) initiating the

program; this SPA could potentially be effective as early as October 1, 2019. As the services

currently are not provided, this initiative would be an investment which the state could

consider as part of a value-based program, for example, to reduce total cost of SUD care.

3. Apply for the Serious Mental Illness/Severe Emotional Disturbance (SMI/SED)

demonstration through a Section 1115 Waiver. The SMI/SED demonstration allows

states to use federal funds to pay for treatment services in residential settings that qualify as

IMDs for individuals with SMI/SED.

187

To access the funds, Missouri would need to design a

program emphasizing quality and value that meets budget neutrality requirements for a

Section 1115 Waiver. Missouri would be expected to achieve a statewide average length of

stay of 30 days for participants receiving care in IMDs. Additional analysis would be required

to understand the net budgetary impact of funding for SMI/SED services provided in IMDs.

4. Apply for CMMI grant funding through the Maternal Opioid Misuse (MOM) and

Integrated Care for Kids (InCK) Models. The models provide states with federal funds to

help the state fund programs to combat OUD in pregnant and postpartum Medicaid

participants and improve behavioral health care for children up to 21 years.

188

MOM

provides up to $64.5 million nationally for implementation, transition, and milestone funding

distributed across up to 12 states; InCK provides $16 million

189

. The state could consider

applying for these grants, which could be (but need not be) seen as two sides of the same

coin. The Notices of Funding Opportunity (NOFO) for both programs are expected in early

2019; applications for funding are likely to be due early in 2019 and funds awarded in late

2019.

186

U.S. Congress: see note 180.

187

U.S. DHHS, “Section 1115 Demonstration Process Improvements,” 2017, see: www.medicaid.gov/federal-policy-

guidance/downloads/cib110617.pdf.

188

CMS, see note 180.

189

CMS, see note 180.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

MEDICAID MANAGEMENT INFORMATION SYSTEM (MMIS)

The current MMIS is a set of ~70 integrated components that plays a fundamental role in most

of functions of the Missouri Medicaid program. Its “core” is a 1979 mainframe system,

maintained and operated by Wipro; a set of additional components are maintained and operated

by Conduent. Three main improvement opportunities were identified. First, while the level of

spending on technology is not misaligned with the needs of a Medicaid system of Missouri’s

size, the functioning of the technology does not meet current or future needs. Its limited

functionalities and the antiquated architecture underlie several of the program’s performance

challenges identified throughout this report. Second, there is an opportunity to increase

alignment between program strategy and the Information Systems group’s (IS) strategy. The

current MMIS replacement plan does not include the specificity required to ensure that the

forthcoming modules will meet future needs. Third, the IS group lacks the range of capabilities

needed to ensure an MMIS replacement trajectory that will deliver that future functionality.

Within this context, the IS group has taken important steps, such as the creation of an outline of

a strategic plan for modular replacement and the prioritization of an Enterprise Datawarehouse

(EDW) and Business Intelligence System (BIS). However, the state could consider a concerted,

integrated effort to set up the MMIS for success. A full end-to-end plan could further define the

current roadmap for modular replacement based upon the functionalities most needed from the

perspective of the future Medicaid program, including prioritization and specified use cases. If

the state would decide to take such an approach, it could consider integrating strategic program

priorities, operating models, capabilities, governance, and environment (e.g., procurement,

FMAP) into the updated end-to-end plan.

Current situation

This section describes the MMIS, the in-flight initiatives and the plans for future improvements

and its costs.

MMIS definition

MMIS includes the Core system operated by Wipro, the Clinical Management Services &

System for Pharmacy Claims & Prior Authorization (CMSP), and the Program Integrity solution

operated by Truven. The MMIS is not managed by the Information Technology Services

Division (ITSD) of the Office of Administration but rather by MHD, with a few notable exceptions:

Financial Cycles and Federal Financial Reporting is managed by Division of Finance and

Administrative Services (DFAS), HCBS Assessment is managed by DHSS, and Provider

Enrollment and Program Integrity is managed by MMAC. The Eligibility Determination systems

(managed by Family Support Division [FSD], Division of Youth Services [DYS], and Children’s

Division [CD]) and Claims Pre-Processing and Adjudication (managed by DMH and DHSS)

were excluded from the analysis.

The “MMIS” refers to a disparate range of technologies that are integrated. The components

tend to be named for the function they support, and they can include staff or vendor staff

activities associated with the technologies. There are also components that largely consist of

(vendor) staff activities rather than technology, as when for example several components are

managed by the same vendor. This broad definition can and does cause confusion in strategic

and tactical discussions, where what “MMIS” means may vary amongst those involved.

Functional and Technical

Missouri’s MMIS consists of a collection of technologies that include ~70 components

supporting a broad range of administrative functions of the Missouri Medicaid Program. These

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

components are supported by three vendors. Wipro manages the core MMIS IBM mainframe-

based system, programmed in COBOL (~7 million lines of code)

190

, originally installed in 1979.

Conduent manages the as-a-service CMSP system

191

which is heavily interconnected with the

Core and supports Care Quality Solutions (inpatient certification, reporting, and provider web

tool), Prescription Delivery (clinical decisions for claims processing and clinical edits), and the

Health Information Exchange. Truven manages the current Program Integrity system

192

which

is interconnected with the Wipro system.

Some of the components are managed by a single vendor while others have shared vendor

responsibilities. Given the architecture and history of the system, many of the ‘components’ are

not partitioned, distinct subsystems but may be highly intertwined within the COBOL code. To

help the planning for future modular replacement, the Information Systems groups has classified

the existing components into the below 11 functional categories.

1. Core Claims / Encounter Processing (administered by MHD): supports core Medicaid

functions such as FFS claims processing, participant web portal, and financial management

(e.g., calculation and transmission of payments, provider specific taxes and

reimbursements, financial summaries). Wipro manages the majority of this functionality

while some components are maintained by ITSD (Medicare buy-in and premium collections

and spend down) and Conduent (participant web portal).

2. Pharmacy and Drug Rebate (administered by MHD), operated by both Conduent and Wipro.

Pharmacy, clinical adjudication, and preferred drug list are operated by Conduent. These

components include pharmacy functions such as managing participant pharmacy benefits,

maintaining and applying the drug formulary, and performing pharmacy pre-certification.

Wipro operates several functions including drug rebate processing and initial claims

validation and pricing.

3. Pharmacy Administration (administered by MHD): primarily manages the drug formulary and

setting supplemental drug rebate amounts. These services are operated primarily by

Conduent.

4. Provider Enrollment (administered by MMAC): is responsible for enrolling, screening, and

monitoring both FFS and managed care providers. It is maintained by Wipro, which

subcontracts parts of this work to LexisNexis.

5. Managed Care Enrollment Broker (administered by MHD): these components are

responsible for the enrollment of managed care participants in plans through a web portal,

associated physical mailings, auto-assignment, and a contact center. The web portal and

auto-assignment system are maintained by ITSD, while the contact centers and any

physical mailings are operated by Wipro.

6. Contact centers (administered by MHD): supports both the provider relations, participant,

and prior authorization contact centers. The provider contact center covers inquiries from

providers around program policies, claim assistance, and claim processing instructions,

while the participant contact center covers inquiries from participants about eligibility,

spenddown, and covered services. The prior authorization contact centers support prior

authorizations over the phone. The Contact centers category is operated by Wipro and

includes the phone system and contact documentation software.

190

This system is hosted by Wipro in a data center in Omaha, NE.

191

This system is hosted by Conduent in data centers in East Windsor, NJ; Sandy, UT; and Richmond, VA.

192

This system is hosted by Truven in a private cloud.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

7. Data, Analytics, and Reporting (administered by MHD): supports the Medicaid program’s

needs to access, analyze, and report on data stored in the MMIS. The current system is

primarily focused on providing the required outputs to Transformed Medicaid Statistical

Information System (T-MSIS)

193

federal financial reporting, and program reporting. Both

Wipro and Conduent manage various analytics and reporting components, aligned with their

business functions.

8. Program Integrity (administered by MMAC): supports the detection of potential Medicaid

fraud, waste, and abuse (FWA) through the analysis of claims data. It is currently operated

primarily by Truven Health Analytics and consists of Data Pro (which runs state-defined

algorithms to detect possible FWA and provides ad hoc reporting) and Truven Advantage

Suite (which provides dashboard reporting). Additionally, the surveillance and utilization

review components are operated by Wipro.

9. Health Information Network (administered by MHD): covers the connection of MMIS to

various Health Information Networks (HINs). The platform, maintained by Conduent, is in

place but is currently not connected to any HINs, pending contract negotiations between the

state and HIN(s).

10. Prior Authorization (administered by MHD): encompasses the automated PA system, the PA

web portal, participant case management, and the processing of various prior

authorizations. The web portal is operated exclusively by Conduent and the case

management and prior authorization processing is handled by both Wipro and Conduent,

with the exception of dental, physician, audiology, and out-of-state services, which are

handled by Wipro.

11. Ancillary / Supporting Services (administered by MHD): these include cross-cutting

components that support other components, such as printing and mailroom, help desks, and

project management. These services may be shared across functions and vendors.

Current replacement plans

CMS has issued guidance for the replacement of MMIS, outlining the criteria for which states

can be eligible for 90/10 federal match of MMIS replacement initiatives.

194

These guidelines

emphasize a modular approach to the acquisition of MMIS modules to encourage reuse,

reduce the need for customization, stimulate and expand the vendor landscape, grow adoption

of shared services, and reduce overall MMIS cost. To meet the CMS criteria, the 11 categories

outlined above are identified as the modules in which Missouri has organized the ~70

functions of its current MMIS.

