Financial Alignment Initiative South Carolina Healthy Connections Prime: First Evaluation Report

September 2019

Financial Alignment Initiative

South Carolina Healthy Connections Prime:

First Evaluation Report

Final

Prepared for

Sai Ma

Nancy Chiles Shaffer

Thomas Shaffer

Centers for Medicare & Medicaid Services

Center for Medicare & Medicaid Innovation

Mail Stop WB-06-05

7500 Security Boulevard

Baltimore, MD 21244-1850

Submitted by

Edith G. Walsh

RTI International

307 Waverly Oaks Road, Suite 101

Waltham, MA 02452-8413

RTI Project Number 0214448.001.007.000.000.006

[This page intentionally left blank.]

FINANCIAL ALIGNMENT INITIATIVE

SOUTH CAROLINA HEALTHY CONNECTIONS PRIME:

FIRST EVALUATION REPORT

RTI International

Catherine Ormond, MS

Matt Toth, MSW, PhD

Jennifer Howard, PhD

Lanre Adisa, BA

Emily Vreeland, MS

Ira Dave, MSPH

Yiyan Liu, PhD

Paul Moore, MA

Emily Gillen, PhD

Giuseppina Chiri, PhD

Wayne Anderson, PhD

Melissa Morley, PhD

Angela M. Greene, MS, MBA

Project Director: Edith G. Walsh, PhD

Federal Project Officers: Sai Ma, Nancy Chiles Shaffer, and Thomas Shaffer

RTI International

CMS Contract No. HHSM-500-2014-00037i TO#7

August 2019

This project was funded by the Centers for Medicare & Medicaid Services under contract no.

HHSM-500-2014-00037i TO #7. The statements contained in this report are solely those of the

authors and do not necessarily reflect the views or policies of the Centers for Medicare & Medicaid

Services. RTI assumes responsibility for the accuracy and completeness of the information

contained in this report.

_________________________________

RTI International is a registered trademark and a trade name of Research Triangle Institute.

Acknowledgments

We would like to thank the State officials who contributed information reflected in this

Evaluation Report through interviews during site visits and quarterly telephone calls. We also

thank the Medicare-Medicaid enrollees, managed care plan staff, consumer advocates, and other

stakeholders who also answered our questions about their experience and perspectives on the

demonstrations. We gratefully acknowledge the many contributions of CMS staff, especially our

Project Officers, Sai Ma, Nancy Chiles Shaffer, and Thomas Shaffer, and our former COR,

Daniel Lehman. We also thank other staff at RTI International, who helped to gather information

and produce tables for this report. Christopher Klotschkow, Michelle Bogus, Catherine Boykin,

and Roxanne Snaauw provided excellent editing and document preparation.

Contents

Section Page

Executive Summary ...................................................................................................................ES-1

1. Overview .................................................................................................................................1

1.1 Evaluation Overview .....................................................................................................1

1.1.1 Purpose ...............................................................................................................1

1.1.2 What it Covers ...................................................................................................1

1.1.3 Data Sources ......................................................................................................2

1.2 Model Description and Demonstration Goals ................................................................4

1.3 Changes in Demonstration Design.................................................................................6

1.4 Overview of State Context .............................................................................................7

2. Integration of Medicare and Medicaid ..................................................................................11

2.1 Joint Management of Demonstration ...........................................................................11

2.2 Overview of Integrated Delivery System ....................................................................12

2.2.1 Healthy Connections Prime Plans ....................................................................12

2.2.2 Provider Arrangements and Services ...............................................................12

2.2.3 Training and Support for Plans and Providers .................................................15

2.3 Major Areas of Integration...........................................................................................16

3. Eligibility and Enrollment.....................................................................................................17

3.1 Introduction ..................................................................................................................17

3.2 Enrollment Process ......................................................................................................17

3.2.1 Eligibility .........................................................................................................17

3.2.2 Phases of Enrollment .......................................................................................18

3.2.3 Passive Enrollment Experience ........................................................................20

3.2.4 Integration of Medicare and Medicaid Enrollment Systems ...........................21

3.2.5 Contacting and Locating Enrollees ..................................................................21

3.3 Summary Data .............................................................................................................22

4. Care Coordination .................................................................................................................25

4.1 Care Coordination Model ............................................................................................25

4.1.1 Assessment .......................................................................................................26

4.1.2 HCBS Waiver Services and Coordination .......................................................29

4.1.3 Care Planning Process......................................................................................30

4.2 Information Exchange ..................................................................................................33

5. Beneficiary Experience .........................................................................................................35

5.1 Introduction ..................................................................................................................35

5.2 Impact of the Demonstration on Beneficiaries ............................................................36

5.2.1 Overall Satisfaction with Healthy Connections Prime ....................................36

5.2.2 New or Expanded Benefits ..............................................................................38

5.2.3 Medical and Specialty Services .......................................................................39

5.2.4 Care Coordination Services .............................................................................40

5.2.5 Beneficiary Access to Care and Quality of Services .......................................41

5.2.6 Person-centered Care and Patient Engagement ...............................................43

5.2.7 Personal Health Outcomes and Quality of Life ...............................................43

5.2.8 Experience of Special Populations...................................................................44

5.2.9 Beneficiary Protections ....................................................................................45

6. Stakeholder Engagement ......................................................................................................49

6.1 Overview ......................................................................................................................49

6.2 Organization and Support ............................................................................................50

6.2.1 Stakeholder Council .........................................................................................50

7. Financing and Payment .........................................................................................................53

7.1 Rate Methodology ........................................................................................................53

7.1.1 Rating Categories and Risk Adjustments ........................................................53

7.1.2 Savings Percentage ..........................................................................................54

7.1.3 Performance Incentives ....................................................................................55

7.1.4 Medical Loss Ratio ..........................................................................................56

7.2 Financial Impact...........................................................................................................56

7.2.1 Early Implementation Experience....................................................................56

7.2.2 Rate Methodology Design Implications ..........................................................57

7.2.3 Cost Experience ...............................................................................................58

8. Service Utilization ................................................................................................................61

8.1 Overview of Benefits and Services ..............................................................................62

8.2 Impact Analyses on the Demonstration Eligible Population .......................................63

8.2.1 Descriptive Statistics on the Demonstration Eligible Population ....................71

8.2.2 Impact Analyses on the Demonstration Eligible Population with SPMI .........71

8.2.3 Service Use for Enrollee and Non-Enrollee Populations in South

Carolina ............................................................................................................73

8.2.4 Service Use by Demographic Characteristics of Eligible Beneficiaries ..........74

9. Quality of Care ......................................................................................................................79

9.1 Quality Measures .........................................................................................................79

9.2 Quality Management Structures and Activities ...........................................................81

9.2.1 State and CMS Quality Management Structures and Activities ......................82

9.2.2 Healthy Connections Prime MMP Quality Management Structure and

Activities ..........................................................................................................82

9.2.3 Independent Quality Management Structures and Activities ..........................83

9.3 Results for Selected Quality Measures ........................................................................83

9.3.1 HEDIS Quality Measures Reported for Healthy Connections Prime

Plans .................................................................................................................83

10. Cost Savings Calculation ......................................................................................................87

10.1 Evaluation Design ........................................................................................................88

10.2 Medicare Expenditures: Constructing the Dependent Variable ...................................90

10.3 Results ..........................................................................................................................93

10.3.1 Regression Analysis .........................................................................................95

vii

10.4 Discussion ....................................................................................................................98

11. Conclusions ...........................................................................................................................99

11.1 Implementation-related Successes, Challenges, and Lessons Learned .......................99

11.2 Demonstration Impact on Service Utilization and Costs ...........................................100

11.3 Next Steps ..................................................................................................................101

References ....................................................................................................................................R-1

Appendices

A Comparison Group Methodology for South Carolina Demonstration Year 1 ................... A-1

B Analysis Methodology ........................................................................................................B-1

C Descriptive Tables ..............................................................................................................C-1

D Sensitivity Analysis Tables ................................................................................................ D-1

E Summary of Predemonstration and Demonstration Design Features for Medicare

and Medicaid Beneficiaries in South Carolina ................................................................... E-1

viii

List of Tables

Number Page

1 Healthy Connections Prime MMP profiles .......................................................................... 13

2 Transition of HCBS waiver functions from State to MMP ................................................. 14

3 Healthy Connections Prime enrollment phases ................................................................... 19

4 Percentage of members that Healthy Connections Prime plans were unable to reach

following three attempts, within 90 days of enrollment, by quarter .................................... 22

5 Key events affecting enrollment and percentage enrolled in Healthy Connections

Prime .................................................................................................................................... 23

6 Total percentage of enrollees whose assessment was complete within 90 days of

enrollment ............................................................................................................................ 28

7 Members with care plans within 90 days of enrollment ...................................................... 32

8 Care coordination staffing ................................................................................................... 33

9 Beneficiary overall satisfaction, 2017 ................................................................................. 37

10 Care coordination, 2017 ....................................................................................................... 41

11 Rate cell Medicaid payments per member per month, demonstration year 1 and

demonstration year 2 ............................................................................................................ 54

12 Savings percentages by demonstration year ........................................................................ 55

13 Average, minimum, and maximum HCC score by MMP during demonstration year

1 (February 1, 2015–December 31, 2016) ........................................................................... 59

14 Summary of South Carolina demonstration impact estimates for demonstration

period (February 1, 2015–December 31, 2016) ................................................................... 62

15 Adjusted means and impact estimate for eligible beneficiaries in the demonstration

and comparison groups in South Carolina through December 31, 2016 ............................. 66

16 Adjusted means and impact estimate for eligible beneficiaries in the demonstration

and comparison groups in South Carolina through the demonstration period

February 1, 2015–December 31, 2016................................................................................. 70

17 Selected HEDIS measures for Healthy Connections Prime plans, 2016 ............................. 84

18 Data sources for monthly Medicare expenditures ............................................................... 90

19 Adjustments to Medicare expenditures variable .................................................................. 91

20 Mean monthly Medicare expenditures for Healthy Connections Prime eligibles and

comparison group, predemonstration period and demonstration period 1,

unweighted ........................................................................................................................... 95

21 Mean monthly Medicare expenditures for Healthy Connections Prime eligibles and

comparison group, predemonstration period and demonstration period 1, weighted .......... 95

22 Demonstration effects on Medicare savings for eligible beneficiaries—Difference-

in-difference regression results, Healthy Connections Prime eligibles and

comparison group ................................................................................................................ 96

23 Adjusted means and overall impact estimate for eligible beneficiaries in the

demonstration and comparison groups, Healthy Connections Prime eligibles and

comparison group ................................................................................................................ 97

List of Figures

Number Page

1 Demonstration effects on service utilization for eligible beneficiaries in South

Carolina—Difference-in-differences regression results for the demonstration

period, February 1, 2015–December 31, 2016 .................................................................... 64

2 Demonstration effects on long-stay nursing facility use for eligible beneficiaries in

South Carolina—Difference-in-differences regression results for the demonstration

period, February 1, 2015–December 31, 2016 .................................................................... 64

3 Demonstration effects on RTI quality of care measures for eligible beneficiaries in

South Carolina—Difference-in-differences regression results for the demonstration

period, February 1, 2015–December 31, 2016 .................................................................... 68

4 Demonstration effects on service utilization for eligible beneficiaries with SPMI in

South Carolina—Difference-in-differences regression results for the demonstration

period, February 1, 2015–December 31, 2016 .................................................................... 72

5 Demonstration effects on quality of care and care coordination for eligible

beneficiaries with SPMI in South Carolina—Difference-in-differences regression

results for the demonstration period, February 1, 2015–December 31, 2016 ..................... 73

6 Percent with use of selected Medicare services ................................................................... 75

7 Service use among all demonstration eligible beneficiaries with use of service per

1,000 user months ................................................................................................................ 76

8 Service use among all demonstration eligible beneficiaries per 1,000 eligible

months .................................................................................................................................. 77

9 Mean monthly Medicare expenditures, predemonstration and demonstration period,

Healthy Connections Prime eligible and comparison group, February 2013–

December 2016 .................................................................................................................... 93

10 Mean monthly Medicare expenditures (weighted), predemonstration and

demonstration period, Healthy Connections Prime eligibles and comparison group,

February 2013–December 2016........................................................................................... 94

[This page intentionally left blank.]

Glossary of Terms

ACL Administration for Community Living

ACSC Ambulatory care sensitive condition

ADL Activities of daily living

CAHPS Consumer Assessment of Healthcare Providers and Systems

CICO Coordinated and Integrated Care Organizations

CLTC Community Long Term Care (SCDHHS’s division previously responsible

for HCBS services for demonstration enrollees)

CMS Centers for Medicare & Medicaid Services

CMT Contract Management Team

CTM Complaint Tracking Module

DinD Difference-in-Difference

DMH South Carolina’s Department of Mental Health

D-SNP Dual Eligible Special Needs Plan

EQRO External Quality Review Organization. For Healthy Connections Prime, it

is the Carolinas Center for Medical Excellence (CCME).