As an overall business strategy for the Missouri Medicaid program is not clearly defined and

integrated with IS’ plans, the detailed three- to five-year end-to-end approach for MMIS

modular replacement (e.g., the timing and requirements for specific modules beyond currently

planned initiatives) has not yet been fully determined.

193

T-MSIS is a data set that CMS requires states to submit which includes data such as: utilization and claims data,

enhanced information on beneficiary eligibility, managed care data, and Medicaid and CHIP expenditure data, see:

www.medicaid.gov/medicaid/data-and-systems/macbis/tmsis/index.html.

194

CMS, “Mechanized Claims Processing and Information Retrieval Systems-Enhanced Funding,” 2016, see:

www.medicaid.gov/federal-policy-guidance/downloads/smd16004.pdf; CMS, “Mechanized Claims Processing and

Information Retrieval Systems – APD Requirements,” 2016, see: www.medicaid.gov/federal-policy-

guidance/downloads/smd16009.pdf; CMS, “CMS-2392-F Mechanized Claims Processing and Information Retrieval

Systems – Modularity,” 2016, see www.medicaid.gov/federal-policy-guidance/downloads/smd16010.pdf.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

The state has started the replacement process with two modules identified as priorities. CMS

has approved the Advance Planning Documents (IAPDs) and contracts have been awarded in

April/September 2018 for a Program Integrity solution and a Business Intelligence

Solution/Enterprise Data Warehouse (BIS/EDW). To support a Medicaid transformation effort,

providing access to MMIS data in a timely and efficient way is considered important. The

BIS/EDW, therefore, is considered a foundational module

195

. The Program Integrity solution

has been contracted to replace current functionality and facilitate incremental improvements.

Cost

Total SFY2019 projected MMIS spending is $85 million, comprised mainly by two main

contractors: Wipro and Conduent ($51 million [60%] and $17 million [20%], respectively).

196

project spending is set to increase from $65 million (SFY2018) to $93 million (SFY2020)

197

driven primarily by net-new spending, such as Truven spending on BIS/EDW and PI (the start

of the modular replacement), a managed care pilot, T-MSIS, Health Information Network

connection, Electronic Health Records, and HCBS support. Current system costs for MMIS

and CMSP remain stable. Projected MMIS spending from SFY2018-SFY2020, by vendor, by

funding source, and by spend type is shown in Exhibit 45.

The SFY2019 weighted average federal match is 73%, up from 69% in SFY2018, and it is

projected to increase to 75% in SFY2020, driven by increased match on implementing new

MMIS modules.

198

In SFY2019, this is projected to split into 90% match for design,

development, and implementation activities (DD&I) ($20 million, or 24% of total); 75% match

for maintenance and operations activities (M&O) ($48 million, or 56% of total), and 50% match

for administrative activities ($17 million, or 20% of total). When the implementation activities

have been completed, funding for components will shift to 75/25. Overall spending on these

items is expected to decrease as the activities shift from DD&I to M&O. From SFY2018-

SFY2020, administrative costs are projected to remain stable, thus decreasing as percentage

of total cost as total cost increases.

195

Interviews with Medicaid program staff members.

196

Missouri DSS, “MHD – FY 18 MMIS Expenditures FINAL 8-13-18,” 2018; Missouri DSS, “MHD – FY 19 MMIS

Expenditures 10-31-18,” 2019.

197

Missouri DSS, “MMIS Spend Plan FY20 Compare to FY19,” 2019.

198

Missouri DSS: see note 197.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 45: PROJECTED MMIS SPENDING THROUGH SFY2020

Potential opportunities for improvement

This section highlights observations based on a high-level assessment of the current state.

• The current MMIS is insufficient for current and future needs, and the antiquated

mainframe technology poses a risk to the program. The limitations of the current

systems prevent Missouri Medicaid from operating at maximum efficiency. Examples

surfaced in other areas of this assessment of the Medicaid program include the following:

the MMIS system rejecting certain MCO data submissions resulting in incomplete MCO

data; incomplete encounter data limiting the effectiveness of MMAC audit reviews; difficulty

in transferring MCO encounter data into MMIS likely resulting in increased error rates;

multiple-format data entry required for pharmacy rebate information increasing rebate

processing time; challenges in eligibility determinations with MEDES data for MCOs

impacting participants’ ability to enroll; difficulty in adding new data fields to the proprietary

layout; and challenges to identifying certain eligibility categories reducing ability to correctly

identify CHIP-eligible children.

To support new initiatives coming from other topical areas, changes would be required in the

MMIS, sometimes with difficult tradeoffs. Generally, there are three categories into which

new initiatives fall.

― Narrow configuration / code changes, where system changes to support an initiative can

be directly made in the current system with minimal disruption. Examples include

additional clinical edits to allow for pharmacy policy adjustments, automation of claim

adjudication driven by lab testing diagnoses, or modifications to MCO auto-assignment

and lock-in logic to expand day-one managed care eligibility and passive enrollment.

$69.7

(75%)

$44.8

(69%)

$19.7

(31%)

$62.6

(73%)

$22.9

(25%)

$22.6

(27%)

$64.5

$85.2

$92.6

$12.7

(20%)

$49.9

(77%)

$50.6

(59%)

$9.1

(11%)

$1.9

$16.7

(20%)

$8.8

(10%)

$51.2

(55%)

$17.6

(19%)

$12.7

(14%)

$11.1

(12%)

$64.5

$85.2

$92.6

$50.6

(55%)

$16.1

(25%)

$45.3

(70%)

$26.9

(29%)

$3.1

(5%)

SFY2018

$47.6

(56%)

$17.4

(20%)

$20.2

(24%)

SFY2019

$15.1

(16%)

SFY2020

$64.5

$85.2

$92.6

OtherConduentWiPro IBM / Truven

Federal State

Vendor

($M)

Funding

source

($M)

Spend type

($M)

Admin (50/50) M&O (75/25) DD&I (90/10)

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

― Limited workaround, where workarounds may exist separate from the system that can

be implemented to support new or updated functionality with relatively few changes

made in the current MMIS system. These workarounds range from a separate

software/service to support VBP analytics, a standalone data intake system to augment

current encounter data, to a supplemental submission flow for X12 data.

― Extensive workaround or rebuild, where a workaround is theoretically possible, but may

involve altering so much of the existing MMIS system that it is worthwhile considering

rebuilding the functionality in a modular replacement instead. Changes this intensive

include drug-level pricing for 340B drugs, DRG classification and payment processing,

and additional rate cells for MCO ABD carve-in.

The state could consider prioritizing the first two categories, as the return on investment of

making these changes is likely better than for the third category. Changes in the third

category could involve a tradeoff between a temporary, possibly costly and higher-risk

investment in the current system while the desired functionality could also be implemented

as part of a forthcoming modular replacement.

Additionally, the decades-old mainframe technology poses a risk to future Missouri Medicaid

success. It is increasingly difficult to maintain the aging core mainframe technology, as the

needed expertise and talent in the marketplace are decreasing. Depreciating technology is

supported by fewer and fewer vendors due to market trends.

Monolithic mainframe systems lead to a lack of agility to make changes in one part of the

system without risking impact to others, as it is especially difficult to fully trace the impacts of

changes in a non-modular system. Ultimately, the risk of mainframe impact inhibits the

ability to quickly make system changes. For example, to make a small update, code

changes may be required in another 50+ locations which may not be simple to edit.

Current MMIS module offerings (such as an off-the-shelf Pharmacy Benefits Management

solution) are built on more modern generations of technology or delivery models (virtualized,

containerized data centers, or cloud services), further risking interoperability between the old

MMIS and new modules.

• Lack of alignment and coordination between Information Systems (IS) and the

Medicaid program. Interviews with IS staff indicate that there is currently no structural

process to incorporate IS in strategic program decisions or to maintain adequate program

awareness of IS challenges and opportunities. In addition, the IS department’s strategic

plan does not appear to be well aligned with program strategic priorities and outcomes, as

these have not been well established, resulting in a lack of detail and prioritization of

program initiatives.

199

The lack of alignment may limit the ability of IS to create MMIS-

related procurements with specific, precise program goals.

• MHD’s IS organization does not map to the needs of a next-generation, modular

MMIS. Based on discussions with IS staff as well as expert interviews across different

states, Missouri’s IS organization appears to not be appropriately staffed and structured to

handle both the day-to-day maintenance and operation of the present MMIS as well as the

planning, implementation, and certification of a new, modular MMIS. Across four areas,

staff and interviewees noted that the IS department lacks key capabilities to support the

new requirements of a modular MMIS.

199

Interviews with Medicaid program staff members.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

― Technology: Covers the capability to define a technical architecture both between and

across vendors, manage several simultaneous procurements and implementations, and

sufficiently understand business process implications on the IS group. Lack of

capabilities within the technology area can result in systems which are not built toward a

centralized architecture, interruptions in current system maintenance, and delays in the

procurement and certification processes. IS currently has one dedicated technical

resource and does not appear to have resources to cover the additional activities

currently slated for the modular replacement. Examples of positions that may provide

these capabilities include technical architects, system operations managers, and

technical managers / project managers.

― Data: Covers the capability to articulate a data governance strategy, align data

management strategy to program goals, and translate between program requirements

and data environment. Missing these capabilities can lead to issues with data quality,

management, and governance. IS does not currently have any resources dedicated to

data capabilities. Positions that could provide these capabilities might include data

architects, and data scientists / engineers.