FFS Fee-for-service

HCBS Home and community-based services

HCC Hierarchical Condition Category

HEDIS Health Effectiveness Information and Data Set

HOS Health Outcomes Survey

HRA Health risk assessment

IC Implementation Council

ICP Individual care plan

IRE Medicare Independent Review Entity

LIS Low income subsidy

xii

LTC Long term care

LTSS Long-term services and supports

MA Medicare Advantage

MARx Medicare Advantage Prescription Drug System

MAXIMUS the enrollment broker for Healthy Connections Prime

MCO Managed care organization

MDS Minimum Data Set

MFFS Managed fee-for-service

MFP Money Follows the Person

MLR Medical loss ratio

MMCO Medicare-Medicaid Coordination Office

MMP Medicare-Medicaid Plan. In South Carolina, the MMPs are Absolute Total

Care (Centene), Advicare Advocate (through August 2016), Molina

Healthcare, and First Choice VIP Care Plus by Select Health of South

Carolina.

MOU Memorandum of Understanding

MT Multidisciplinary team, also known as Interdisciplinary Care Team

OSA Office for the Study of Aging at the University of South Carolina

NF Nursing facility

PACE Program of All-Inclusive Care for the Elderly

PCP Primary care physician or provider

Phoenix SCDHHS’ electronic Medicaid HCBS waiver case management system

PMPM per member per month

POC Plan of care

Prime Advocate Healthy Connections Prime’s ombudsman, who is also the LTC

Ombudsman for the State.

SCDHHS South Carolina Department of Health and Human Services

xiii

SDRS State Data Reporting System

SHIP State health insurance program

SPA State Plan Amendment

SPMI Serious and persistent mental illness

Waiver-like services Flexible benefits that MMPs may provide to Healthy Connections Prime

enrollees. They are often used prior to enrollment into an HCBS waiver.

xiv

[This page intentionally left blank.]

ES-1

Executive Summary

The Medicare-Medicaid Coordination Office and the Innovation Center at the Centers for

Medicare & Medicaid Services (CMS) have created the Medicare-Medicaid Financial Alignment

Initiative to test, in partnerships with States, integrated care models for Medicare-Medicaid

enrollees. CMS contracted with RTI International to monitor the implementation of the

demonstrations and to evaluate their impact on beneficiary experience, quality, utilization, and

cost. The evaluation will include a final aggregate evaluation report and individual State-specific

evaluation reports.

South Carolina and CMS launched the Healthy Connections Prime demonstration in

February 2015 to integrate care for Medicare-Medicaid beneficiaries throughout the State. Four

health plans were competitively selected to operate Medicare-Medicaid Plans (MMPs). MMPs

receive capitated payments from CMS and the State to finance all Medicare and Medicaid

services. MMPs also provide care coordination, a new palliative care benefit, and flexible

benefits.

Beneficiaries who are 65 years or older and living in the community are eligible for the

demonstration, which operated in 39 of the 46 counties in the State as of December 2017.

Participants in three home and community-based services (HCBS) waiver groups are also

eligible to participate.

This First Evaluation Report for the South Carolina demonstration describes

implementation of the Healthy Connections Prime demonstration and early analysis of the

demonstration’s impacts. The report includes findings from qualitative data for February 1, 2015

through December 31, 2017, with key qualitative information through mid-2018, and

quantitative results from February 1, 2015 through December 31, 2016. Data sources include key

informant interviews, beneficiary focus groups, the Consumer Assessment of Healthcare

Providers and Systems (CAHPS) survey results, Medicare claims data, the Minimum Data Set

nursing facility assessments, MMP encounter data, and other demonstration data. Future analyses

also will include Medicaid claims and encounters as those data become available.

Highlights

Eligibility and Enrollment

• In December 2017, approximately 20,726 beneficiaries were eligible for the

demonstration and 11,511 beneficiaries were enrolled. Beneficiaries eligible for

passive enrollment into the demonstration included Medicare-Medicaid beneficiaries

who received their Medicare benefits under fee-for-service. Medicare-Medicaid

beneficiaries enrolled in Medicare Advantage could also choose to disenroll from

their Medicare Advantage plan and opt into the demonstration.

• Initially in 2013, MMPs developed financial plans, built infrastructure, and hired staff

to support Healthy Connections Prime based on the State’s estimate of 53,000 eligible

beneficiaries, which included beneficiaries enrolled in Medicare Advantage plans

who were ineligible for passive enrollment. By 2016 and 2017, although MMPs had

ES-2

streamlined staffing and support for the demonstration, two of the three plans cited

financial concerns in sustaining their participation in the demonstration, due to lower

than expected enrollment and higher than expected costs.

• The demonstration experienced delays and pauses in passive enrollment due to

legislative actions at the start, and subsequently while the State’s leadership

considered including the demonstration population in a new managed long-term

services and supports (MLTSS) program. These delays resulted in a lack of

momentum to build enrollment and, consequently, a concern by MMPs and

stakeholders about the viability and future of the demonstration.

• In July 2018, CMS, the State, and all three plans agreed to extend the demonstration

through December 2020. The agreement included a plan to increase the eligible pool

by introducing passive enrollment to Dual Eligible Special Needs Plan (D-SNPs)

enrollees. It also reduced the savings percentages applied to the capitated rates that

CMS and the State pay to the plans effective 2018 and eliminated some

administrative requirements of the plans.

Care Coordination

• Evaluation of social determinants of health is an important factor in the

demonstration’s home-based enrollee assessments along with the beneficiary’s

mental, cognitive, and physical health. Care coordinators connect enrollees with

community-based groups to provide services that can enable enrollees to live safely in

their homes in this very rural State. Plans provide varying degrees of additional

flexible benefits primarily focused on nursing facility diversion.

Beneficiary Experience

• 2017 CAHPS survey results show that 54 to 66 percent of enrollees rated their

satisfaction with the three plans as a 9 or 10 (with 10 being the highest). Similarly,

almost all RTI focus group participants in South Carolina rated their plans a 9 or 10.

• 2016 and 2017 focus group participants were consistent in stating that the lack of

copays was the most important factor to improved access to care and quality of life.

Many noted that the funds not spent on copays were now used for food and other

essential items. Others noted that they were no longer turned away at providers’

offices for not having sufficient funds to receive care.

• Participants in focus groups generally knew their care coordinator by name and

described their reliance on their services. Focus group participants described ways in

which some care coordinators took extraordinary measures to accompany members to

appointments to ensure they received needed care.

• The volume of complaints and appeals has remained low throughout the

demonstration. The State’s long-term care ombudsman serves as the demonstration’s

Prime Advocate or ombudsman. In addition to working with plans to resolve disputes,

ES-3

she participates in the contract management team, and provides outreach to potential

enrollees. Plans reported the Prime Advocate is a major demonstration asset.

• Among those individuals who disenrolled from Healthy Connections Prime, surveys

and member exit interviews showed beneficiary preference to stay with their current

providers who were not in the MMPs’ networks. MMPs experienced some difficulty

attracting a wide provider network due, in part, to fear of managed care, and the

reluctance of some established health care networks to participate in the

demonstration.

Service Utilization

• As measured across all eligible beneficiaries, the demonstration resulted in a 19.0

percent reduction in the probability of inpatient admissions, a 16.9 percent reduction

in the probability of skilled nursing facility admission, an 18.5 percent reduction in

the probability of overall ambulatory care sensitive condition (ACSC) admissions,

and a 19.8 percent reduction in the probability of chronic ACSC admissions (see

Table ES-1). Conversely, the demonstration resulted in a 27.2 percent increase in the

probability of any new long-stay nursing facility use.

These reported changes, which

although appear large in percentage terms in comparison to the comparison group, are

actually quite small in absolute terms, reflecting usually less than a one percentage

point change in the probability of service use. Section 8 on service utilization and

Appendix B contain an explanation of the research design and populations analyzed.

• Results for those with serious and persistent mental illness on the probability of

inpatient or skilled nursing facility admission or ER visits were in the same direction

and to a similar degree as those for all eligible beneficiaries, as well as for the number

of evaluation and management visits.

This finding should be interpreted with caution as State-level eligibility determination delays for long-stay

nursing facility approval could have contributed to this finding. During the period covered by this report, long-

stay nursing facility residents were excluded from being eligible to enroll into the demonstration. However, due

to delays in determining nursing facility eligibility, individuals newly requiring long-stay nursing facility

services were enrolled in the demonstration, thus raising the probability of a long-stay nursing facility stay in the

demonstration group. Potentially, if beneficiaries with those long-stay nursing facility stays had been more

quickly identified, they would have been ineligible for demonstration enrollment.

ES-4

Table ES-1

Summary of South Carolina demonstration impact estimates for demonstration period

(February 1, 2015–December 31, 2016)

(p < 0.1 significance level)

Measure

All demonstration eligible

beneficiaries

Demonstration eligible

beneficiaries with SPMI

Probability of inpatient admission Decreased Decreased

Probability of ambulatory care

sensitive condition (ACSC)

admissions, overall

Decreased NS

Probability of ACSC admissions,

chronic

Decreased NS

All-cause 30-day readmissions NS Decreased

Probability of emergency room (ER)

visit

Decreased Decreased

Preventable ER visits Decreased NS

30-day follow-up after mental health

discharges

NS NS

Probability of skilled nursing facility

(SNF) admission

Decreased Decreased

Probability of any new long-stay

nursing facility (NF) use

Increased N/A

Physician evaluation and

management (E&M) visits

Decreased Decreased

N/A = not applicable; NS = not statistically significant; SPMI = serious and persistent mental illness.

SOURCE: RTI analysis of Medicare and Minimum Data Set data.

Cost Savings

• The results of preliminary Medicare cost savings analyses using a difference-in-

differences regression approach do not indicate savings or losses due to the South

Carolina demonstration over the period February 2015–December 2016. The cost

savings analyses do not include Medicaid data due to current data unavailability, but

these data will be incorporated into future calculations as they become available.

1. Overview

1.1 Evaluation Overview

1.1.1

Purpose

The Medicare-Medicaid Coordination Office (MMCO) and the Innovation Center at the

Centers for Medicare & Medicaid Services (CMS) have created the Medicare-Medicaid

Financial Alignment Initiative to test, in partnerships with States, integrated care models for

Medicare-Medicaid enrollees. The goal of these demonstrations is to develop person-centered

care delivery models integrating the full range of medical, behavioral health, and long-term

services and supports (LTSS) for Medicare-Medicaid enrollees, with the expectation that

integrated delivery models would address the current challenges associated with the lack of

coordination of Medicare and Medicaid benefits, financing, and incentives.

This report on the South Carolina capitated model demonstration under the Medicare-

Medicaid Financial Alignment Initiative, called Healthy Connections Prime, is one of several

reports that will be prepared over the next several years to evaluate the demonstration. CMS

contracted with RTI International to monitor the implementation of the demonstrations under the

Financial Alignment Initiative and to evaluate their impact on beneficiary experience, quality,

utilization, and cost. The evaluation includes a final aggregate evaluation (Walsh et al., 2013)

and individual State-specific evaluation reports.

The goals of the evaluation are to monitor demonstration implementation, evaluate the

impact of the demonstration on the beneficiary experience, monitor unintended consequences,

and monitor and evaluate the demonstration’s impact on a range of outcomes for the eligible

population as a whole and for special populations (e.g., people with mental illness and/or

substance use disorders, LTSS recipients). To achieve these goals, RTI collects qualitative and

quantitative data from South Carolina each quarter; analyzes Medicare and Medicaid enrollment,

claims, and encounter data; conducts site visits, beneficiary focus groups, and key informant

interviews; and incorporates relevant findings from any beneficiary surveys conducted by other

entities. In addition to this report, monitoring and evaluation activities will also be reported in

subsequent evaluation reports.

1.1.2

What it Covers

This report analyzes implementation of the Healthy Connections Prime demonstration

from its initiation on February 1, 2015 through December 31, 2017. For this reporting period,

quantitative data based on Medicare claims and encounter data and the nursing facility Minimum

Data Set 3.0 through 2016 are included. To capture relevant information generated at the

conclusion of the demonstration period or immediately afterward, this report also includes key

updated qualitative information through mid-2018. It describes the South Carolina Healthy

Connections Prime demonstration key design features; examines the extent to which the

demonstration was implemented as planned; identifies any modifications to the design; and

discusses the challenges, successes, and unintended consequences encountered during the period

covered by this report. It also includes data on the beneficiaries eligible and enrolled, geographic

areas covered, and status of the participating Medicare-Medicaid Plans (hereafter referred to as

Healthy Connections Prime plans or MMPs). Finally, the report includes data on care

coordination, the beneficiary experience, stakeholder engagement activities, and, to the extent

that data are available, analyses of utilization and quality, and a summary of preliminary findings

related to Medicare savings results in the first demonstration year.

1.1.3

Data Sources

A wide variety of information informed this First Evaluation Report of the Healthy

Connections Prime demonstration. Data sources used to prepare this report include the

following:

Key informant interviews. The RTI evaluation team conducted in-person site visits in

South Carolina in July 2015 and July 2016 and conducted in-depth telephone interviews in

October and November 2017. The team interviewed the following types of individuals: State

policy makers and the South Carolina Department of Health and Human Services (SCDHHS)

staff, CMS and State contract management team (CMT) members, Ombudsman Program

officials, MMP officials, MMP care coordinators, hospital and nursing facility providers,

advocates and other stakeholders.

Focus groups. The RTI evaluation team conducted eight focus groups in South Carolina

in 2016 and six in 2017. A total of 64 enrollees and 14 proxies participated in the RTI focus

groups. Participants were assigned to groups based on their LTSS and behavioral health services

use, race, ethnicity, and primary language. Focus groups were not conducted with beneficiaries

who opted out of the demonstration or who disenrolled.

Surveys. Medicare requires all Medicare Advantage plans, including Healthy

Connections Prime plans, to conduct an annual assessment of the experiences of beneficiaries

using the Medicare Advantage Prescription Drug CAHPS survey instrument. The survey was not

conducted in 2015 or 2016 due to low enrollment in the demonstration during those years. The

2017 survey for Healthy Connections Prime was conducted in the first half of 2017 and included

the core Medicare CAHPS questions and 10 supplemental questions added by the RTI evaluation

team. Survey results for a subset of 2017 survey questions are incorporated into this report.

Findings are available at the Healthy Connections Prime plan level only. The frequency count for

some survey questions may be suppressed because too few enrollees responded to the question.

Comparisons with findings from all Medicare Advantage plans are available for core CAHPS

survey questions but not for the RTI supplemental questions.

Demonstration data. The RTI evaluation team reviewed data provided quarterly by

South Carolina through the State Data Reporting System (SDRS). These data included eligibility,

enrollment, and information reported by South Carolina on its stakeholder engagement process,

accomplishments on the integration of services and systems, any changes made in policies and

procedures, and a summary of successes and challenges. This report also uses data for quality

measures reported by Healthy Connections Prime plans and submitted to CMS’ implementation

contractor, NORC.

2,3

Data reported to NORC include core quality measures that all Medicare-

Medicaid Plans are required to report, as well as State-specific measures that Healthy

Connections Prime plans are required to report. Due to some reporting inconsistencies across

plans in 2015, plans occasionally resubmit data for prior demonstration years; therefore, these

data are considered preliminary.

Demonstration policies, contracts, and other materials. This report uses several data

sources, including the Memorandum of Understanding (MOU) between the State and Centers for

Medicare & Medicaid Services (CMS) (Centers for Medicare and Medicaid Services and State of

South Carolina, 2013; hereafter, MOU, 2013); the three-way contract (U.S. Department of

Health & Human Services and the South Carolina Department of Health & Human Services,

2014; hereafter, South Carolina three-way contract, 2014); the Summary of Contract Changes

dated November 1, 2017, contract amendments dated July 1, 2018, and materials available on the

CMS website: https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-

Coordination/Medicare-Medicaid-Coordination-

Office/FinancialAlignmentInitiative/SouthCarolina.html; documents available on the Healthy

Connections Prime website: https://msp.scdhhs.gov/SCDue2/site-page/trainings-and-

presentations; data reported through the State Data Reporting System [RTI, SDRS]), and

documents on the CMS Medicare-Medicaid Coordination website (Centers for Medicare &

Medicaid Services, 2018).

Conversations with CMS and SCDHHS officials. To monitor demonstration progress,

the RTI evaluation team engages in periodic phone conversations with SCDHHS and CMS.

These might include discussions about new policy clarifications designed to improve plan

performance, quality improvement activities, and contract management team actions.

Complaints and appeals data. Complaint (also referred to as grievance) data are from

three separate sources: (1) complaints from beneficiaries reported by Healthy Connections Prime

plans to SCDHHS, and separately to CMS’ implementation contractor, NORC; (2) complaints

received by SCDHHS or 1-800-Medicare and entered into the CMS electronic Complaint

Tracking Module (CTM);

and (3) qualitative data obtained by RTI on complaints. Appeals data

are based on data reported by MMPs to SCDHHS and NORC, for Core Measure 4.2, and the

Medicare Independent Review Entity (IRE). Data on critical incidents and abuse reported to

SCDHHS and CMS’ implementation contractor by Healthy Connections Prime plans are also

included in this report.

Although a discussion of the South Carolina MMPs is included, this report presents

information primarily at the Healthy Connections Prime demonstration level. It is not intended to

Data are reported for February 2015 through December 2017.

The technical specifications for reporting requirements are in the Medicare-Medicaid Capitated Financial

Alignment Model Core Reporting Requirements document, which is available at

https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-

Medicaid-Coordination-Office/FinancialAlignmentInitiative/MMPInformationandGuidance/MMP

ReportingRequirements.html.

Data are presented for the time period February 2015 through December 2017.

assess individual plan performance, but individual plan information is provided where plan level

data are the only data available, or where plan level data provide additional context.

Service utilization data. Evaluation Report analyses used data from many sources. First,

the State provided quarterly finder files containing identifying information on all demonstration

eligible beneficiaries in the demonstration period. Second, RTI obtained administrative data on

beneficiary demographic, enrollment, and service use characteristics from CMS data systems for

both demonstration and comparison group members. Third, these administrative data were

merged with Medicare claims and encounter data, as well as the Minimum Data Set.

Although Medicaid service data on use of LTSS, behavioral health, and other Medicaid-

reimbursed services were not available for the demonstration period and therefore are not

included in this report, CMS administrative data identifying eligible beneficiaries with serious

and persistent mental illness (SPMI) was available, so that their Medicare service use could be

presented in this report. Future reports will include findings on Medicaid service use once data

are available.

1.2 Model Description and Demonstration Goals

The goals of Healthy Connections Prime are to alleviate fragmentation and improve

coordination of services for Medicare-Medicaid enrollees, enhance quality of care, and reduce

costs for both the State and the Federal government. This demonstration aims to improve the

entire beneficiary care experience by engaging beneficiaries in their care and ensuring that they

have the choice to self-direct services as appropriate. Improving the beneficiary experience can

lead to system-wide benefits such as better quality, improved transitions between care settings,

fewer health disparities, reduced costs for payers, and the elimination of cost shifting between

Medicare and Medicaid (MOU, 2013, p. 3). Healthy Connections Prime plans are responsible for

delivery and coordination of all medical, behavioral health, and LTSS for their enrollees, as well

as providing care coordination and the new palliative care benefit.

Period of demonstration. Healthy Connections Prime began on February 1, 2015 and

was due to expire on December 31, 2018. In July 2018, CMS, the State, and all three MMPs

agreed to extend the demonstration until December 31, 2020 (CMS, 2018a).

Financial model. Healthy Connections Prime plans are paid a blended, risk-adjusted

capitated rate covering all Medicare and Medicaid services. Medicare Parts A and B and

Medicaid payments reflect the application of savings percentages and quality withholds (see

Section 7.1, Rate Methodology).

Eligible population. Beneficiaries with full benefit Medicare and Medicaid and who are

age 65 or over and living in the community at the time of enrollment, are eligible for the

demonstration. This had been a population of special interest to the State which had reduced its

nursing facility admissions by increasing home and community-based services (HCBS). Except

for the State’s Program for All-Inclusive Care for the Elderly (PACE) that only covered four of

the State’s 46 counties, there had been no previous effort to integrate HCBS with primary care

and behavioral health services. The demonstration provided this opportunity.

Beneficiaries participating in the Community Choices, HIV/AIDS, or Mechanical

Ventilation Dependent waiver programs for HCBS are eligible for the demonstration. Individuals

who transition from a nursing facility into the community, and are otherwise eligible, may elect

to enroll in the demonstration. Individuals already enrolled who later enter a nursing facility may

remain in the demonstration (MOU, 2013, p. 10). Any Healthy Connections Prime enrollee

residing in a nursing facility for more than 90 days before transition from that facility also may

qualify for the State’s Money Follows the Person Rebalancing demonstration, called Home

Again. Upon transition, enrollees must enter one of the three participating waivers included in

Healthy Connections Prime. Once Healthy Connections Prime enrollees return to the

community, the MMP must offer them any enhanced HCBS covered through the Home Again

program (South Carolina three-way contract, 2014, p. 97).

Healthy Connections Prime plans. SCDHHS and CMS initially contracted with four

plans, originally known as Coordinated and Integrated Care Organizations (CICOs), Absolute

Total Care (Centene), Advicare Advocate, Molina Healthcare, and First Choice VIP Care Plus by

Select Health of South Carolina. In early 2016, after State-contracted testing showed that

beneficiaries preferred the name Medicare-Medicaid Plan (MMP), the CICO name was rarely

used. Advicare’s managed care product was purchased by WellCare Health Plans, Inc., and

withdrew from the demonstration effective August 31, 2016, leaving three MMPs in the

demonstration.

Geographic coverage. Initially, the demonstration operated in 41 of the State’s 46

counties. After Advicare’s departure, 39 of the 46 counties were covered by at least one of the

three remaining MMPs. At the end of 2017, the three MMPs each covered from 29 to 39 of the

46 counties. MMPs were unable to contract with sufficient numbers of providers in seven

counties in this very rural State; therefore, those counties are not included in the demonstration.

Beneficiaries in the counties with one MMP may opt into the demonstration, but, during the

period of this report, may not be passively enrolled. In July 2018, the State made the decision to

allow eligible beneficiaries in counties in which one plan operates to be passively enrolled into

that plan beginning fall 2018.

Care coordination. Care coordination is a central feature of the South Carolina

demonstration and plans are required to provide care coordination services to all enrollees. It

begins with a home visit and comprehensive assessment, followed by the development of a care

plan. A multidisciplinary team, dependent upon the needs and wishes of the enrollee and led by a

care coordinator, delivers care coordination to all enrollees (see Section 4.1, Care Coordination

Model).

Benefits. The demonstration provides coordination of primary care, acute care,

behavioral health services, and LTSS, and a new palliative care benefit. Beneficiaries enrolled in

the HCBS waivers receive their coordinated waiver benefits under the demonstration.

Enrollees who have a history of hospitalization or emergency department use due to pain

or symptom management and who have been diagnosed with certain conditions, such as cancer,

end-stage renal disease, or congestive heart failure, may receive the new palliative care benefit. It

is designed to improve the enrollee’s quality of life by providing pain management together with

curative treatment and, according to the State, by “adding an extra layer of support to the

enrollee.” The State has engaged various organizations to assist in developing education for

providers and plans and adopted use of the Charlson Comorbidity Index (Charlson, n.d.). Fine-

tuning the reporting specifications and tracking services has been particularly challenging in

providing this benefit according to both State officials and plans. By the end of 2017, the

demonstration was focused on the following activities: determining evidence of increase in

member advance care planning; tracking members from palliative care to hospice; and assessing

pre and post palliative care costs (SCDHHS, 2017). For more on this topic, see Section 5,

Beneficiary Experience).

Supplemental benefits and waiver-like services. Plans have supplemental benefits

available to members enrolled in their Healthy Connections Prime products that vary by plan.

Examples of such benefits include gym memberships, coverage of eyewear expense (e.g., $150

every 2 years), and a monthly allowance for over-the-counter drug products. At their discretion,

plans may also provide flexible benefits, known in South Carolina as “waiver-like” services to

enable enrollees to postpone institutionalization or prevent higher levels of care for frail

beneficiaries. These benefits typically include home modifications, respite, personal care or other

services and are provided to members who do not meet the level of care requirements to

participate in one of the three HCBS waivers included under the demonstration (SCDHHS). For

more on use of flexible benefits, please see Section 5, Beneficiary Engagement and Section 7,

Finance and Payment.

New service delivery models. A key feature of the South Carolina demonstration is the

transition of HCBS waiver roles and responsibilities from the State to the MMPs. The SCDHHS

Community Long Term Care (CLTC) division previously had responsibility for these functions

for the three waiver populations included in the demonstration. The phased transition process

was designed to systematically introduce these functions to the MMPs while preserving the

State’s authority until each MMP demonstrated readiness. For more on this activity, please see

Section 2.2.1, Provider Arrangements and Services.

Stakeholder engagement. The Healthy Connections Prime State team has engaged

stakeholders with meetings and presentations throughout the early design and implementation

phases of the demonstration. The Prime team and its contractors have convened work groups to

focus on particular aspects of the demonstration design and made presentations to providers,

provider and beneficiary organizations, and other stakeholders. The State’s contractors engaged

in outreach to stakeholders include SC Thrive, a Medicaid beneficiary outreach organization; the

Office for the Study of Aging at the University of South Carolina, a research and training

organization; and the State’s long-term care ombudsman, who is also the demonstration’s Prime

Advocate or ombudsman.

1.3 Changes in Demonstration Design

The demonstration’s three-way contract called for opt-in enrollment to begin on

January 1, 2015, and for passive enrollment to begin in April 2015. The South Carolina

legislature passed a proviso in spring 2015 to delay passive enrollment until April 2016, thus

requiring the demonstration to have 14 months of opt-in enrollment, resulting in a very limited

enrollment during that period (see Section 3, Eligibility and Enrollment for a discussion on this

topic). This delay also caused a postponement in the transition of the HCBS waiver function to

the MMPs from the State. Initially, these functions were to roll out in three phases, from January

2015 through December 2016 (South Carolina three-way contract, pp. 262–73). The delay in the

start of the demonstration and of passive enrollment necessitated a longer transition of the HCBS

waiver functions from the State to the MMPs that began in February 2015 and concluded in

April 2017. One of the HCBS waiver functions did not transfer from the State to the MMPs: by

mutual agreement between the MMPs and the State, oversight of beneficiary self-directed

attendant care workers continued to be delivered by the State. In November 2017, the three-way

contract was re-executed to update the demonstration year dates and to bring the contract into

alignment with certain State and Federal guidelines and program requirements (CMS, 2017a).