― Contract management: Covers capabilities such as technical assessment of bid

responses, alignment between RFP/contract outcomes to program goals, and

certification of multiple modules simultaneously. Without these capabilities, contracts

may lack conciseness and precision, risking suboptimal functionality. Additionally,

certification cycle time may increase without appropriate certification resources. IS

currently does some limited contract management through OA and its project managers,

but OA does not currently have dedicated resources to compose xAPDs (PAPDs,

IAPDs, and OAPDs). Example positions that could provide these capabilities include:

contract managers, xAPD writers/budget managers, and dedicated resources for various

contracts and certification tasks.

― Vendor management and accountability: Covers capabilities such as holding vendors

accountable to contract and program outcomes/deadlines and facilitating cross-vendor

cooperation. Without these capabilities, the risk of vendors delivering suboptimal or

incompatible functionality, or failing to meet milestone deadlines, is increased, especially

as the number of vendors is likely to increase. IS currently has a limited number of

project managers but does not appear to be sufficiently resourced to support upcoming

modular replacement activities. Additional project managers and a clear governance

structure (including who is making decisions regarding areas such as infrastructure,

policy, or participant interactions) is an industry best practice without which vendors may

not be able to align to a unified vision and work cohesively.

Potential initiatives for consideration

1. Improve alignment between IS and program. Missouri could consider adopting the

following industry best practices in place in many other states.

The state could include an integrated perspective across both IS and the program in both

strategy development / planning and day-to-day operations. This would include having IS

representation at key program meetings to advise on technical implications and feasibility of

various program decisions. In this way, IS would be able to inform and advise on

implications of program decisions, introduce novel ideas, provide insights in IS-driven needs

as well as opportunities ahead of time. Additionally, it would allow IS to keep business

informed about in-flight initiatives to take into account during program decision-making.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Additionally, the state could conduct joint planning exercises to ensure that IS timelines are

in accordance with program-desired outcome delivery dates of both technical and functional

requirements. These exercises could also help IS explain the choices that the program may

have to make and the implications of those choices. These decisions, which tend to be

made by the program, will impact the delineations between modules as well as the

sequencing of module implementation. Holding these planning exercises may help tighten

the feedback loop for IS to explain the choices that the program may have to make and the

implications thereof.

Lastly, the state could consider including specific desired functional/program outcomes in

procurement documents (e.g., RFPs, vendor contracts) such that vendors are operating

against both technical specifications required by IS and functional specifications required to

drive targeted, prioritized program outcomes. The additional specificity may help ensure that

IS day-to-day tactical actions are more closely aligned.

2. Evaluate the current modular replacement strategy and define an updated strategy

informed by clear strategic direction from the program and reflecting better alignment

to the market, other states, and CMS. First, Missouri could reevaluate the structure of

modules used in the current replacement plan, realign it closer to the modules

recommended by CMS and those utilized in other states further along in their MMIS

modernizations, and map to solutions offered in marketplace. Additionally, finalized modules

may be aligned to program priorities (e.g., the decision to carve pharmacy in or out of

managed care would alter the future Pharmacy module). To create an illustrative example of

a potential module alternative, several interviews were conducted with experts both in

Medicaid and in the MMIS industry to understand the common module structures and

market offerings. In Exhibit 46, a sample alternative module option is displayed, along with

Missouri’s current module structure as well as the common marketplace modules.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 46: ILLUSTRATIVE EXAMPLE OF ALTERNATIVE MODULES

Next, the state could conduct further rigorous planning to help create a roadmap aligned to

program priorities with IS input on feasibility. Four sample evaluation criteria were created

that could be used to develop a heatmap of module priorities: program

priority/value/service delivery strategy (e.g., impact of module and functionality on program

priority and service delivery strategy, specific program outcomes driven by the module, and

urgency and criticality of value unlocked by module), solution availability (e.g., maturity and

competitiveness of marketplace, variety of marketplace solutions available), resources

(e.g., available resources to dedicate, potential upfront and ongoing costs), and complexity

(e.g., dependencies on upstream, downstream, other modules, or other departments and

change required).

Through a sample planning exercise conducted with IS staff, each module from the

potential alternative module option was evaluated against the sample criteria, considering

factors such as updated vendor landscape information, incompletely defined program

decisions, populations affected by module change, and ongoing procurements in other

departments. The preliminary illustrative roadmap created (shown below in Exhibit 47)

could be updated as program priorities are clarified and strategic decisions (e.g., pharmacy

carve in/out, additional managed care population carve-in) are made by the program.

Additionally, any roadmap could be validated at multiple levels and pressure-tested for

feasibility, with many dependencies (e.g., vendor responses, CMS approvals).

Core MMIS/Encounter

Prior Authorization

Pharm./Drug Rebate

Pharm. Admin

Enrollment Broker

Call Centers

Provider Enrollment

Program Integrity

BIS – EDW

Health Info Network

Ancillary Services

MO Modules

Core MMIS/Encounter

Pharmacy

Provider Management

Program Integrity

BIS – EDW

Third-Party Liability

Care Management

Prior Authorization

System Integrator

Financial Man.

Enrollment Broker

Call Centers

Health Info Network

Commercially available

vendor solutions

Core MMIS/Encounter

Prior Authorization

Pharmacy

Enrollment Broker

Member Call Center

Provider Management

Program Integrity 1

BIS – EDW

Health Info Network

Ancillary Services

Third-Party Liability

Care Management

System Integrator

Program Integrity 2

Potential Alternate

Potential Alternate Rationale

Third-Party Liability:

Consolidated standalone solutions

exist

Care Management: off-the-shelf

integrated solutions available, split

from Prior Authorization

Pharmacy: Best practice across

states to have single pharmacy

module

Call Centers: Best practice is to

have member call center

outsourced and provider call

center bundled with claims

processing or provider

management

Program Integrity 2: Next-

generation capability added in a

new module separate from

Program Integrity 1 module

System Integrator: Future

requirement due to increase in

vendor complexity

Ancillary Services: Services not

directly related to technology

which should be handled

separately

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 47: ILLUSTRATIVE MODULE SEQUENCING

Based on clarified program priorities and a value assessment of the current modular

replacement strategy, the state could then define their clear, updated, end-to-end strategy.

3. Strengthen IS capabilities through hiring, partnering for talent, and

retraining/upskilling. DSS could consider prioritizing upskilling IS to complement the

currently available skill sets with capabilities focused on technology, data, contract

management and vendor management and accountability. Training, hiring, or outsourcing

individual expertise are all possible routes towards this goal.

4. Optimize insourcing vs. outsourcing. Increasingly, Medicaid leaders across the country

are confronted with the need to make informed decisions about what MMIS activities to keep

in-house and what to outsource. This decision is particularly critical given that most

agencies are making greater use of managed care, implementing value-based purchasing at

scale, and/or replacing the business information system platforms they use for eligibility

determinations, claims processing, and provider management. All these changes have

significant impact on the component’s required functionalities. CMS guidance would suggest

that a best practice is to keep policy and infrastructure-related decisions in house, allowing

for additional oversight and agility. Other activities, such as handling participant interactions

may be more efficiently outsourced or delivered through a hybrid model. To make this

determination, the state could evaluate factors such as strategic priorities, existing talent,

and vendor availability.

High complexity, with

interdependencies or upstream

milestones required

Relatively standalone,

moderate criticality and

difficulty

Critical needs

or quick wins

Foundational, relatively

independent

(In-progress)

Case Management

(DMH): In-progress of

creating / releasing RFP,

TBD

▪ BIS – EDW

▪ Program Integrity 1

▪ Pharmacy / Drug Rebate:

Foundational piece with

several vendors offering

modular solutions; not

tightly integrated with Core

MMIS

▪ Health Information

Network: Federally

mandated with immediate

deadline

▪ Provider Enrollment:

Important piece with

simplified procurement

through NASPO, impacts

providers but not members

▪ System Integrator:

Necessary across new

modules

▪ Enrollment Broker:

Implementation timeline

can shift due to managed

care carve ins; affects

potentially majority of

population

▪ Third-Party Liability:

Some initiatives

dependent on TPL

▪ Call Center: Easier to

procure, but depending on

carve ins/outs, vendor may

need to interface with

several different modules /

systems

▪ Prior Authorization: Will

require business process

updates / improvements

prior to implementation;

dependent on new Core

MMIS for new PA layouts

▪ Core MMIS / Encounter

Data Processing: Long

implementation time and

vendors are essentially the

same as monolithic MMIS

providers; market does not

currently have many

modular providers

▪ Program Integrity 2: Not

technically difficult to

implement, but need BIS –

EDW and Program Integrity

1 in place prior to

implementation

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

OPERATIONS

The Family Support Division (FSD) and MHD are responsible for critical participant- and

provider-focused functions for the Medicaid program such as eligibility determination, participant

enrollment, provider enrollment, prior authorizations/medical management, claims processing,

and general participant and provider queries and escalations. Cross-cutting support functions,

such as contact centers and data and analytics, support these customer-focused tasks. The

functions are executed through a mix of staff and vendor contracts.

In this section, the performance and operational efficiency of three functions identified by state

staff as having relevant opportunities for improvement are discussed: managed care enrollment,

claims processing, and contact centers.

200

Compared to other states and viewed from the customer-focused functional level (the integrated

process of participant enrollment from eligibility determination to MCO enrollment, for example),

actual work processes often appear fragmented, process steps seem poorly integrated and

best-practice management principles are variably applied. Individual staff participants tend to

have deep knowledge about their own responsibilities but much less insight into the overall

processes and responsibilities therein. Perceived inefficiencies in handoffs between different

parts of the organization (such as manual rework) are often accepted as “inevitable” or

“unavoidable.” Currently, outsourced roles do not appear to be optimally integrated or managed

to ensure high performance of functions.

Potential initiatives identified, if employed, may help improve suboptimal service provided to

participants and providers, create efficiencies in deployment of scarce staff, reduce frustrations

in the workforce, and realize savings while improving vendor performance. Initiatives could

include process optimization, with redesign starting from the perspective of the client(s);

automation improvements and improved contract management. Improvements made to address

the gap with best practices across the different functional areas could lead to up to 15% to 20%

improvements in productivity. This may create staff capacity that could be redeployed for other

purposes, to improve program effectiveness and workforce satisfaction.

201

Current situation

Participant managed care enrollment

Participant enrollment processes within MHD aim to enroll children and pregnant women found

eligible by FSD into the managed care program.

202

Once a participant is found to be MCO-

eligible, the Information Technology Services Division (ITSD) passes that information to the

Enrollment Broker (EB). The information the EB sends is determined based on whether the

applicant is a “state care and custody” individual, a pregnant woman, and/or “all other”

individuals. “State care and custody” applicants are auto-assigned to a Medicaid plan. They

receive letters with this information, additional information about switching plans if they wish,

and enrollment guides. Pregnant women and all other individuals receive a welcome letter with

an enrollment form and an enrollment guide. Pregnant women also receive a health risk

assessment. If they do not decide within seven days, they are auto-assigned to an MCO. All

other MCO-eligible participants have 15 days to choose an MCO before they are auto-assigned.

200

Interviews with Medicaid program staff members.

201

Analysis of DSS data versus industry benchmarks.

202

Interviews with Medicaid program staff members; for participants not in managed care no additional enrollment is

required. Once found eligible, notices are triggered, and these participants are covered as long as they keep up to

date on any premium payments, spend downs, etc.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

While the majority of process steps are automated, errors with inbound data feeds and

interactions with other state agencies drive manual interventions to ensure participant

enrollment is accurate and timely. Communications with several other state agencies take place

and are coordinated by the MHD team over email and phone correspondence.

The participant enrollment function serves as a critical interface with FSD. The process consists

of four steps (see Exhibit 48). Eligibility determination is performed within FSD, and managed

care enrollment is performed as a downstream process within MHD through an enrollment

broker. The function relies heavily on upstream FSD systems for data and information feeds.

MHD receives enrollment information from five systems, with the majority of volume driven by

FSD systems. MEDES, the primary system for managed care eligibility, supplies participant

information such as age, income, pregnancy status, etc. Once automated processes have run,

any exceptions and errors are handled within MHD participant enrollment staff through manual

intervention and communication with other state agencies such as FSD.

EXHIBIT 48: PARTICIPANT ENROLLMENT WORKFLOW

The function also engages other state services: Information feeds from the Department of

Corrections, DSS, and the Children’s Division are received monthly for manual enrollment

updates. Information supplied contains data on incarceration dates, status in foster home

programs (flagged as runaway), etc.

Claims operations

Claims operations include processes from intake of claim files, prior authorizations, claim

adjudication and finalization.

203

Claims operations involve multiple stakeholders across claims

203

Interviews with Medicaid program staff members.

FSD and other

state agencies

Data provided

to MO Health

Net for MCO

enrollment

Errors are manually

adjusted to ensure

accuracy

Enrollment updated based

on information from

department of corrections,

child services, etc.

Errors require

manual processing

Enrollments

process begins

automatically

Error reconciliation

Upstream eligibility

determination

Periodic enrollment

status updates

Errors in enrollment data are

manually reconciled

FSD and other

agencies determine

eligibility

Participant status updated

based on information from

state agencies

ITSD and MMIS process

upstream enrollment data

MCO enrollment

Process

workflow

Process

description

▪ FSD, child services,

youth services and

other agencies

determine eligibility

for Medicaid

enrollment

▪ Participant

information is

passed to MO

Health Net for MCO

enrollment

▪ Eligibility data received

from state agencies is

ingested and checked

for accuracy

▪ Accurate records begin

the enrollment process

through data intake by

ITSD and MMIS

systems

▪ However, data errors

drive need for manual

reconciliation

▪ Manual adjustments are

made to erroneous data

inputs to ensure accurate

enrollments are processed

▪ Information exchange is

often over email and phone

with various agencies

▪ Participant status

changes are often

received from

department of

corrections, child

services and other

agencies

▪ Updates are reflected

through manual updates

to enrollment systems

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

operations, IT, and vendor resources. These operations are both largely automated and

outsourced, in line with industry best practices (see Exhibit 49).

EXHIBIT 49: CLAIMS INTAKE, ADJUDICATION, AND PAYMENT PROCESSING WORKFLOW

The claims operations function cuts across core claims operations, IT systems (MMIS), and

other adjacent processes such as prior authorization. The claim intake is largely digital with

limited need for OCR or manual intake in MMIS. The sources of manual intake include out-of-

state provider claims, some drug claims, and some DME claims. Various labor-intensive

process steps such as manual adjudication and prior authorization have been outsourced and

MHD staff focus on issue resolution with vendors and providing expertise to allow vendors to

adjudicate claims appropriately. Although the MMIS system is outdated (see previous section),

its core FFS claim processing functionality has over the years become well-aligned with

current work processes and needs. The system yields high auto adjudication rates, reducing

manual work and improving accuracy for participant and provider stakeholders.

Claims operation processes are largely outsourced: vendors are utilized across almost all

process components, including exception adjudication and medical record review.

Contact Centers

DSS manages two large clusters of contact centers. Some contact centers are outsourced; the

largest, internal FSD contact centers are composed of ~350 FTEs. FSD contact centers

handle calls related to food stamps, health care, child care, and child support for families.

Internal contact centers employ full-time state staff members, many of whom served as case

workers before the state moved away from the practice of assigning recipients case workers.

The internal center within FSD handles queries related to food stamp benefits and processing

in addition to a dedicated tier of agents who conduct interviews for food stamp and child care

Adjudication

Pre-Cert /

authorization

Payment and

communications

EDI

gateway

Claims

Accums

Claim

finalized

Manual

adjudication

Auto

Adjudication

Pre-

auth

Provider

payment

EOB

Phone

EDI

EMR

and

benefit

data

Pre-auth

Adjustments and

reprocessing

Claims

Intake

checks

Claim payment calculated and

data synced

Provider verifies

eligibility

Patient and provider

informed of adjudication

decision

837 files

received

Claim

routed to

adjudication

platform

Intake and routing

Process

workflow

Process

description

▪ Pre-authorizations

are received

electronically or

manually

▪ Processed both

internally and by

vendors

▪ Process is largely

automated but

some manual

decision making is

required

▪ Intake includes both

receipt and preparation

of claims for

adjudication

▪ Process is largely

automated with specific

claim types such as out

of state claims triggering

manual intervention

▪ Vendors engaged for

mailroom and OCR

activities

▪ Clean claims are passed to

adjudication system for

finalization

▪ Edits built into system

appropriately adjudicate

claims as to be paid or

denied

▪ Small portion of claims are

pended requiring manual

adjudication

▪ After claim adjudication,

finalized claims are

passed to payment

processes

▪ Appropriate payments

and notifications are

triggered to both

participants and

providers

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

eligibility, while outsourced FSD contact centers handle MAGI-related requests. The FSD

contact centers handled 3.2 million calls in 2018.

204

The constituent health services (CHS)

contact centers are largely outsourced. Both internal and external contact centers handle

queries related to the state-run Medicaid program for participants and providers.

205

See

Exhibit 50 for a breakdown of staff members across DSS contact centers.

EXHIBIT 50: MISSOURI DSS CONTACT CENTERS BREAKDOWN

206

Incoming calls to FSD contact centers are routed to one of five tiers based on call reason.

~50% of FSD incoming calls are classified as Tier 1 (family support helpline): basic information

requests, queries about outbound communication (annual review letters) and case status

requests. ~30% of FSD incoming calls are classified as Tier 3 (food stamp interviews), which

handle mandatory food stamp interviews. The remaining ~20% of calls are routed to Tier 2, 4,

or 5 and are requests for live case-processing over the phone (e.g., for issuance of food-stamp

benefits) or MAGI hearing requests (see Exhibit 51). Wait times average ~10 minutes but can

reach over an hour for each tier.

207

204

Analysis of state data, 2018; interviews with Medicaid program staff members.

205

Interviews with Medicaid program staff members.

206

Analysis of state data, 2018.

207

Analysis of state data, 2018.

Constituent

health

services

(~100 FTEs)

Family

Support

Division

(~500 FTEs)

Technical helpdesk

Participant

services

Provider

relations

Service

authorization

In-house

Outsourced

107

Technical Helpdesk

MAGI Helpline

Tier support

desk

Eligibility

and enrollment

~350

FTE

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 51: OVERVIEW OF FSD CALL TYPES

208

Internal FSD contact center operations are spread across 10 contact centers located

throughout the state. These average a size of ~30-50 FTEs in each location; training, hiring

and other support functions are centralized and require staff member travel. Exhibit 52 shows

the number of staff members as well as the average handle time (AHT) per location.

208

Interviews with Medicaid program staff members; analysis of state data, 2018.