In July 2018, CMS and the State announced changes to the South Carolina demonstration

that include: (1) extending the demonstration to December 31, 2020 with all three MMPs,

(2) allowing passive enrollment of eligible beneficiaries who are enrolled in Medicare Advantage

plans, (3) allowing passive enrollment of eligible beneficiaries who reside in counties with only

one MMP, (4) reduction to shared savings effective 2018, and (5) a reduction in some reporting

and other administrative requirements. These changes fall outside the reporting period of this

report and will be discussed in future reports.

1.4 Overview of State Context

This section discusses South Carolina’s experience with managed care, available waiver

services, and effects of changes in State leadership. For a summary of predemonstration and

demonstration design features for Medicare and Medicaid enrollees in South Carolina, see

Appendix E.

State leadership and MLTSS consideration. In fall 2016, the State’s Medicaid Director

announced a suspension of passive enrollment into the demonstration, effective January 1, 2017.

The State engaged external subject matter experts to assess whether an MLTSS program,

including a population not limited to those over age 65, would be more practical for the State.

Subsequently, the Medicaid Director left that position and passive enrollment in the

demonstration was restored in August 2017, but discussions about possibly transitioning the

demonstration to a MLTSS program continued throughout 2017.

A new Medicaid Director was not confirmed until early 2018. During the period from fall

2016 through the informant interviews conducted in fall 2017, MMPs, providers, and

stakeholders were uncertain whether the demonstration would continue beyond December 2018.

It is not clear whether demonstration enrollees were aware of what stakeholders commonly

called “a cloud of uncertainty” over the demonstration. In July 2018, when CMS and the State

announced an extension of the demonstration through 2020, State officials explained that the

State was continuing to study the ramifications of implementing an MLTSS product for all

dually eligible beneficiaries in the State. Any implementation would begin after the close of the

demonstration.

Experience with managed care. From the beginning of the Healthy Connections Prime

demonstration, the RTI evaluation team heard reports that managed care was relatively new in

the State and that providers were generally distrustful of contracting with managed care

organizations, including Healthy Connections Prime plans. Individual providers, such as home

health agencies and nursing facility owners, were particularly hesitant to engage with plans and

often opted for single case contracts for an individual enrollee, rather than contracting more

broadly with the MMP. The basis of their concern was fear of not being paid on time or as

agreed. Providers and stakeholders reported that this concern was well founded. For example,

large practices and small individual providers told of not being paid for services rendered despite

prior authorizations or approvals by Medicare Advantage plans operating in the State.

Healthy Connections Prime plans reported that Medicare Advantage plans and their

brokers provided negative or misleading information to beneficiaries about the viability of the

demonstration. This activity increased during the period of uncertainty beginning in fall 2016,

when the State paused the passive enrollment process to consider a possible change to MLTSS.

The State advised MMPs to collect details about such incidents and inform CMS and the State.

CMS researched these incidents using a “secret shopper” contractor to pose as beneficiaries. The

Center for Medicare issued a technical assistance letter to one managed care plan and contacted

another plan’s account manager to clarify instructions and guidance regarding appropriate

marketing. Similar activity from other managed care plans, agents, and brokers continued and

CMS pursued all complaints; however, obtaining first-hand evidence for these types of activities

is extremely difficult.

The State had originally introduced managed care in 1996 through risk-based managed

care organizations (MCO) serving children, pregnant women, and adults with disabilities who

were not dually eligible. Services included primary, acute, some specialty care, and outpatient

behavioral health services. Enrollment was limited until 2006, when Healthy Connections

Choices, an enrollee support program, was introduced. Operated by the State’s enrollment

vendor, Maximus, Healthy Connections Choices advised Medicaid beneficiaries that they could

choose from three Medicaid delivery models: an MCO, a primary care case management

program, or the traditional fee-for-service (FFS) option. In 2011, South Carolina required

Medicaid beneficiaries to enroll in either an MCO or a Medical Home Network (MHN) and in

January 2014, the MHNs were transitioned to MCOs. Exempted from mandatory Medicaid

managed care are children in foster care, children with disabilities, Medicaid waiver enrollees,

beneficiaries receiving institutional care, and dual eligible beneficiaries.

The State contracts with five plans to provide services to Medicaid managed care

enrollees: Absolute Total Care (Centene), Molina of South Carolina, BlueChoice Health Plan,

WellCare of South Carolina, and First Choice Select Health. These MCOs pay providers

capitated rates for each enrollee assigned to their practices. Three of these MCOs (Absolute

Total Care, Molina of South Carolina, and First Choice Select) are Healthy Connections Prime

demonstration plans. Advicare was initially contracted as an MMP, but when its Medicaid

managed care product was purchased by WellCare in August 2016, it withdrew from the

demonstration. Accountable Care Organizations (ACOs) do not have a significant presence in

South Carolina, according to State officials.

In 2017, approximately 243,030 beneficiaries, about one-quarter of South Carolina’s

Medicare beneficiaries, received care through Medicare Advantage plans. UnitedHealth group,

Humana, and Aetna accounted for 92 percent of the coverage. Medicare-Medicaid beneficiaries

participating in Medicare Advantage may receive care through four special needs plans for

dually eligible beneficiaries (D-SNP) operating in South Carolina. In 2017, approximately

24,364 individuals received care through D-SNPs (Kaiser Family Foundation, 2017). Three

PACE plans are available in the State; enrollment in those plans was 439 as of December 2017

(CMS, 2018b).

HCBS waivers. For individuals with disabilities or complex care needs, South Carolina

operates nine HCBS waivers. Each waiver targets individuals with a unique set of disabilities

and provides appropriate services to them. Beneficiaries participating in three of the nine

waivers, the HIV/AIDS, Mechanical Ventilation Dependent, and Community Choices HCBS

waivers, are eligible to participate in the South Carolina demonstration. The Community Choices

waiver is designed for individuals who require a nursing home level of care and assistance with

their activities of daily living. These waiver enrollees may direct services including hiring their

own personal assistance, respite and companion service providers. Beneficiaries from the

remaining six HCBS waivers (Intellectual Disabilities and Related Disabilities, Head and Spinal

Cord Injury, Community Supports, Medically Complex Children’s, Pervasive Developmental

Disorder, and Psychiatric Residential Treatment Facility Alternative Chance) are not eligible to

participate (MOU, 2013).

Federal funding. South Carolina received $4.5 million in Federal funds in September

2014 to support Healthy Connections Prime planning and implementation. The State used these

funds to contract vendors to assist with developing the demonstration. Vendors include

Maximus, the State’s enrollment broker; Ikaso Consulting for marketing support; SC Thrive for

beneficiary education and outreach; and the University of South Carolina to develop and provide

courses for the MMPs, providers, and stakeholders. Federal funding for planning and

implementation continued in the second demonstration year ($2.8 million) and the State

contributed $0.9 million.

The State Long Term Care Ombudsman office, which functions as the demonstration’s

ombudsman or Prime Advocate, was awarded $246,493 in 2015 by CMS in collaboration with

the Federal Administration for Community Living (ACL). This office received $297,254 in 2016

and the same amount in 2017 to support these services.

[This page intentionally left blank.]

2. Integration of Medicare and Medicaid

This section provides an overview of the management structure that was created to

oversee the implementation of the demonstration and discusses in greater detail the organization,

geographic coverage areas, and the four initial Healthy Connections Prime plans that were

selected to integrate and deliver the Healthy Connections Prime. It also provides a general

description of the other functions (e.g., care coordination, eligibility, enrollment, quality

management, and financing) that SCDHHS, CMS, and the plans had to coordinate or integrate as

part of the implementation of the demonstration. Later sections provide more in-depth discussion

of the implementation successes and challenges associated with the integration of these

functions.

2.1 Joint Management of Demonstration

The CMS-State contract management team (CMT) monitors the MMPs and oversees

nearly every aspect of the demonstration including marketing, enrollment, grievances and

appeals, and the MMPs’ compliance with providing certain services within required timeframes.

The CMT is responsible for monitoring to ensure that State and CMS policies are integrated, to

respond to MMPs’ questions and concerns, and to identify areas for technical assistance.

The CMS team members initially included two representatives from the MMCO, one

Consortium for Medicare Health Plans Operations (CMHPO) staff, and one Consortium for

Medicaid and Children’s Health Operations (CMCHO) staff. By 2017, the MMCO and CMCPO

staff members participated. State members of the CMT include several core Healthy Connections

Prime team members, and, depending upon the discussion topic for the meeting, other key State

Highlights

• Medicare-Medicaid Plans (MMPs) provide the administrative structure to integrate the

financing and delivery of primary and acute care, behavioral health, and long-term

supports and services (LTSS).

• The slow rollout of the demonstration, due to the passive enrollment delay, provided

time for the MMPs to receive training and hone their skills in LTSS and HCBS waiver

services.

• Advicare, one of the four original MMPs, left the demonstration effective August 31,

2016. Most enrollees who had been covered by Advicare were transferred to other

MMPs. As of December 2017, 39 of the 46 counties were covered by the

demonstration.

• MMPs satisfied their provider network requirements despite hesitant participation by

some providers—including hospitals, nursing facilities, and personal care providers—

due to their fear of managed care. MMPs have used single-case contracting to provide

continuity of care of enrollees and to build relationships with providers.

staff. The State’s Prime Advocate or ombudsman for Healthy Connections Prime, who is in the

Long Term Care Ombudsman office in the State’s Office on Aging, joined the CMT in March

2015.

Noting their limited Medicare expertise, State staff reported they relied on the knowledge

of the CMS CMT members to provide details and context of various Medicare regulations and

requirements. MMP interviewees reported that they appreciated the CMT’s quick responses to

policy and procedural issues as a result of having Medicare, Medicaid, and State staff available

in the same meeting. All parties remarked on the effectiveness of this “team approach.” The team

initially met bi-weekly with each MMP to review activities, to provide them with feedback

reports to help them benchmark their performance, and to respond to questions.

As the demonstration has evolved, so have the activities of the CMT. By 2017, the CMT

was meeting monthly with each plan. The monthly, State-led operational meeting with all plans

discusses the MMPs’ operations reports that show monthly trends of (1) the ratio of care

coordinators to active enrollees; (2) enrollee counts of those receiving flexible or waiver-like

services; (3) open and closed grievances and appeals; (4) issues tracking by low, medium, and

high priority areas, and other relevant topics. This report also includes the results of MMPs’ exit

interviews with enrollees who have disenrolled from each plan.

2.2 Overview of Integrated Delivery System

2.2.1

Healthy Connections Prime Plans

After a competitive bidding process and extensive readiness reviews, South Carolina and

CMS contracted with four Medicare-Medicaid Plans (MMPs), (originally known as Coordinated

and Integrated Care Organizations or CICOs) to authorize, integrate, and coordinate the full

range of Medicare and Medicaid services including primary, acute, and behavioral health care,

and institutional and community-based LTSS, and a new palliative care benefit. The four MMPs

had extremely varied backgrounds; two had extensive experience with Medicaid managed care in

the State. Three did not have State experience integrating Medicare and Medicaid services in a

product such as a D-SNP; one had minimal experience (see Table 1). At the start, the

demonstration covered 41 of the 46 counties; after the Advicare departure 39 counties were

covered. For further discussion of the Advicare departure and the effect on enrollment, see

Section 3, Eligibility and Enrollment.

2.2.2

Provider Arrangements and Services

Plans noted that outside of the three main metropolitan areas (Charleston, Columbia, and

Greenville), most of the counties are rural and have a limited range of specialists, yet MMPs are

required to contract with a full complement of providers in most specialties in each county. Plans

pointed out that some counties do not have cardiologists or oncologists, which precluded the

MMP from operating in particular counties. MMPs believed that these requirements were too

stringent given the number of small counties in the State, and noted that providers located in a

neighboring county might be closer to a beneficiary than providers located in the beneficiary’s

home county. Nevertheless, MMPs contracted with a sufficient number of primary and specialty

providers to fulfill the CMS minimum network adequacy requirements prior to the start of the

demonstration.

Table 1

Healthy Connections Prime MMP profiles

Parent

company

Healthy

Connections

Prime MMP

No. of

counties

served as

of January

2016*

No. of counties

served as of

December 2017

experience with

Medicare

Advantage or

D-SNP

experience with

Medicaid

managed care

National

chain

Centene

Corporation

Absolute Total

Care (ATC)

32 35 No Yes Yes

Spartanburg

Regional

Healthcare

Advicare

Advocate**

39 0 No Yes No

Molina

Healthcare

Molina

Healthcare of

South Carolina

22 29 No Yes Yes

AmeriHealth

Caritas

Select Health of

South Carolina

38 39 Yes Yes Yes

* Total of 46 counties

** Advicare left the demonstration after its purchase by WellCare Health Plans, Inc., in August 31, 2016.

SOURCE: SCDHHS monthly enrollment dashboards (South Carolina 2018).

Provider reluctance. As discussed in Section 1.4, Overview of State Context, some

providers were reluctant to contract with MMPs, explaining that this was due to the poor

reputation of Medicare Advantage plans operating in the State. Other providers noted their lack

of experience with, and a general fear of, managed care plans. These factors contributed to

MMPs’ difficulty in developing robust provider networks with which they could develop value-

based purchasing agreements. See Section 7, Financing and Payment for more on this topic.

Con

tin

uity of care. Although skilled nursing facilities, individual providers, and some

large healthcare networks declined to enter into contractual arrangements with MMPs, they

chose to enter into single case contracting with MMPs as needed. The demonstration’s continuity

of care provision allows for enrollees to continue seeing their previous provider up to six months

prior to transitioning to a Healthy Connections Prime provider (SCDHHS, 2017). In RTI’s

annual site visit interviews, the team consistently heard from MMPs that most providers readily

engaged in single case agreements for enrollees as needed. The MMPs used these agreements to

build relationships with providers that often led to full contracts.