Call type

Tier 1

Tier 2

Tier 3

Tier 4

Tier 5

Description

▪ Basic questions about

applications, eligibility,

minor system changes,

and SS updates

▪ Case processing for food

stamps and MHABD

▪ Interviews for programs

including food stamps and

child support

▪ Hearing request for MAGI

▪ Case processing for

temporary assistance,

childcare, and food stamps

of specific participant

groups (e.g., disabled)

FTEs in

each tier

160

AHT Per

Tier (mins)

10.0

15.3

29.2

23.3

16.0

% of

calls

53%

11%

30%

Average wait

times (mins)

13.0

10.9

8.0

9.0

Max wait

times (mins)

67.0

62.2

60.9

61.5

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

EXHIBIT 52: FTE BREAKDOWN BY LOCATION – FSD INTERNAL CONTACT CENTERS

209

The AHT varies across locations: there is a 40% variation between the location with the

highest and lowest handle times. Workforce management practices are limited across the

centers. Each location operates for ~12 hours, leading to support ratios of one supervisor to

eight frontline agents and one manager to 45 frontline agents.

Outsourced operations deliver several key services to the Missouri Medicaid program: MHD

participant and provider communications, which handles incoming calls related to participant

and provider queries related to the Medicaid program (e.g., premiums, cost estimates); service

authorizations; and the technical helpdesk for the MHD program.

210

In addition, within FSD, the outsourced contact center specializes in MAGI enrollment and the

eligibility helpdesk, including calls related to program eligibility, enrollment of new participants

into the Medicaid program, inquires related to MAGI and case updates for the existing MAGI

programs.

Potential opportunities for improvement

This section identifies potential opportunities to improve the efficiency and outcomes of

Missouri’s operations functions.

209

Analysis of state data, 2018. ATT: average talk time. ACW: after call work (avg. time for the agent to wrap up

call-related tasks).

210

Analysis of state data, 2018.

Number of FTEs, # AHT in Tier 1, secondsLocation

Jennings

Hillsboro

Chouteau

St. Charles

Cape

Midtown

Page

St. Mary’s

Sikeston

Boone

406

478

410

490

545

538

649

593

478

392

-40%

ACWATT

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

Participant managed care enrollment and claims operations

• Limited KPI tracking and dashboarding: There is limited tracking of Key Performance

Indicators (KPIs) across key functions. Also, compared to best-in-class payors, contract

management could be improved in areas such as claim handle time, adjustment rate, and

timeliness of payment at a granular level.

211

• Staff members, particularly those involved in the participant enrollment process,

perform a significant number of repetitive manual tasks: Some of these tasks include

incarcerated participant disenrollment, runaway children closeouts in ITSD (while they

remain eligible), immediate enrollment of women who have just given birth, and error

reconciliation.

212

Significant manual intervention is required to ensure enrollment

information is correct and up to date.

For claims operations, the need for manual and resource intensive interventions is limited

primarily to medical record review and claims adjudication. The adjudication of complex

claims typically requires manual intervention to ensure the right amount is being paid for

relevant/appropriate services provided to participants. In some cases, this implies

requesting medical records, that may be reviewed by clinicians, to ensure that services

provided conform to existing policies. Such cases can require significant time as multiple

records are received piecemeal from providers and only a subset may be matched to the

correct claim. Even if the medical record was available, the content/document

management process is not always able to link the record with the corresponding claim.

• Upstream processes and outputs drive errors and limit scope for process

automation: Challenges with the quality of data received from upstream systems drives

manual intervention (e.g., duplicate DCNs, deceased eligibility, >9 months pregnancies).

The information received can be erroneous and needs to be manually resolved by MHD

staff.

213

Resolutions involve review of data received (e.g., re-coding a pregnant participant

as female instead of male) and communication with agencies providing data to gain

clarifications (e.g., managing multiple DCNs). Data updates for participants are often

received through email and must be manually inputted or modified within enrollment

systems including ITSD and MMIS. Several of these issues are currently being addressed

by FSD.

• Participant correspondence processes are not integrated across FSD and MHD:

Parallel communication with participants drives repeat, out-of-order, and therefore

potentially confusing communication to participants. For example, participants receive

eligibility notices and premium notifications separately from both FSD and MHD. FSD

notifications do not inform individuals that they might have to pay a premium and instead

only inform them that they are eligible for Medicaid. Subsequently, secondary premium

notifications from MHD may go unnoticed, resulting in poor response rates and potential

disenrollment.

• Staff members perform significant tasks within participant enrollment: Compared to

those at best-in-class payors, staff members perform a significant number of tasks across

processes such as managed care enrollment, disenrollment, and error correction, which in

other state Medicaid programs are often managed by vendors. Currently, therefore,

211

Interviews with Medicaid program staff members.

212

Interviews with Medicaid program staff members.

213

Interviews with Medicaid program staff members.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

100

internal staff spend much of their time performing tasks that could potentially be

outsourced instead of, for example, focusing on quality assurance of the processes.

Contact centers

• In the FSD internal contact center, current non-phone self-service options are

limited, leading to a high number of live contacts and high wait times: Limited

alternative self-service options (e.g., chat or SMS bots) are available for answering basic

questions, which creates high call volume in Tier 1. In addition, no status notification

systems are in place to inform the participants about the status of food stamp applications

or document requests, leading to requests for over-the-phone case processing and status

updates. The combination of these factors leads to high incoming call volume in Tier 1,

leading to high wait times and sub-optimal customer and staff member experience (e.g.,

waiting 60 minutes for a 1-minute answer to a question). Addressing the above

opportunities could drive a 15% to 25% reduction in incoming call volume while

significantly enhancing participant experience during the food stamp interview process.

214

In addition, the self-service options could reduce call volume in Tiers 2, 3, 4, and 5 through

notification and tracking of claims processing, scheduling hearing requests via email or

chatbots, and alerting an individual when their application is ready for interview via SMS or

email. Self-service options require thoughtful design, as suboptimal website and chat bot

design may reduce participant experience rather than improve it.

• FSD outbound communications (e.g., letters, review requests) and applications

highlight live customer support options that result in high wait times in FSD contact

centers: Many communications focus on providing customer support via phone and do not

guide participants to alternate resources such as the website or clarify the frequently asked

questions in the communication. For example, the Missouri food stamp application form

bolds the customer support number and asks a participant to call as soon as possible,

even though the participant needs to wait 24 hours after submission for the form to be

uploaded in the system.

215

In comparison, Florida food stamp application forms more

clearly explain the processing window and guide participants to resources on the web (see

Exhibit 53).

216

Missouri could adopt a combination of these approaches to improve

participant access to timely support.

214

Analysis of state data, 2018.

215

Missouri DSS, “886-0460 (9-16) Application for Food Stamp Benefits,” see:

dss.mo.gov/fsd/formsmanual/pdf/fs1.pdf.

216

Florida state government, “Access Florida Application,” 2016.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

101

EXHIBIT 53: FOOD STAMP APPLICATION EXAMPLE

217

• Customer experience is impacted due to both variability in average handle time for

calls and high wait times: In each tier, bottom-quartile agents require 1.7 times longer to

handle an average call in comparison to the top-quartile agents. For example, in Tier 3 top-

quartile agents handle an interview in ~18 minutes, while bottom-quartile agents take ~30

minutes (see Exhibit 54).

218

Top-performing agents may handle a call in ~30-40% lower

handle time compared to lower-performing agents.

219

This could indicate gaps in training

and coaching processes, which prevent delivery of a consistent experience. In addition, in

the internal contact centers, workforce management practices are not deployed to match

incoming call volume to expected staffing of agents in each tier; this likely leads to high

wait times in certain tiers during peak times (e.g., >60-minute wait time in Tier 1 on

Monday mornings). Currently, Tier 1 has 67 dedicated agents; although analysis indicates

at least 100 agents may be required to meet demand. Conversely, Tier 5 has 32 agents

although only 15 may be required to meet demand.

220

217

Missouri DSS and Florida state government: see note 215 and note 216.

218

Analysis of state data, 2018.

219

Analysis of state data, 2018.

220

Analysis of state data, 2018.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

102

EXHIBIT 54: AVERAGE HANDLE TIME VARIABILITY – FSD INTERNAL CONTACT CENTER

• In the internal contact centers, multiple locations limit the efficiency of support

functions: Many locations operate at low scale which leads to performance variability,

challenges in support services, and underutilization of facility space.

221

Several locations

have significantly lower management spans in comparison to the industry benchmark ratio

of 15 frontline agents per supervisor (see Exhibit 55).

222

Multiple locations not only lead to

higher cost to serve but also to diminished staff member experience due to reduced scale.

For example, staff members receive less training because of the need to travel to a central

location, and there is inconsistent operating experience in centers.

221

Internal contact center site observations; Interviews with Medicaid program staff members.

222

Analysis of state data, 2018.

422

1,620

Average

335

402

487

4th quartile

1st quartile

3rd quartile

2nd quartile

+45%

1,109

1,479

1,784

+61%

Tier 3Tier 1

AHT per agent quartile (Seconds/call, Aug- Oct 2018)

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

103

EXHIBIT 55: AGENT, SUPERVISOR, AND MANAGER STAFFING LEVELS

• Policies governing the outsourced contact center operations lead to rework for

internal state staff members: In MAGI contact center operations, policies prohibit

outsourced contact centers from submission of MAGI case updates in case of change of

coverage. This tends to lead to a case transfer to internal case processing teams, who

rework the case from the beginning, doubling work and increasing processing time, as well

as potentially frustrating customers. Currently, 38% of calls to the MAGI contact center

must be transferred for internal processing.

223

Similarly, a policy to transfer calls from

MHD reception for internal escalation as opposed to directly to the vendor leads to

disruptions in workflow for internal agents and could potentially be simplified to improve

customer and staff member experience.

• Dual operating environment of in-house and outsourced operating model: The state

currently manages both in-house and outsourced operations in its current contact center

operations. Currently, Missouri’s state in-house operations and outsourcers have similar

operating costs. The state has approximately the same set of resources dedicated to

managing in-house operations and for contracts for outsourced operations. This may lead

to a dual focus of administrative resources and limited opportunity to focus and hone

expertise in either of the skillsets.

223

Analysis of state data, 2018.

5.6

Lowest

span

(Boone)

Highest

Span

(Cape)

Average

(across all

locations)

8.8

8.3

Current Avg.

Benchmark

~15

35.0

60.0

44.4

Avg.

Benchmark

Current

~90

Number of agents per supervisor, #

Number of agents per manager, #

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

104

Potential initiatives

Based on observations and engagement of Missouri’s leaders and functional process owners,

potential initiatives spanning organizational process optimization, automation and digitization,

and sourcing optimization have been identified.

Implementation of outlined initiatives could drive opportunity for MHD across the following axes:

• Enhanced customer experience: Automation, digitization and process improvement

could drive improved experience for participants and reduce pain points.

• Improved staff utilization and satisfaction: Elimination of repetitive, manual processes

and reduction of error corrections could help staff contribute to other high value processes.

FTE capacity created could potentially be used to address limited staffing in other core

operations functions.

• Optimized vendor spending: Assessment of current vendor contracts and management

of existing work types could help reduce administrative expense and vendor performance.

Initiatives that deal with technological improvements and capabilities should likely be

considered in conjunction with the initiatives discussed in the section on MMIS.

Potential organizational process optimization initiatives

1. Develop process guides for staff member efficiency improvement and error

reduction: Currently, a limited number of job guides exist to guide staff member on best

practices for repetitive tasks. The creation of such guides could drive improved efficiency,

reduce errors, and greatly shorten time to ramp up new staff members on core manual

processes. Specifically, the enrollment process has many potential tasks that could benefit

from the creation of a guide, such as incarcerated disenrollment and error reconciliation.

Guides that provide step-by-step instructions on how to complete these tasks could be

readily created and may be greatly beneficial.

2. Develop job aides for high-volume tasks: High-volume tasks that are currently performed

through experience and on-the-job learning have the potential to be standardized and

expedited through the creation of job aids. (Job aids are basic decision trees, checklists,

planning tools that support work and activity by guiding or directing tasks at hand.) For

example, within claims operations, the team could create job aids who provide algorithmic

guides specifically for top edits and manual adjudications.

3. Implement workforce management: In contact centers, for example, shifting staff

members across tiers and optimizing staffing in each tier could achieve reduced wait time on

the phone (i.e., average speed of answer of less than 60 seconds against current average of

over 10 minutes). To achieve this reduction in wait time, potential solutions include moving

agents from Tiers 3, 4, and 5 to Tier 1 and ensuring the right shrinkage factors (e.g.,

absenteeism) are factored into the staffing model. Workforce management principles could

also improve efficiencies within claims processing by further aligning staff members to

specializations by skill level and claim type. This could improve processing times and staff

productivity.

4. Adopt performance management practices: In contact centers, the state could coach

toward behaviors that drive high talk-time and quality and reduce variability in average

handle time across each tier to achieve a ~10% to 12% reduction in average handle time in

each tier and improve customer experience. The state could adopt best-in-class

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

105

performance management practices, including defining clear agent goals and KPIs and

increasing structured coaching and uniform meeting cadence. To help define clear goals

and a holistic set of KPIs, scorecards could be updated with realistic goals against important

KPIs. (See Exhibit 56 for typical measurements that are leveraged in contact center

environments and shared across contact centers to hold individuals/teams accountable for

their role in creating a positive participant experience.) With clearer scorecards and KPIs,

coaching could become more structured and efficient, driving better customer experience

and lower wait time through reduction in average handle time.

KPIs and dashboarding could be equally essential in participant enrollment and claims

processing (see Exhibit 57 for typical KPIs). In the claims tracking process, for example,

organizational leadership could increase its effectiveness if it could have access to critical

KPIs such as auto-adjudication rate, adjustment rate, percentage of claims paid as billed,

rate of denial by denial reason code and denials overturn rate.

EXHIBIT 56: EXAMPLE BEST PRACTICE CONTACT CENTER KPIS

Key KPIs

Service/

Quality

Efficiency

Volume

Dimension

Category Metric Typical definition

▪ Resolution rate ▪ Repeat transaction rate

▪ FCR

▪ % SR requests/calls with a repeat call/SR within 24 hours

▪ % of calls resolved during the first call

▪ CSAT ▪ End-user satisfaction

▪ End-user dissatisfaction

▪ TNPS

▪ On a scale of 1-5, % of customers that have rated 4 or 5 (satisfied/

very satisfied)

▪ % of customers choosing “dis-satisfied (2)” or “very dissatisfied (1”)

▪ Promoters – Detractors

▪ Escalation & transfer

▪ Volume

▪ Escalation and transfer

rates

▪ Transactions offered

▪ % of answered calls escalated /transferred to Tier 2/ other

departments/desks

▪ Availability ▪ Attendance

▪ Schedule adherence

▪ CSR showing up for work on their scheduled day

▪ Time CSR is available / time they are scheduled to work

▪ Utilization ▪ Utilization

▪ Occupancy ratio

▪ (Talk+Hold+ACW)/ Total paid time

▪ Talk and wrap time divided by logged time

▪ Average handle time ▪ Average handle time ▪ (Talk+Hold+ACW)/ Total number of answered calls. Average

processing time for deferred transactions

▪ Transactions quality ▪ Quality score

▪ Critical accuracy score

▪ Call monitoring score(%) (average score on call

monitoring/maximum score)

▪ % of monitored calls without any fatal errors

▪ On time (for deferred

transactions)

▪ Service level

▪ Average time late

▪ % of transactions processed within targeted cycle time

▪ Average time late of transactions which are outside of targeted cycle

time.

▪ Speed of answer ▪ Service level

▪ Avg. speed of answer

▪ % of intervals that achieve the target service level

▪ Total wait time/ Total number of answered calls

▪ Abandoned rate ▪ Abandoned rate after the

IVR

▪ Abandoned rate during

the IVR

▪ Calls dropped after it reaches an CSR/Calls offered

▪ Calls dropped during IVR, before it reaches an CSR (excluding self-

serve calls)/Calls offered

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

106

EXHIBIT 57: EXAMPLE PARTICIPANT ENROLLMENT AND CLAIMS PROCESSING KPIS

5. Within the participant enrollment flow, integrate mailer and correspondence process

with FSD: The state could implement process change to integrate correspondence of

premium notices with FSD eligibility notices to drive improved response rate. This could

require a simple process change to implement the inclusion of the first premium notice in the

same envelope as the eligibility notice. This may significantly reduce non-responses to

premium notices.

6. Improve medical record matching to reduce incorrect denials in participant

enrollment: Matching medical records to appropriate claims to minimize inaccurate denials

could drive significant improvement in enrollment accuracy and reduce downstream rework.

7. Improve accumulator accuracy to help manage spend down errors: Spend down

inaccuracy drives significant billing errors and inbound inquiries. The state could consider

setting up a team to minimize spend down on out of sync scenarios, which could help

minimize errors.

8. Assess prior authorization (PA) list for high pass rate codes and optimize through

quarterly refreshes: State staff could conduct analysis to identify drivers of manual PAs

and ensure quarterly list refreshes. This could minimize manual PAs for high pass rate

codes.

9. Redesign root-cause drivers (e.g., participant communication & notification) to

reduce call volume to contact centers: The state could institute ongoing processes to

Participant

enrollment

Claims

Payor value

chain function

Metric subcategory

Average enrollment processing days

Enrollment accuracy

Specific metric

ID cards ▪ Percent of participants that received cards before effective date

Summary metrics ▪ Enrollment cost per total transaction

▪ Total enrollment transactions per FTE per year

▪ Total automated electronic transactions as a % of total member

transactions

Billing ▪ Total number of bills sent per participant

Claims type rates ▪ Total suspension rate

▪ Total claims auto-adjudication rate

▪ Total adjustment rate

▪ Denied claims rate

Speed of processing ▪ Average payment period in days

▪ Average inventory in days

Percent of claims processed within

the following days of receipts

▪ 0-14 days

▪ 15-30 days

▪ 31-60 days

▪ > 60 days

Timing of claims ▪ Claims turnaround time (TAT)

Summary metrics ▪ Suspended (manual) claims processed per FTE per year

▪ Cost per suspended (manual claim)

▪ Claims processed per claims FTE per year

▪ Cost per claims processed

Quality ▪ Dollar accuracy percent

▪ Frequency accuracy percent

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

107

address root-cause drivers currently leading to increased call volumes. There are several

near-term initiatives that could lead to the reduction of call volume. For example, the state

could consider redesigning forms and letters to guide to digital channels and highlight

additional communication resources available. This communication could provide clear and

updated guidelines on when to call the contact center after application submission. In the

case of interviews, the state could consider asking the applicants to wait 24 hours to allow

for the appropriate processing time before calling.

10. Revise policy guidance on MAGI helpdesk to avoid rework: The state could review

internal policies that currently prohibit outsourced MAGI helpdesk agents from completing

case updates in situations when a change of coverage occurs (currently, these changes

must be completed by an internal agent). Also, the state could consider updating policies to

enable MHD staff members to transfer to Wipro, when applicable, to reduce downstream

rework in internal case processing team.