Behavioral health providers. Plans contracted with the State Department of Mental

Health (DMH) for the provision of behavioral health services in their community health centers

located in each county. Because of limited capacity of these providers and low utilization by

aging beneficiaries, the State encouraged MMPs to also contract with private behavioral health

providers. MMPs may, but are not required to, contract with the Department of Alcohol and

other Drug Abuse Services, which provides treatment for substance use; they also contract with

private providers for these services.

By summer of 2016, of the more than 9,000 enrollees, only 149 (1.7%) were receiving

services from DMH providers. By 2017, the State and the plans reported low overall utilization

of mental health and behavioral health services. Challenges included the State’s lack of capacity

to serve the aging population; the DMH’s reported reluctance to treat beneficiaries who have a

primary diagnosis of dementia; and providers’ focus on disruptive behavior, while conditions

such as depression were under-treated (please see Section 5, Beneficiary Experience for more

on this topic).

Transition of HCBS waiver services. A key feature of the South Carolina

demonstration is the transition of HCBS waiver roles and responsibilities from the State to the

MMPs. The SCDHHS Community Long Term Care (CLTC) division has had responsibility for

these functions for the three waiver populations included in the demonstration. The phased

transition process, developed with significant stakeholder input, was designed to systematically

introduce these functions to the MMPs while preserving the State’s authority until each MMP

demonstrated readiness. Table 2 shows the transition schedule for HCBS waiver functions.

Table 2

Transition of HCBS waiver functions from State to MMP

HCBS waiver functions

Phase I

February 2015

Phase II

September 2016

Phase III

April 2017

HCBS provider credentialing/monitoring State State

State; MMP may

choose responsibility

HCBS provider contracts State MMP MMP

HCBS care plan development & authorization

State with MMP

formal input

MMP with State

approval

MMP with State

approval

Oversight of State waiver case manager’s

participation in multidisciplinary team

MMP MMP MMP

HCBS provider rate setting State

MMP within State

guidelines

MMP within State

guidelines

HCBS claims processing via Care Call and

provider payments

State MMP MMP

LTC level of care assessments State State State

LTC level of care reassessments State MMP MMP

Self-directed attendant care and related

functions

State State State

HCBS = home and community-based services; LTC = Long term care; MMP = Medicare-Medicaid Plan.

Per mutual agreement, the State has retained responsibility for processing HCBS provider payments; MMPs’

Medicaid capitation includes HCBS; therefore, these payments are subsequently reconciled

Effective July 1, 2018, MMPs are responsible for approving the LTC reassessments.

The July 1, 2018 contract amendment lists the MMP as having this responsibility for Phase III although the State

will continue to have responsibility.

SOURCES: CMS amendment to South Carolina three-way contract (July 1, 2018) and South Carolina Healthy

Connections Prime Demonstration contract re-execution (November 1, 2017).

Phases 1, 2, and 3 were originally scheduled to be completed by December 2015, June

2016, and December 2016, respectively, but were adjusted due to the delay in passive

enrollment. Approximately 200 of the 1,800 beneficiaries (11.1%) who opted into the

demonstration during the first year were HCBS waiver participants. After the April and July

2016 passive enrollment waves, approximately 1,500 waiver participants joined the

demonstration. By December 2017, 1,738 waiver participants were enrolled. Section 4, Care

Coordination discusses the transition of HCBS waiver enrollee care management from provision

solely by the CLTC case managers to joint provision with the MMP care managers.

2.2.3

Training and Support for Plans and Providers

The State engaged the University of South Carolina’s Office for the Study of Aging to

train MMPs on HCBS and LTSS service provision, services that were new to them. This and

other trainings (e.g., assessment process, care planning, and dementia care), began in 2015 and

continued throughout 2016. Electronic copies of the trainings were subsequently available to

MMP staff. MMPs reported that the training was comprehensive and well-delivered. In addition

to trainings provided by the State’s contractor, MMPs track and report ongoing trainings their

staff receive on monthly reports submitted to the CMT. Providers such as nursing facility staff

and personal care providers, who were new to managed care, were also provided formal training

on the demonstration by the State’s training contractor. Plans and providers were also invited to

attend learning collaboratives and other related trainings provided by the Office for the Study of

Aging at the University of South Carolina. For more on this topic, please see Section 6,

Stakeholder Engagement.

An important element of MMP training for HCBS service provision related to Phoenix,

the State’s automated HCBS waiver case management and service authorization system. This

data system maintains records of essential functions, including all intake, assessment, and care

planning activities; it includes a section for home assessment data, caregiver supports, and

quality indicators. Phoenix has capacity for provider notes, correspondence among users, and

various features (e.g., alerts for follow-up appointments) can be built in to ensure compliance

with Federal regulations, State policies, or programs. The system also includes a method to

identify waiver participants at risk for missed in-home visits and those most at risk in the event

of natural disasters. CLTC program staff and developers expanded Phoenix and tailored it to the

needs of the demonstration.

Plans are required to use Phoenix as a centralized electronic record system to document

all assessments, individual care plans, provider information, caregiver support systems, waiver

case management, and quality assurance activities for demonstration enrollees. The State

maintains the system and may monitor enrollee data entered by plans and use it to generate

reports. Security is tailored to allow users access only to needed sections.

The Care Call feature of Phoenix automates prior authorizations of specific services and

provides real time service monitoring and billing. For example, upon starting and ending their

tasks, home care workers call a toll-free number to document services delivery or they may use

an application in their smartphones to report their hours worked for a particular enrollee. If Care

Call’s GPS system detects that the worker is not in the vicinity of the enrollee’s home, it triggers

a review of the worker’s hours and functions. Services are linked with the prior authorization to

ensure appropriateness of services and Care Call generates weekly electronic billing to MMIS

for those services.

2.3 Major Areas of Integration

Benefits. Beneficiaries may enroll in the demonstration, through the State’s enrollment

broker, to receive integrated Medicare and Medicaid benefits. Their benefits, including the new

care coordination, palliative care, and additional flexible benefits are coordinated by one MMP.

See Section 1.2, Model Description and Goals and Section 4, Care Coordination.

Quality. Four activities comprise the demonstration’s quality oversight: (1) joint

monitoring and oversight by the State and CMS, (2) external quality review activities conducted

by the External Quality Review Organization, (3) quality and performance improvement

initiatives undertaken by the plans, and (4) quality reporting and measurement (see Section 9,

Quality of Care).

Financing. Healthy Connections Prime plans are paid a blended, risk-adjusted capitated

rate covering all Medicare and Medicaid services. Medicare Parts A and B and Medicaid

payments reflect the application of savings percentages and quality withholds (see Section 7,

Financing and Payment).

3. Eligibility and Enrollment

3.1 Introduction

This section provides an overview of the enrollment process for Healthy Connections

Prime. Eligibility for the demonstration, enrollment phases, and the passive enrollment process

are included in this section. Enrollment data are presented and factors influencing enrollment

decisions are also discussed.

3.2 Enrollment Process

3.2.1

Eligibility

Full benefit Medicare-Medicaid enrollees aged 65 or older and living in the community at

the time of enrollment are eligible to enroll in Healthy Connections Prime. Beneficiaries

participating in the Community Choices, HIV/AIDS, or Mechanical Ventilation Dependent

waiver programs for home and community-based services (HCBS) are eligible for the

demonstration. Individuals who transition from a nursing facility into the community, and are

otherwise eligible, may elect to enroll in the demonstration. Individuals already enrolled who

later enter a nursing facility may remain in the demonstration (MOU, 2013, p. 10). Any enrollee

residing in a nursing facility for more than 90 days before transition from that facility also may

Highlights

• Legislative action that delayed the start of passive enrollment for 1 year prohibited the

demonstration from growing and expanding during this critical period.

• After passive enrollment occurred in two waves in 2016, it continued for beneficiaries

who aged in or otherwise became eligible for the demonstration. This supported

sustained enrollment as other enrollees left the demonstration when they became

ineligible through loss of Medicaid, moved out of the demonstration counties, or died.

• While the State considered a change to another model that might include the

demonstration population, it suspended passive enrollment for 7 months beginning

January 2017. This resulted in reduced overall enrollment and created concern among

plans and stakeholders about the demonstration’s viability. It was not clear if

beneficiaries were aware of this uncertainty.

• From the beginning of the demonstration through 2017, plans and stakeholders

reported negative behavior by Medicare Advantage plan representatives. Reportedly,

representatives provided incorrect information to beneficiaries about the status and

future of Healthy Connections Prime with the intent of attracting beneficiaries away

from the demonstration. CMS investigated these allegations.

• Limitations of the State’s eligibility system posed challenges for identifying eligible

beneficiaries, particularly those in nursing facilities.

qualify for the State’s Money Follows the Person (MFP) Rebalancing demonstration, called

Home Again (South Carolina three-way contract, 2014, p. 97). As of December 2017, no

demonstration enrollees had also qualified for or participated in the MFP demonstration.

During the initial planning stages of the demonstration, the State estimated 53,600

beneficiaries might be eligible for the demonstration. This figure was adjusted downward for a

number of reasons. While the State considered approximately 17,000 enrollees receiving

comprehensive benefits in other Medicare products (e.g., Medicare Advantage) eligible to opt in,

the evaluation does not consider these enrollees eligible for the demonstration while they are

enrolled in another product. Another 7,581 beneficiaries were residents of counties where none

of the MMPs had an approved provider network at the beginning of the demonstration. In

October 2015, the State estimated that approximately 27,000 beneficiaries met the CMS and RTI

definition for eligibility in the demonstration. This number included approximately 6,000

beneficiaries participating in the low income subsidy (LIS) program for their Medicare Part D

benefits. The number of eligible beneficiaries slowly decreased, and by mid-2016 the number

stabilized to approximately 20,000. Table 5 shows the number of eligible beneficiaries in the

span of the demonstration through December 2017.

MMPs entered into the three-way contracts, developed their Healthy Connections Prime

products, and contracted with providers, based on expectations of approximately 53,600

beneficiaries who would be eligible for the demonstration. With enrollment limited to opt-in

enrollment during the first year, enrollees in each of the four MMPs ranged from approximately

200 to 600 and totaled approximately 1,800. By December 2017, enrollment totaled

approximately 11,500, with 3,000 to 5,100 enrollees in each of the three plans. During RTI

interviews in each year, MMP officials expressed their concern and frustration about their

increased level of effort and reporting requirements with so few enrollees. Plans had expected at

least 12,000 enrollees each, but had fewer than half of that number. For further discussion on this

topic, see Section 7, Financing and Payment.

3.2.2

Phases of Enrollment

According to the three-way contract, the MMPs were to begin opt-in enrollment on

January 1, 2015. Subsequently, this was delayed by 1 month. SCDHHS sent an opt-in welcome

letter to all beneficiaries eligible for Healthy Connections Prime in late 2014 for the start of the

demonstration on February 1, 2015. Approximately 1,200 enrollees signed up in the first month

and that number increased and held to approximately 1,800 throughout the first year of the

demonstration prior to the initiation of passive enrollment. This number included about 200

HCBS waiver enrollees.

In the spring of 2015, shortly before the mailing of 60-day notices to eligible

beneficiaries to begin passive enrollment, the State legislature introduced legislation to delay the

start of passive enrollment until April 1, 2016, stipulating that participation in Healthy

Connections Prime would be limited to individuals who affirmatively elect to participate (South

Carolina Senate, 2015). Therefore, the demonstration had 14 months of opt-in enrollment prior to

the start of passive enrollment. This unexpected legislation upended the Healthy Connections

Prime team’s passive enrollment plans and the MMPs’ expectation for enrollments and required

both to drastically revise their projections and rework their strategies.

To encourage beneficiaries to opt into the demonstration during these 14 months prior to

the start of passive enrollment, the State engaged in several activities. The demonstration team

and its contractors made presentations at events to educate potential enrollees, approached

providers to join the demonstration, revised beneficiary materials to improve readability, and

surveyed disenrollees to understand their disenrollment reasons. Although there was marginal

success in enrollment opt-ins during this period, there were also significant disenrollments due to

beneficiaries becoming ineligible, moving out of coverage areas, and death. The State also

conducted outreach to LIS beneficiaries to attract them to the demonstration during their annual

enrollment period. The State reported that this effort to enroll this difficult to reach population

was not successful.

Passive enrollment began with two waves in April and July 2016 in the upstate and

coastal counties and the HCBS population (see Table 3). The 46 counties in the State are

designated geographically as upstate (or inland) counties and coastal counties. The former are

situated in the northwestern region that includes Columbia, the State capitol; coastal counties are

located in the southeastern part of the State that includes Charleston. Enrollees in the upstate

region were passively enrolled with an effective coverage date of April 1, 2016, and those in the

coastal counties began their coverage on July 1, 2016. Also included in the July coverage date

were beneficiaries who receive HCBS waiver services from SCDHHS. After the two initial

waves, ongoing passive enrollment continued in November 2016 as beneficiaries became eligible

for the demonstration (e.g., when they moved into demonstration counties, became dually

eligible for Medicare and Medicaid).