Potential automation and digitization process optimization initiatives

11. Implement macros and automation to replace repetitive manual tasks: The state could

identify repetitive manual tasks and build simple macros/automation to reduce manual

intervention. Batch enrollment corrects or incarcerated participants disenrollment could be

executed automatically, for example. Implementation would require both macro

development and inbound data manipulation. Creating macros or process automation

routines that utilize database queries and pre-set algorithms to perform defined tasks such

as participant information updates or error reconciliation could significantly reduce the

manual intervention required. Engaging vendors to identify and build use cases for

automation could drive efficiency gains: in some quick-win cases, technologies like optical

character recognition (OCR) could be implemented within 6 to 12 months, while more

complex implementations (e.g., machine learning to improve auto adjudication rates) could

take 12 to 24 months.

12. Improving upstream systems to help reduce manual rework: Erroneous information

feeds drive significant rework within MHD processes. Engaging FSD leadership to drive

changes in these upstream systems (especially MEDES) could significantly reduce rework

within participant enrollment function. Improved data formats (e.g., pipe-delimited flat files

rather than email-based information) could provide basis for rapid system updates,

eliminating manual processes.

13. The state could engage inbound data stream owners to align on data feed formats:

Currently, data from various other state agencies is primarily received via email. Convening

leaders to align on unified and simple data exchange formats (some best-in-class payors

use pipe-delimited files) could allow for easy and automated intake into ITSD systems. This

could reduce manual workarounds and potential for error.

14. Invest in improvement of auto adjudication rates: The state could conduct detailed

analysis to assess current drivers of manual adjudication – such as edits, medical policies,

system issues – and inbound data issues, in addition to implementing improvements in the

claims systems to improve auto-adjudication rates. For example, the state could consider

identifying top edits that trigger manual adjudication and determining modifications to edits

that could drive claims to be auto-adjudicated. For example, if an edit requires an

assessment of a particular attachment or medical record and it is found that such claims are

paid with a high pass rate, removing that requirement could eliminate need for manual

intervention.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

108

15. Implement issue and project tracking system: Currently, issue and request management

is done through email. The state could consider transitioning to a ticket-based management

system that provides real time tracking, escalation paths and pan-organizational

transparency.

16. Build digital participant engagement platform: Transitioning traditional communication

channels to a digital medium for high-impact communications such as premium notices

(e.g., e-pay functions), ID cards, or explanation of benefits (EOB) delivery could drive

improved participant engagement. The state could consider investing to get ahead of

developing participant digital preference trends and drive adoption for new enrollees over a

five-year horizon.

224

Some of these participant engagement practices could help promote

self-service and reduce call volumes for the contact center and save on costs incurred due

to existing communication using traditional channels, such as printed ID cards and EOBs.

Given the proliferation of different modes of communication, it could also be helpful to

note/flag the preferred method of communication during the enrollment process.

17. Provide self-service options for Tier 1 calls to reduce live calls and wait times: The

state could consider investing in new self-service channels for the resolution of simple

issues like status of cases or food stamp eligibility through alternative channels to reduce

call volume by ~15% to 20% in Tier 1. This could ease the load and reduce the current peak

wait time for 60 minutes significantly. The state could consider potential quick wins like

website-based self-service options and SMS-based notifications and bots to provide quick

answers to simple requirement questions. Likewise, chatbots in the website could potentially

answer queries on case status and document uploads (e.g., an AI-based chat service with

limited human intervention).

Sourcing optimization initiatives

18. Evaluate engaging additional vendors: The state could consider engaging a vendor for

improved participant address management. Significant correspondence challenges stem

from dead or out-of-date addresses. Engaging vendors to conduct address reconciliation

and quality improvement is a best-in-class payor practice and can help better engage

participants. Ensuring mail is sent to correct participant addresses could also drive cost

savings by reducing rework and through postage and printing cost reduction.

In addition, the state could consider engaging vendors to maximize value added work

performed by the department’s participant enrollment team. MHD could consider the

potential to outsource enrollment correction processes and assess the current reliance on

participant enrollment team to solve routine and complex enrollment issues.

19. Define future operating model for state contact centers to balance in-house vs

outsourcing options: As part of future review of the state’s contact center operation, the

state could consider three options for future operating model for contact centers:

• Focus on operations excellence and operate contact centers internally with only

strategic outsourced vendor partners as required

• Focus on contract excellence and move to a primarily outsourced model with retention

or strategic contact center operations in-house

224

McKinsey, “Healthcare’s digital future,” 2014, see: www.mckinsey.com/industries/healthcare-systems-and-

services/our-insights/healthcares-digital-future.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

109

• Continue blended operating model with focus on operations excellence with internal

center and focus on contract excellence for outsourced operations with adequate

resources to manage in-house operations and outsourced operations

All options could facilitate providing best-in-class contact center services to participants

and providers. In the first two options, the state could choose to dedicate its resources to

focus on either on operational excellence or contract excellence; in a blended model, the

state would likely need to ensure that appropriate resources are dedicated to each area

against a shared resourcing model with dual focus. To determine the best fit option for the

long term, the state could evaluate each of these options in view of cost, quality of service,

strategic fit, administrative priorities, and operational agility to determine the best choice

going forward.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

110

Implementation Considerations

Without significant changes, Medicaid spending may comprise 26% to 30% of state general

revenues by 2023. To bring growth of Medicaid spending in line with the level of economic

growth of the state while preserving access for participants and avoid reducing eligibility or

coverage, significant savings would be necessary. In the preceding pages, eight programmatic

and functional areas were analyzed, and descriptions of the current state, potential opportunities

for improvement, and potential initiatives were provided.

Were Missouri to fully and effectively address the opportunities and potential initiatives outlined

in this report, total gross savings to the program (including federal and state share) could total

up to $0.5 billion to $1.0 billion by SFY2023 (net of potential reinvestments in the delivery

system and in the Medicaid program’s operations). These potential savings are not meant to

represent an absolute reduction in Medicaid program spending but rather a meaningful

reduction in the growth rate of the program to bring it in line with economic growth of the state.

In addition, many initiatives focus on reducing cost growth through improving participant

outcomes and experience. Adopting such a transformation agenda could make the program

more financially sustainable and reduce fiscal pressure that may arise in the event of a

recession or changes in federal financing.

Potential initiatives are wide-ranging, including operational improvements to bring the program

up to date with common practices among other state Medicaid programs, as well as

implementing best practices and more transformational changes. The following entails some of

the choices the state may consider in selecting the portfolio of initiatives that will comprise

Missouri’s approach to Medicaid transformation. Also outlined below is a summary of the key

requirements for implementation.

Approach to Portfolio Selection

Broadly, the state could balance two approaches to controlling spending. One approach

commonly adopted by both public programs and managed care would rely primarily on

controlling the unit prices paid for services and seeking to curb utilization through payor

decisions regarding clinical necessity. This approach could reduce costs and drive efficiency

across provider types, readily realizes savings, with limited associated technical complexity. Yet

this approach may lead to provider resistance and does not provide an incentive to improve

patient outcomes. Finally, mainly focusing on rates and volume would likely be only a temporary

solution: as one of the root causes of the problem – the underlying FFS payment mechanism –

would not be addressed, fragmentation and growth of volume may continue to exist, potentially

leading to the need for further rounds of budgetary tightening.

In the second approach, the state would seek to adopt innovative value-based payment and

care delivery models that reward providers for quality and efficiency of the total care delivered to

patients. This approach may support more transformational changes in care delivery, with

corresponding improvements in patient outcomes and experience. A key focus would be

reducing costs through improved outcomes for participants: strengthening primary care,

integrating behavioral and physical care, emphasizing independent living at home (with

community support where needed), and addressing social determinants of health. This

approach would pay for the outcomes that matter to participants rather than volume and would

stimulate the transparency of provider performance. The approach is likely to require greater

commitment of resources and longer to generate impact, given the need for providers to adopt

not only the new payment models but also to adopt new capabilities and implement changes in

clinical practices.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

111

To align the growth of Medicaid expenditures with the state’s economic growth could require a

combination of these approaches, balancing and prioritizing shorter- with longer-term needs and

strategic goals. Regardless of the balance chosen, there is a range of “no regret,” operational

initiatives that the state may consider bringing policies and operations up to speed with common

practices, including state-of-the-art fraud, waste, and abuse (FWA) as well as third-party liability

(TPL) methodologies, targeted use of utilization management, as well as improvements in

contact centers and other internal administrative processes. Adoption of common and leading

practices in these areas will address outlier practice patterns and inefficiencies, generate near-

term savings, improve customer experience, increase workforce satisfaction and reduce

pressure on the rest of the system while longer-term, more transformational changes are being

implemented.

Any substantial portfolio of initiatives would demand careful planning and execution, and

investments to support the transformation and build new capabilities. Key requirements for

effective design and implementation of Medicaid transformation include: strong and visible

executive leadership; effective stakeholder engagement; commitment to fact-based decision

making supported by robust data; upskilling of key agency staff; a well-resourced transformation

office; and modernization of the program’s technological infrastructure.

Requirements for Implementation

Whichever approach the state adopts, any substantial portfolio of initiatives will demand careful

planning and execution, and thoughtful investments in new capabilities. Many states

underestimate the resources needed and the challenges that may be encountered in the

implementation process.

The assessment of the state’s Medicaid program revealed that the Departments responsible for

the Medicaid program are aware of many of the opportunities identified and would embrace an

ambitious transformation plan. Yet both leadership and staff are also acutely aware of the

challenges the state will face in effecting changes. While there is significant institutional

knowledge that will greatly benefit the state’s efforts, few have experience with managing large-

scale transformations. In addition, many of the potential initiatives will require technical

knowledge based on experiences outside Missouri. Finally, most initiatives will require building

upon operational processes that themselves were identified as needing improvement, as well as

outdated technology and data and analytics infrastructure.

Based on our experience, key requirements for successful implementation would include strong

executive leadership, a detailed and objective fact base, and extensive stakeholder

engagement. In parallel, significant attention to upskilling key agency staff and improving

technical abilities (ranging from MMIS functionality to data and analytics to the digitization of key

operational bottlenecks) will be necessary to ensure success and sustainability of

improvements. The following briefly describes these key requirements in turn.

Strong and visible executive leadership. Successful transformation of the Medicaid

program will require active and visible leadership from the Governor’s Office, the Medicaid

Director, and other agency leaders and senior staff across DSS, DHSS, and DMH, as well as

additional support from the Office of Administration. In most states, the Medicaid Director

would be the owner of the overall transformation and regular Steering Group meetings, which

could include other agency directors/commissioners, the Medicaid CFO, and Governor’s Office

representative(s), for example. Such a steering group could lead not only through decision-

making but also through role-modeling for senior staff, creating a sense of urgency, adopting

creative solutions to problems, and communicating a compelling change story.

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

112

The planning and execution of the individual initiatives could be grouped into workstreams

which could be owned by agency leaders or senior staff (“sponsors”) responsible for the areas

impacted by the workstream. The exact configuration of this group will be highly state-specific,

but a portfolio derived from this current assessment could require sponsor roles of the

Medicaid COO, CFO, and CIO, in addition to the Managed Care Director, the Value-Based

Payment lead, and the Program Integrity lead, among others. Not allocating sufficient time to

these roles is one of the most frequent reasons for implementation failure. In other states,

sponsors will spend 20% to 30% of their time for several years per workstream. The Medicaid

Director will likely have to commit to an on-the-ground leadership role in the transformation for

the majority of his/her time.

A well-resourced transformation office. To realize the implementation of an ambitious

portfolio of initiatives, a well-resourced transformation office (TO) has proven to be essential.

The TO commonly sits outside of the normal line organization and explicitly operates with a

with a clear – and bold – mandate of the executive leadership. The function of the TO goes

well beyond the traditional Project Management Office to help the executive leadership, the

steering group, and the workstreams to achieve their goals by fulfilling several core roles:

• Drive action, help clarify goals, balance priorities and coordinate between initiatives

• Help create and execute the initiative- and workstream specific implementation plans

• Create, maintain, report on, and further develop the fact base for initiatives per workstream

• Perform advanced data and analytics functions for the workstreams

• Support the stakeholder engagement process in its different forms

• Manage resources, timelines, internal- and external meetings and realization of the targets

• Facilitate training of agency staff

To fulfill all these roles, a TO will need sufficient resources. Successful states draw on talented

agency staff as well as on subject matter experts who may bring knowledge of best practices

from other states, payors, or industries to build the TO. The TO also requires rigorous project

management and in-depth experience with all the dimensions of the change process. The TO

should be led by a full-time, sufficiently senior member of the senior leadership with

experience in change management.

Upskilling of key agency staff. The Missouri Medicaid program benefits from agency staff

who are not only committed to the performance and sustainability of the program, but who

collectively possess significant institutional knowledge. At the same time, staff and leadership

realize that the future state (and the change process needed to get there) will require

knowledge and skills that are currently absent or in short supply. Without these new

capabilities, no change efforts can hope to be sustainable.

The more ambitious the agenda, the more critical it is that Medicaid agencies develop strong,

end-to-end talent capabilities – including the ability to attract, develop, deploy, reward, and

retain top talent. Key skills and roles that would require investment (in number of individuals

and/or level of skills) are, for example:

• Project management

• Lean or other business process redesign

• Vendor contracting

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

113

• Data and analytics

• Performance management

• Outcomes transparency

• Payment innovation

• Communication

• Technical and data architects

• Data scientists / engineers

• Technical and systems operations management

• Contract management

Best practices in other states include optimally building on existing talent, investing in training

(including on-the-job), redeployment, and re-training, as well as recruitment of new leaders

and staff. As several of the initiatives address the optimization of the Department’s own

operations, staff may become available for new organizational roles.

Detailed and objective fact base. A firm footing in data is necessary to set measurable

goals, and track progress and ROI. In addition, a solid base of objective facts is the foundation

for effective decision making. The opportunities outlined in this report are based on a breadth

of research and analysis conducted over the past three months. Going forward, detailed

design and implementation of initiatives will demand an even richer fact base to

• Determine the improvement opportunities per initiative in more detail and set goals

(including financial targets, outcomes of care, customer experience);

• Analyze options for granular initiative design decisions, ranging from the clinical criteria to

be incorporated into new medical policies for utilization management; methods for adjusting

new payment models for patient risk and severity; planning internal contact center redesign;

or rebasing reimbursement rates;

• Apply risk adjustment, set target budgets, calculate shared savings/losses, quality

outcomes and bonus payments;

• Forecast and track possible impacts on provider finances from changes in reimbursement;

• Forecast and track possible impacts of relevant initiatives on rural health and the safety net

(financial, access and quality); and

• Create transparency of care costs and outcomes (value) per subpopulation, key conditions,

regions and other relevant dimensions;

Stakeholder engagement. Each of the potential initiatives discussed in this report has the

potential to affect participants, providers, and other stakeholders, placing a premium on

transparency and proactive communication. Certain initiatives – chiefly those associated with

reimbursement, value-based payment, and possible changes in the scope of managed care –

pose more significant implications for stakeholders and therefore likely demand a collaborative

process for design and implementation of changes. Without adequate stakeholder

engagement, those impacted by the planned changes are more likely to experience change as

something happening to them rather than as something that was co-shaped by them as

partners in the change process. Also, without adequate stakeholder engagement, initiatives

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

114

may fail to be adequately grounded in the reality of care delivery and payment, and the

experience of participants.

Effective stakeholder engagement will take different forms, ranging from a statewide working

group consisting of stakeholder leaders, committees focused on specific initiatives or cross-

cutting topics (“regulatory issues,” or “APM quality measures”), a clear communication plan,

interactive web-based discussion forums, informative webinars, regional stakeholder

conferences, and training opportunities.

Experience has shown that successful stakeholder engagement starts with a shared narrative

about the need for change, and a strong fact base underpinning key decisions. Clarity of goals

is essential. Subsequently, initiatives should be fleshed out and design decisions and the

implementation plan should be discussed with stakeholders. Stakeholder engagement should

continue during the implementation itself as well as during the first year(s) of rollout, and

progress toward goals should be jointly monitored. As the implementation and go-live of

initiatives always run into unforeseen issues, obstacles, and opportunities, having the ability to

jointly address these is a great benefit.

Technology

The existing MMIS poses one of the key challenges for any substantial portfolio of initiatives the

state may want to implement. Many of the initiatives will require functionalities that the current

MMIS does not offer. If the planning and implementation does not take these limitations

sufficiently into account, these initiatives may fail to achieve their intended goals. That said, the

current state of the MMIS does not have to hamper achieving ambitious goals. Functional

limitations can be addressed in three ways:

1. Limited configuration or code changes. For many of the utilization management or pharmacy

initiatives, for example, or the further modernization of hospital outpatient FFS

reimbursement, minor changes to the existing MMIS will suffice. These can be incorporated

in the initiative’s implementation plan and – after ensuring that the combination of required

MMIS changes is feasible – executed.

2. Adding new, rapidly deployable functionalities with high program but low system impact. A

cross-cutting need for the transformation process as well as the program, as a whole, is

improved access to the claims data, improvement of data quality and the analytical

capabilities needed to generate the fact base mentioned above. This ranges from the

identification of improvement opportunities, monitoring of APM spending, risk adjustment,

calculation of shared savings, creation of reports for stakeholders to the tracking program

transformation goals. Vendors that support large-scale transformation processes in

Medicaid programs tend to be able to ingest states’ data (and potentially and quickly deploy

the analytics required. In addition, using state-of-the-art, off-the-shelf technology and agile

design, high-impact digitization of key operational bottlenecks can be achieved at limited

cost.

3. Planning the initiative as dependent on the MMIS replacement process. Some functionalities

are difficult to realize without significant program and system impact. The current MMIS, for

example, cannot support DRGs or drug-level pricing for 340B drugs, which are both

potential initiatives. Although analytical capabilities to create DRGs and identify these drugs

could be rapidly deployed, these functionalities would have to go much beyond analyses:

providers would have to be paid using DRGs and different drug payment schedules, and

existing program integrity algorithms (such as claims edits) would need to be changed. Such

cases could lead the state to delay or deprioritize the initiative or opt for an alternative

Missouri Department of Social Services PREDECISIONAL

Rapid Response Review – Assessment of Missouri Medicaid Program

115

initiative to achieve the goal (in this case, help reduce inpatient spending) with less system

impact. As highlighted in the section on MMIS, the state could reassess the MMIS

replacement strategy and module requirements in the light of the prioritized initiatives, thus

simultaneously solving some of the MMIS’ key limitations and facilitating the transformation’s

success.

□ □ □

As described, the Missouri Medicaid program faces significant fiscal pressure, assuming

continuation of historically observed increases in program spending, outpacing growth with

state general revenues, with the potential for further exacerbation based on both economic

and regulatory risks. There are opportunities for Missouri to dramatically improve the

effectiveness and efficiency of the program. The potential range of initiatives as previously

outlined are ambitious in scale and scope. Addressing these opportunities and initiatives will

require a thoughtful approach to portfolio selection and investment of significant resources.

However, with sufficient leadership and commitment to long-term change, Missouri has the

potential to dramatically improve the quality and efficiency of its Medicaid program and in so

doing protect the financial sustainability of the program for future generations.