Table 3

Healthy Connections Prime enrollment phases

Population

Opt-in

enrollment

began

Initial passive

enrollment

schedule

Passive

enrollment

occurred

Ongoing

passive

enrollment

paused

Ongoing

passive

enrollment

resumed

Upstate (inland) region February 1, 2015 June 1, 2015 April 1, 2016 January 1, 2017 August 1, 2017

Coastal region February 1, 2015 August 1, 2015 July 1, 2016 January 1, 2017 August 1, 2017

HCBS enrollees February 1, 2015 October 1, 2015 July 1, 2016 January 1, 2017 August 1, 2017

N.B., dates are effective enrollment dates. Ongoing passive enrollment refers to passive enrollment of beneficiaries

who are newly eligible for the demonstration.

SOURCE: State enrollment schedules.

Beneficiaries who were passively enrolled received 60- and 30-day notices that explained

the benefits of the demonstration. SCDHHS utilized an intelligent assignment algorithm to

identify the best MMP to meet the needs of enrollees. The methodology included existing

provider relationships, previous history with an MMP product, and household members who

were assigned to a MMP (South Carolina three-way contract, 2014, p. 29). Beneficiaries could

select an MMP; if they did not select one, they were assigned, using the intelligent assignment

algorithm, to one that contracted with the prospective enrollee’s usual source of care. Enrollees

could opt out of the demonstration prior to passive enrollment or disenroll at any time; they

could also change MMPs, and beneficiaries who opt out of Healthy Connections Prime may

reenroll at any time as long as they still meet the eligibility requirements.

Advicare departure. Advicare’s MMP was included in the demonstration from the start.

When Advicare’s Medicaid managed care product was purchased by WellCare, Inc., the MMP

was disbanded and the company withdrew from the demonstration effective August 31, 2016.

The State and Advicare began informing enrollees and beneficiaries slated for passive enrollment

into Advicare as early as June 2016. Approximately 3,100 Advicare enrollees and prospective

enrollees were absorbed by the three remaining MMPs in counties that had at least two MMPs.

The remaining 460 Advicare members affected by the departure were enrolled into Healthy

Connections Medicaid and fee-for-service Medicare with Part D plans, because they lived in

counties where no or only one MMP was available. Beneficiaries living in counties in which one

MMP operated (420 beneficiaries) were given the opportunity to opt into the demonstration.

Passive enrollment suspension. In fall 2016, while the State considered a change to an

MLTSS model for this population, the Medicaid Director suspended passive enrollment into

Healthy Connections Prime effective January 2017. Subsequently, a new administration decided

to proceed with ongoing passive enrollment which resumed in August 2017. Please see

Section 1.4, Overview of State Context for background on this topic.

3.2.3

Passive Enrollment Experience

During the April and July 2016 waves of passive enrollment, approximately 7,100 new

beneficiaries were enrolled into Healthy Connections Prime, bringing total enrollment to 9,002

by August 2016, approximately 44 percent of those eligible.

The State undertook several efforts to understand the reasons for enrollees disenrolling

from the demonstration after they had opted in, and for opting out after they had been passively

enrolled. The State’s contractor, Clemson University, conducted a survey in spring 2015, and SC

Thrive, the State’s outreach contractor, conducted a second survey in July 2016 after the first

wave of passive enrollment. The State also analyzed its own data on cancelations and opt-outs to

understand any emergent patterns. Later, in spring 2017, MMPs conducted exit interviews of

their members who had disenrolled. All analyses found that beneficiaries opted out of the

demonstration primarily because they were satisfied with their provider and their provider was

not in the network of any of the MMPs.

Secondary issues for disenrollment had to do with confusion about some of the language

used in marketing materials and concerns about loss of Part D benefits. During the 2016 passive

waves, State officials, plans, and stakeholders reported confusion among beneficiaries when

individuals scheduled for passive enrollment received disenrollment notices from their Part D

prescription drug plans before they received Healthy Connections Prime passive enrollment

notices and materials. This generally prompted prospective enrollees to cancel their enrollment,

believing they would lose their Part D prescription drug plan. To mitigate this concern, CMS and

the State worked together to revise notices, making it clear that the demonstration included Part

D benefits. CMS worked to develop new scripts for 1-800-Medicare staff. When misinformation

from Part D plans was reported, CMS followed up with “secret shopper” calls. After the first

wave of passive enrollment MMPs began making early welcome calls to enrollees within 3 days

of receiving new enrollee data from the State. During the calls, MMP staff educated the new

enrollees and confirmed all benefits, including the Part D benefits that they have through the

MMP. This activity has continued for all subsequent new enrollees.

Another effort to improve beneficiary understanding related to language and marketing

came from beneficiary feedback regarding confusion about the demonstration’s benefits and how

they differ from the State’s Medicaid program, Healthy Connections. This confusion prompted

some beneficiaries to disenroll from the demonstration because they did not distinguish Healthy

Connections from Healthy Connections Prime. To try to differentiate the two products and

explain the benefits of Healthy Connections Prime, the demonstration team redesigned the

demonstration website and developed new marketing and outreach materials written in a simpler

language and format.

3.2.4

Integration of Medicare and Medicaid Enrollment Systems

The Healthy Connections Choices Customer Service Center is operated by the State’s

enrollment vendor, Maximus, which was trained by the State on the details of Healthy

Connections Prime. Beneficiaries may call the service center for options counseling and to

enroll. Maximus developed generic materials presenting enrollment options for eligible enrollees

(South Carolina SCDHHS, n.d., p. 2). The options counselor, SC Thrive, is also responsible for

assisting enrollees who wish to transfer to a different MMP.

SC Thrive is a beneficiary advocacy organization that provides education and outreach

activities throughout the State about a range of Federal and State programs. Contracted by the

State to add online enrollment in Healthy Connections Prime and outreach to eligible

beneficiaries, it provides individual in-person options counseling with someone to answer

questions and concerns, walk beneficiaries through the enrollment process, and provide a “warm

hand-off” (e.g., a person-to-person transfer from the counselor) to the enrollment broker to

complete the enrollment process. Additional information on beneficiary education and

engagement is provided in Section 5, Beneficiary Engagement.

3.2.5

Contacting and Locating Enrollees

As the first step in coordinating care for new demonstration enrollees, a care manager

contacts the enrollee to conduct a health risk assessment. (Please see Section 4.1.1, Assessments,

for more information on this topic). Table 4 displays the percentage of members that MMPs

were unable to reach within the first 3 months of enrollment. During 2015, these percentages

were relatively low, reflecting that all enrollees had opted into the demonstration in that year.

However, when passive enrollment began in the second quarter of 2016, MMPs began to have

difficulty contacting enrollees due to incomplete, erroneous, or outdated contact information.

The percentage of members that MMPs were unable to reach within 90 days of enrollment

increased to a high of 26.4 percent (979 of 3,707 enrollees) in quarter 3 of 2016.

Plans used vendors or their own care managers or community navigators to visit the

previous addresses and ask neighbors where to find the new enrollees. They found that some had

moved to a nursing facility, a community residential care facility, or the home of a family

member. Because there were fewer new enrollees after passive enrollment was suspended in

January 2017, MMPs were able to reach their new enrollees more easily. But after passive

enrollment was resumed in August 2017, plans had difficulty reaching all 3,870 new enrollees

whose 90

day of enrollment occurred in quarter 4 of 2017, and the percentage of members that

MMPs were unable to reach increased to 38.7.

Table 4

Percentage of members that Healthy Connections Prime plans were unable to reach

following three attempts, within 90 days of enrollment, by quarter

Quarter CY 2015 CY 2016 CY 2017

Q1 N/A 6.6% 25.8%

Q2 7.6% 21.9% 14.4%

Q3 4.7% 26.4% 13.3%

Q4 6.8% 19.3% 38.9%

CY = calendar year.

NOTES: N/A indicates data are not applicable. Advicare withdrew from the demonstration and is not included in

data after Q3 2016.

SOURCE: RTI analysis of MMP reported data for Core Measure 2.1 provided to RTI as of October 2018. The

technical specifications for this measure are in the Medicare-Medicaid Capitated Financial Alignment Model Core

Reporting Requirements document, which is available at:

https://www.cms.gov/Medicare-Medicaid-

Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-Coordination-

Office/FinancialAlignmentInitiative/MMPInformationandGuidance/MMPReportingRequirements.html

3.3 Summary Data

As of December 2017, approximately 11,511 beneficiaries were enrolled in the Healthy

Connections Prime demonstration, representing about 56 percent of the eligible population.

Table 5 displays how key events, discussed in this section, have affected enrollment in Healthy

Connections Prime. By January 2016, after 12 months of opt-in only enrollment, total enrollment

was 1,773 or approximately 7 percent of those eligible for the demonstration. The April and July

2016 waves of passive enrollment increased enrollment to more than 44 percent of those eligible,

but after the departure of one MMP and the suspension of passive enrollment, the percentage of

eligible beneficiaries enrolled in the demonstration dipped to approximately 37 percent by July

2017. After passive enrollment resumed, enrollment reached more than 11,000 from August

through December 2017.

Table 5

Key events affecting enrollment and percentage enrolled in Healthy Connections Prime

Enrollment dates and key events

Beneficiaries

eligible for the

demonstration Enrollment

Percentage

enrolled

January 2016; after 12 months of opt-in only enrollment 24,809 1,773 7.1

August 2016; after April and July 2016 passive enrollment waves 20,288 9,002 44.4

October 2016; effect of August 2016 Advicare departure 20,802 7,862 37.8

January 2017; suspension of passive enrollment 20,898 8,981 43.0

July 2017; effect of 7 months without passive enrollment 20,901 7,818 37.4

August 2017; passive enrollment resumed 20,320 11,468 56.4

December 2017; ongoing passive enrollment 20,726 11,511 55.5

SOURCE: RTI State Data Reporting System; data are reported on the last day of each month.

As of December 2017, State data indicate that approximately 16 percent of enrollees had

opted into the demonstration; the remainder had been passively enrolled. Select Health provided

care to approximately 44 percent of enrollees, and ATC covered 30 percent; the remaining 26

percent were covered by Molina. Although churning had been reported by MMPs in 2016 with

enrollees switching plans frequently, this had stabilized and there were no reports of churning in

2017.

State data also show that 15 percent of all enrollees were enrolled in one of the three

HCBS waivers (two enrollees were in the Mechanical Ventilator Dependent waiver; 30 enrollees

were in the HIV/AIDS waiver, and 1,706 enrollees were in the Community Choices waiver).

Only two percent of all enrollees were in a skilled or a long-stay nursing facility as of December

2017 (SCDHHS, 2017).

[This page intentionally left blank.]

4. Care Coordination

4.1 Care Coordination Model

This section provides an overview of the demonstration requirements related to the care

coordination function, including assessment processes; use of multidisciplinary teams (MT) and

the development of individual care plans (ICP); delivery of care coordination services; and the

role of care managers. The experience of Medicare-Medicaid Plans (MMPs) are included in this

section as is the care coordination of LTSS and behavioral health services and data exchange.

On July 1, 2018, CMS announced contract amendments that include a reduction in a

number of requirements for conducting assessments (CMS, 2018a). These changes will be

discussed in future RTI reports. The descriptions below cover the period through 2017.

Care coordination, a central function of Healthy Connections Prime, is provided by the

plans to all enrollees through care coordinators, referred to in South Carolina as care managers,

and multidisciplinary care teams. The plans’ care coordination models are intended to be person-

centered, to promote enrollees’ ability to live independently, and to coordinate the full set of

Medicare and Medicaid benefits, including medical, behavioral health, LTSS, and social support

Highlights

• Every Healthy Connections Prime enrollee has a care manager to coordinate needed

services identified by the enrollee, provider, or care manager. Plan community health

navigators support the manager and the enrollee by connecting enrollees with

community-based services.

• In-home assessments, including the evaluation of social determinants of health, are an

essential component of developing an individual care plan for demonstration

enrollees. Care plans, which include language, culture, and service history of each

enrollee, and identify medical, behavioral, functional, and psychosocial needs of the

enrollee, are developed by the enrollee, his/her family supports, the care manager, and

providers.

• Enrollees who also participate in the three home and community-based waivers have

access to both a plan care manager and a State waiver case manager to manage their

medical, behavioral health, and long-term services and supports needs. By 2017, the

roles and responsibilities of these two types of managers have been clearly delineated

to improve support for enrollees.

• Phoenix, the State’s electronic case management and service authorization system,

was modified to meet the demonstration’s needs, including recording all care planning

activities. Plans reported the system was cumbersome and redundant with their own

data systems. In summer 2018, the State agreed to reduce some reporting

requirements.

services. The State’s Division of Community Long Term Care (CLTC) case managers have

extensive experience in coordinating care for HCBS waiver enrollees, supported by the State’s

Phoenix case management and Care Call reporting systems. Plan care managers are required to

utilize these systems for assessments, documentation and coordination of services for

demonstration enrollees (South Carolina three-way contract, 2014, p. 57).

Care managers. Care managers, the primary point of contact for the enrollee, are

responsible for ensuring that the initial screen, comprehensive assessment, and ICP are

completed pursuant to the timeframes outlined in the three-way contract. This individual is also

the lead member of the multidisciplinary team, who engages all relevant providers and supports

to ensure coordinated care is provided to the enrollee. Care managers serving enrollees at

moderate to high risk must have a clinical background; those serving enrollees assessed as low

risk must have a bachelor’s degree at a minimum. Plans train all care managers in person-

centered planning, cultural and disability competencies, compliance with the Americans with

Disabilities Act, independent living and recovery, and wellness philosophies (South Carolina

three-way contract, 2014, p. 54). In addition, care managers participate in Dementia Dialogues,

“a structured dementia education program that supports a person-centered dementia care

framework developed by the Office for the Study of Aging and the University of South

Carolina” (Healthy Connections Prime, 2018). This training better equips care managers to

recognize the signs and symptoms associated with Alzheimer’s Disease and related disorders

(ADRD). It also teaches effective communication with caregivers of those affected by ADRD

and trains staff and caregivers how to respond appropriately to challenging behaviors by training

both “staff and caregivers to ‘step into the world’ of the person with dementia” (Curtis, 2017).

The care manager makes referrals, monitors outcomes, and communicates regularly with

the enrollee or the enrollee’s designated representative or informal caregiver about medical,

psychological, and social needs. These check-ins may be via phone or in person, depending upon

the enrollee’s preferences, and are conducted monthly for high risk enrollees and every 90 and

120 days for those stratified as low or moderate risk, respectively. An enrollee may request a

change in care managers at any time.

Healthy Connections Prime enrollees also have access to community health navigators.

Although the role of community health navigators is not defined in the three-way contract, all

three plans noted the importance of these individuals who are regionally located across the State.

Navigators are knowledgeable about local community or faith-based organizations that provide

supports, they work under the direction of the care manager and arrange for services or

accompany the enrollee to medical appointments. When visiting new enrollees in remote rural

areas, plans reported that it was more efficient to conduct the assessments with both the care

manager and the navigator present, to enable both individuals to develop rapport with enrollees

and begin connecting them to services immediately. Navigators may also accompany enrollees to

appointments with physicians and other providers.

4.1.1

Assessment

Risk stratification and the comprehensive assessment. Within the first 30 days of

enrollment, all enrollees participate, via phone or in person, in an initial health screen that

collects medical, psychosocial, LTSS, function, and cognitive needs, and determines low,

moderate, or high risk. All HCBS waiver enrollees are considered high risk. Each new enrollee

also participates in a comprehensive assessment of medical, behavioral health, community-based

or facility-based LTSS, and social needs within 60 days for high or moderate risk, or 90 days if

the enrollee was deemed low risk through the screener. Plans may elect to complete the

comprehensive assessment and the screener at the same time if they are completed within 60

days. Plans consistently reported that this was their preference, saying that it was more efficient

for both the enrollee and the care manager. The initial health screens were developed by

SCDHHS together with the plans, whereas the State’s assessment tool is required to be used for

the comprehensive assessment. Reassessments are required to be conducted every 12 months or

upon certain trigger events as specified in the three-way contract, e.g., transitions, changes in

diagnoses, etc. (CMS, September 5, 2014, pp. 58–63).

The comprehensive assessment includes social, functional, medical, behavioral, wellness

and prevention domains, caregiver status and capabilities, as well as the enrollee’s preferences,

strengths, and goals. All plans hired care managers on staff, but because the demonstration

covers extensive rural areas in the State, some plans initially contracted with home care and in-

home medical care service agencies located throughout the State to conduct the screenings and

assessments. Plan care managers generally conducted assessments except during waves of

passive enrollment, during which plans also engaged contractors in order to meet the assessment

completion requirements.

Plans reported some challenges reaching new enrollees to welcome them to the plan and

to conduct the initial health screen and assessment. Plans sent postcards or visited enrollees

whose enrollments had been completed without telephone numbers. CMS contract management

team members offered to provide addresses from the Social Security database, and the Medicaid

agency supplied provider contacts for the plans to obtain current contact information. Care

managers and community health navigators were generally able to reach new enrollees within

mandated time frames, according to both the plan interviewees and the CMS contractor’s

monitoring reports. By 2017, all plans had completed assessments within the required time

frames in order to meet the quality target. Please see Section 9, Quality of Care, for more

information on quality measures.

Table 6 shows the proportion of enrollees willing to participate in the assessment process,

who could be reached, and who completed assessments within 90 days of enrollment. Since the

second year of the demonstration, the proportion has consistently stayed above 90 percent. When

passive enrollment waves occurred, the percentage of all enrollees with assessments completed

within 90 days of enrollment declined (59.9 percent in quarter 3, 2016; 48.9 percent in quarter 4,

2017), reflecting the challenges in reaching enrollees (Section 3.2.5, Contacting and Locating

Enrollees).

Table 6

Total percentage of enrollees whose assessment was complete within 90 days of enrollment

Quarter

Total number of enrollees

whose 90th day of enrollment

occurred within the reporting

period

Assessment completed within 90 days of enrollment (%)

All enrollees

All enrollees willing to participate

and who could be reached

2015

Q1 N/A N/A N/A

Q2 1,470 72.8 80.8

Q3 321 74.1 79.3

Q4 177 83.1 89.6

2016

Q1 226 89.4 97.6

Q2 3,824 65.6 93.0

Q3 3,707 59.9 95.1

Q4 1,281 72.8 96.5

2017

Q1 1,874 63.8 92.1

Q2 181 75.1 95.8

Q3 180 73.9 91.7

Q4 3,870 48.9 95.3

NOTES: N/A indicates data are not applicable. Advicare withdrew from the demonstration and is not included in

data after Q3 2016.

SOURCE: RTI analysis of MMP reported data for Core Measure 2.1 pro

vided to RTI as of October 2018. The

technical specifications for this measure are in the Medicare-Medicaid Capitated Financial Alignment Model Core

Reporting Requirements document, which is available at:

https://www.cms.gov/Medicare-Medicaid-

Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-Coordination-

Office/FinancialAlignmentInitiative/MMPInformationandGuidance/MMPReportingRequirements.html.

Home assessments and social determinants of health. Home assessments are not

delineated in the three-way contract; however, all plans and State officials noted that these were

essential when conducting comprehensive assessments. Interviewees reported that when visiting

new enrollees, particularly in remote rural areas, assessors evaluate the availability of basic

resources to meet the enrollees’ daily needs. Such resources included the availability of food,

presence of a working refrigerator, indoor plumbing, and safe housing features such as secure

doors. When these resources are lacking, community health navigators and the care managers

identify resources in the community to improve such conditions. The State, MMPs, and

providers all viewed assessments of these social determinants of health as essential for the

demonstration population because of rural isolation, lack of family or other informal caregiver

support, and general conditions of poverty of some enrollees. In addition to initial assessments,

enrollees must be reassessed within 365 days from the previous assessment. In 2016, MMPs

became responsible for the annual reassessments and reported frustration with this requirement,

believing that reassessments were unnecessary and time consuming for the majority of the

population. In late 2017, the State reported discussions of enrollee “assessment fatigue” from the

MMPs, especially for enrollees also being assessed for long-term services and supports (LTSS).

At that time, the State was examining the assessment requirements specifically for enrollees who

had been assessed as low risk.

LTSS assessment. Should long-term care needs be identified during the comprehensive

assessment, plan care managers input referrals for LTSS assessments into Phoenix within 24

hours of completion of the comprehensive assessment. The State’s CLTC care managers then

conduct LTSS assessments using a multifunctional tool that collects cognitive, functional, and

activities of daily living data about individuals who may require access to nursing facility or

HCBS waiver services. Level of care, service eligibility, and the service plan are recorded in

Phoenix, which can be accessed by all relevant parties. In 2017, CLTC care managers continued

to provide initial LTC assessments and determine LTC level of care during the Healthy

Connections Prime demonstration; however, plans became responsible for the annual

reassessments in 2016 (see Table 2, Transition of HCBS waiver functions from State to

MMPs).

4.1.2

HCBS Waiver Services and Coordination

Waiver service transition. As discussed in Section 2.2.1, Provider Arrangements,

transition of HCBS waiver services to the MMPs was a goal of Healthy Connections Prime.

Early in the demonstration, CLTC maintained responsibility for all waiver activities including

assessment, development, and monitoring of service plans for HCBS waiver participants who

enrolled in Healthy Connections Prime. After the plans developed expertise in these activities in

2016, the State transitioned most HCBS roles and responsibilities to the MMPs (see Table 2,

Transition of HCBS waiver functions from State to MMP).

As of December 2017, approximately 1,700 waiver participants were enrolled in the

demonstration, the vast majority of whom were in the Community Choices waiver. At that time,

applications for another 200 demonstration enrollees were being evaluated for HCBS waiver

participation. In fall of 2017, plans and the State noted that although there was not a formal wait

list, the HCBS waiver participation evaluation process was extremely protracted.

See Section 9,

Quality of Care, for a discussion on the issue of plans providing “waiver-like services” for these

enrollees while being reimbursed at the community rate.

Waiver participants may be passively enrolled or opt into Healthy Connections Prime, or

they may become eligible for one of the three HCBS waivers after enrollment in the

demonstration. As part of waiver participation, a CLTC waiver case manager is assigned to

support the long-term care needs of the waiver participant. When a waiver participant is also a

demonstration enrollee, the CLTC case manager works together with the plan’s care manager. In

2016, State and stakeholder interviewees reported that the delineation of responsibilities and

communication between the Healthy Connections Prime plan care managers and the CLTC

waiver case managers had become challenging. Insufficient clarity on the roles resulted in

tensions and inefficiencies. The State addressed this by convening multiple training sessions with

both groups of managers to discuss their roles and responsibilities and developed materials

providing guidance, outlining the roles and intersection of each type of manager (SCDHHS,

In 2018, the State worked to improve wait times by streamlining the process and engaging its outreach vendor to

assist enrollees with the waiver application process. Outreach work is expected to begin in 2019.

2017). The two types of managers are meant to work together, with the CLTC waiver case

manager providing expertise in LTSS, and the plan care manager providing ongoing support to

the enrollee regarding overall health care concerns, initially through in-home assessment and

subsequently with regular check-in interactions. All services authorized by the MMP are

reported to the CLTC case manager and they are input into Phoenix. In late 2017, as a result of

the State’s proactive approach, the concern about respective roles subsided. The State

subsequently surveyed both types of care managers and found that they each recognized and

understood the expertise and role of the other.

Transitional care. Initially Healthy Connections Prime team members worked with State

resources, stakeholders, and plans to develop transition planning policies and procedures to assist

enrollees and providers with enrollee management across settings (e.g., from home to nursing

facility). These include the identification of a key provider contact for clinical coordination

throughout the transition; inclusion of the multidisciplinary team in planning, coordination and

reassessment; coordination with the State’s Money Follows the Person (MFP) demonstration, as

appropriate; and protocol development for medication monitoring. Plan care managers report

transitions to and from hospital, nursing facility, and community settings in Phoenix, and plans

report transition data in their quarterly reports to the CMS implementation contractor (State-

specific reporting requirement S.C. 2.6). However, a CMT member reported in 2018 that it had

come to their attention that care managers are not always aware of hospitalizations. RTI will

cover this topic in future reports.

4.1.3

Care Planning Process

Interdisciplinary Care Teams (ICT)

The interdisciplinary care team, referred to as MTs in South Carolina, initially was

composed of the enrollee, his or her designated representative, the enrollee’s care manager,

primary care physician, and may include other physicians, physician assistants, long-term care

providers, nurses, specialists, pharmacists, behavioral health specialists, and/or social workers

appropriate for the enrollees’ diagnoses, health conditions, and community support needs. Led

by the care manager, the team may assist develop, implement, and periodically review the

enrollee’s ICP, which is designed to meet the enrollee’s medical, behavioral, LTSS, and social

needs. The plans have reported difficulty engaging primary care physicians (PCP) in the MTs

because PCPs are not compensated for participation. When an enrollee’s PCP is unable to

participate in the MT, the plan’s medical director may participate instead. In all scenarios, the

enrollee signs off on their own care plans. The November 2017 contract amendment made PCP

attendance at these meetings optional.

MTs authorize services and these authorizations may not be modified by the plan;

enrollees and their designated representatives may appeal these service determinations. MTs

meet in person or in phone meetings, and they may communicate within the Phoenix system;

they focus on both medical and social needs. All members of the MT are expected to be familiar

with principles of person-centered care, cultural competence, accessibility, independent living

and recovery, and wellness. If MT members are unfamiliar with these topics, plans are required

to provide training (South Carolina three-way contract, 2014, p. 46). The State allows flexibility

in how the MTs operationalize their meetings. By the fall of 2017, plans described these

meetings as “part of the culture” of developing care plans for their enrollees.

The ICP includes language, culture, and service history of each enrollee, and identifies

medical, behavioral, functional, and psychosocial needs of the enrollee. Within 90 days of

enrollment, the ICP is developed as much as possible by the enrollee and his/her family supports,

the care manager, providers, and members of the multidisciplinary team. Measurable short and

long-term services and goals, preferences, and expected outcomes are the key elements of ICPs

that are updated in Phoenix and monitored according to risk level. ICPs of high, medium, and

low risk enrollees are monitored by the plan at 30, 60, and 120 days, respectively, and concerns

are addressed by the MT.

Table 7 presents the total number and percentage of enrollees with a completed ICP for

each quarter of 2015–2017, beginning quarter 2, 2015. Among all enrollees, the percentage of

enrollees with an ICP completed within 90 days of enrollment varied over the course of the

demonstration. Enrollees who the MMPs were able to reach within the first 90 days and who

agreed to complete a care plan ranged from a low of 69 percent when passive enrollment began

(quarter 2, 2016) and increased steadily to 94 percent in the last quarter (quarter 4. 2017) of this

period. (Table 4 in Section 3.2.5, Contacting and Locating Enrollees, displays percentages of

new enrollees who plans were unable to reach within the first 90 days).

Care Coordination at the Plan Level

Plans are required to report to CMS’ implementation contractor (NORC) the ratio of care

managers to enrollees. As shown in Table 8, in 2015 there were 24 full time equivalent care

managers in Healthy Connections Prime plans who were responsible for approximately 90

enrollees each. By 2017 the number of care managers increased to 94, each with an average case

load of 132 enrollees. Plans reported challenges retaining care coordinators due to the high

demand for this type of position in the State and long driving distances. Care manager turnover

was high in the first 2 years (29 percent and 36 percent in 2015 and 2016 respectively), but

decreased to 12 percent by the end of 2017. Each plan organizes its care coordination staff

differently, with some using vendors for initial assessments, particularly during passive

enrollment periods, as well as utilizing support staff to make phone call reminders and conduct

other supportive tasks. Community health navigators are not included in these figures.

Table 7

Members with care plans within 90 days of enrollment

Quarter

Total number of enrollees whose

90th day of enrollment occurred

within the reporting period

Care plan completed within 90 days of enrollment (%)

All enrollees

All enrollees not documented as

unwilling to complete a care plan

or un-reachable

2015

Q1 N/A N/A N/A

Q2 1,506 82.7 89.1

Q3 322 74.8 84.3

Q4 181 86.7 95.7

2016

Q1 230 90.0 97.6

Q2 3,836 51.3 69.1

Q3 3,711 57.3 90.5

Q4 1,331 71.3 92.7

2017

Q1 2,072 65.8 90.6

Q2 190 64.7 85.4

Q3 184 71.2 89.1

Q4 4,226 48.0 94.0

N/A = data are not applicable.

SOURCE: RTI Analysis of MMP reported data for State-specific measure SC 2.1 provided to RTI as of October

2018. The technical specifications for this measure are in the Medicare-Medicaid Capitated Financial Alignment

Model South Carolina-Specific Reporting Requirements document, which is available at:

https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-

Coordination-

Office/FinancialAlignmentInitiative/MMPInformationandGuidance/MMPReportingRequirements.html

Table 8

Care coordination staffing

Calendar year

Total number of care

coordinators (FTE)

Percentage of care

coordinators assigned

to care management

and conducting

assessments

Member load per care

coordinator assigned

to care management

and conducting

assessments

Turnover rate

(%)

2015 24 83.3 90.30 29.2

2016 99 92.9 121.01 35.7

2017 94 92.6 132.06 12.0

FTE= full-time equivalent.

SOURCE: Analysis of MMP reported data for Core Measure 5.1 provided to RTI as of October 2018. The technical

specifications for this measure are in the Medicare-Medicaid Capitated Financial Alignment Model Core Reporting

Requirements document, which is available at:

https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-

and-Medicaid-Coordination/Medicare-Medicaid-Coordination-

Office/FinancialAlignmentInitiative/MMPInformationandGuidance/MMPReportingRequirements.html.

4.2 Information Exchange

In South Carolina, Phoenix (the electronic Medicaid HCBS waiver case management and

service authorization system) was modified to meet the demonstration’s needs, including to

maintain records of all intake, assessment, and care planning activities. Please see Section 2.2.2,

Training and Support of Plans and Providers for a discussion on Phoenix.

Phoenix includes information on home assessments, caregiver supports, and quality

indicators, and has capacity for provider notes, correspondence among users, and other features

(e.g., alerts for follow-up appointments) that can be modified to ensure compliance with Federal

regulations and State policies and programs. MMPs are required to use Phoenix as a uniform

centralized electronic record system to document all activities relating to enrollees’ care (e.g.,

ICPs, provider information, waiver case management, and quality assurance activities). In

principle, any MT member may be trained and access the Phoenix system to read or input notes

on their particular enrollees; however, according to State officials, it appeared that the primary

users of Phoenix within the first 3 years of the demonstration, other than State staff, were the

HCBS providers and plan care managers.

In annual interviews, plans reported that they each have their own extensive member data

management system and that Phoenix was duplicative and not useful to them. Furthermore, plans

made significant and costly efforts to ensure their data systems were conversant with Phoenix to

fulfill their data reporting requirements. During the 2018 discussions with the RTI team

regarding the demonstration extension, CMS noted that some of the plans’ reporting

requirements through Phoenix have been scaled back. This will be discussed in more detail in

future reports.

[This page intentionally left blank.]

5. Beneficiary Experience

5.1 Introduction

Improving the experience of beneficiaries who access Medicare- and Medicaid-covered

services is one of the main goals of the demonstrations under the Financial Alignment Initiative.

Many aspects of Healthy Connections Prime are designed expressly with this goal in mind,

including emphases on working closely with beneficiaries to develop person-centered care plans,

delivering all Medicare and Medicaid services through a single plan, providing access to new

and flexible services, and aligning Medicare and Medicaid processes.

This section highlights findings from various sources that indicate the levels of

beneficiary satisfaction with Healthy Connections Prime overall; it also describes beneficiary

experience with new or expanded Healthy Connections Prime benefits, medical and specialty

services, care coordination services, access to and quality of care, person-centered care, and

quality of life. For beneficiary experience, we draw on findings from the Consumer Assessment

of Health Plan Survey (CAHPS), and RTI focus groups and stakeholder interviews. Please see

Section 1.1.3, Data Sources and Methods for details about each data source. South Carolina

focus group findings in 2016 and 2017 were very similar. Therefore, unless otherwise noted, the

2016 and 2017 focus group findings are reported together. This section also provides information

on beneficiary protections, data related to complaints and appeals, and critical incident and abuse

reports. The section includes information, where available, on the experience of special

populations.

Highlights

• CAHPS survey results and RTI focus group findings indicate that the majority of

enrollees are satisfied with services provided under Healthy Connections Prime.

• The majority of focus group participants felt they were part of their care planning

team, and appreciated their care managers’ regular check-ins, assistance in accessing

needed services, and support in resolving issues such as improper billing.

• The majority of focus group participants were aware of, and many had used, benefits

such as plan supplemental benefits (e.g., gym memberships), plan flexible or ‘waiver

like’ service benefits (e.g., home modifications), and community-based organization

benefits (e.g., Meals on Wheels).

• Availability of, and access to, certain services continue to be a challenge for many

Healthy Connections Prime enrollees in this very rural state. Focus group participants

and site visit interviewees mentioned challenges with access to specialty providers,

including behavioral health providers, and the availability of aging-specific services,

(e.g., vision, dental, hearing, and dementia care such as day centers).

5.2 Impact of the Demonstration on Beneficiaries

This section summarizes the findings of focus groups, beneficiary surveys, and

stakeholder interviews reflecting beneficiary experiences with service delivery and quality of life

under Healthy Connections Prime. Beneficiary experiences related to the early enrollment

process, including experiences of beneficiaries who chose to opt in, opt out, or who were

passively enrolled, are discussed as part of Section 3, Eligibility and Enrollment.

5.2.1

Overall Satisfaction with Healthy Connections Prime

CAHPS results for 2017 and RTI focus group findings for 2016 and 2017 indicate that

the majority of enrollees are satisfied with Healthy Connections Prime. Focus group participants

rated their satisfaction very high and were especially pleased with the elimination of copayment

costs.

Table 9 presents 2017 CAHPS data for three beneficiary satisfaction measures across

MMPs participating in the demonstration in South Carolina: (1) the percent of beneficiaries that

rated the health plan as a 9 or 10 (on a scale of 0 to 10, 10 being the best rating), (2) the percent

that rated the drug plan as a 9 or 10, and (3) the percent that reported the health plan “usually” or

“always” gave them the information they needed. We provide Medicare Advantage (MA)

benchmarks, where available, understanding that MA and demonstration enrollees may have

different health and sociographic characteristics, and these differences could affect the results.

Among the three MMPs, beneficiaries’ satisfaction with their health plan ranged from 54 to 66

percent. Although the ratings for two of the MMPs are comparable to the national percentages

for MA and MMP contracts, beneficiaries enrolled in another MMP reported a lower rating of

their health plans (54 percent). In contrast, a higher percentage of respondents across all three

MMPs reported higher satisfaction with their drug plans, with the percentage rating their drug

plan a 9 or 10 ranging from 65 to 68 percent. Likewise, consistent with national ratings, a high

percentage of respondents across all three MMPs reported “usually” or “always” receiving

information they needed from their health plan.

Table 9

Beneficiary overall satisfaction, 2017

CAHPS survey

item

National

distribution—all

MA contracts

National

distribution—all

MMP contracts

Absolute Total

Care (ATC)

Molina

Healthcare of

South Carolina

Select

Healthcare of

South Carolina

Percent rating

health plan 9 or

10 on scale of 0

(worst) to 10

(best)

(n=188,484)

(n=14,662)

(n= 454)

(n=318)

(n=350)

Percent rating

drug plan 9 or 10

on scale of 0

(worst) to 10

(best)

(n=172,033)

(n=14,087)

(n=410)

(n=315)

(n=319)

Percent reporting

that health plan

“usually” or

“always” gave

them information

they needed

(n=168,247)

(n=13,879)

(n=452)

(n=298)

(n=344)

MA = Medicare Advantage; MMP = Medicare-Medicaid Plan.

SOURCE: CAHPS data for 2017

RTI focus group participants expressed overall satisfaction and characterized services

under Healthy Connections Prime as better than those previously received. Participants said they

were grateful to be in the demonstration because they no longer had copayments for provider

visits or medications. In 2017, when asked to rate the plan on a scale of 0 to 10 with 10 being the

best, almost all of the focus group participants ranked their plan between eight and 10. Examples

from beneficiaries expressing satisfaction with their Healthy Connections Prime plan in 2017

follow:

It’s [combined Medicare and Medicaid] just better. You understand it better. You

don’t have to have two cards. And then when you go to the drugstore they say,

‘This one pays for that. That pays for that. And then this one don’t pay for that.’

And then it ain’t like that with your one card.

The most I like about the plan is they’re nice people, and you don’t have [a]

copayment, and you can call from 8:00 in the morning until 8:00 at night if you’ve

got a problem. I don’t think there’s anything about them that I don’t like.

Without [health plan], it’s been almost impossible for me to get [services] because

I don’t have the money. A lot of things you go to, right away you ain’t got no

insurance, ain’t nothing happening. [Now] You feel like they really care, that they

want to … make you feel better and do a job right.

Some focus group participants expressed satisfaction more specifically with the benefits

that they could now access with the demonstration. One said: “[I] used to pay money and didn’t

know whether I was going to eat that month. But now I don’t have to worry about medication. I

can go to the doctor, where I used to have to wait until I [got] the money.”

Participants provided recommendations for improvements to benefits as well. In

particular, participants said that benefits should have been expanded for more dental, vision, and

transportation services. For example, South Carolina’s Medicaid insurance covers the costs of

tooth extractions but does not cover dentures, a need of many older adults; none of the MMPs

added supplemental dental benefits. Similarly, eye exams are covered by South Carolina’s

Medicaid insurance but only two of the three plans added supplemental eyewear benefits. One

2017 focus group participant summed it up by saying, “What I least like about it is the dental and

eye care. They don’t add any of those in. Even hearing aid ... I wish that was a part of it, too.

Because you can’t see, you can’t hear, you can’t eat, you’re dead.”

5.2.2

New or Expanded Benefits

The three plans use flexible benefits or “waiver-like services” as they are known in South

Carolina, differently. Two plans noted that they use the benefits strategically to enable enrollees

to remain in their homes rather than in nursing facilities. Although the total monetary value of

these benefits is not available, according to MMP self-reported data counts, 164 ATC enrollees,

31 Select enrollees, and five Molina enrollees have received some type of “waiver-like service”

or flexible benefit from their plans during calendar year 2017.

Benefits that most 2017 focus group participants knew were available to them included

plan supplemental benefits (e.g., extended vision services), community-based organization

benefits (e.g., Meals on Wheels), respite services (for HCBS recipients), and flexible benefits

(e.g., home modifications, etc.). Participants said they were pleased with extra benefits provided

by some plans; some noted that gym membership, a plan supplemental benefit, was useful to

maintain a healthy weight. Participants also reported receiving gift cards for attending

informational sessions at the plan or for receiving mammograms or colonoscopies. One

participant explained a helpful monthly over-the-counter drugstore supply program offered by

one plan:

But they also have another plan where you can get $25 worth of over-the-counter

medicines ... they give you a list, and it shows you the pricing. You can get things

like multivitamins, aspirin, skin lotion, stuff like that.

State officials indicated that approximately half of enrollees who would be appropriate

for the new palliative care benefit were receiving those services. Reporting and tracking this

service was challenging, and State and plan officials noted that there are few providers in the

State who are trained in providing this benefit. In order to improve awareness and access to this

benefit, the State has collaborated with the Center for Palliative Care to develop enrollee and

physician educational materials to improve messaging regarding the difference between the

palliative benefit and other benefits, such as hospice. See Section 1.2, Model Description and

Goals for information on this benefit.

Likewise, another participant expressed appreciation for flexible or “waiver-like” benefits

that enabled her to remain in her home. She explained that if she needed additional home

improvements or modifications, those services would be available to her: “I get pest control

[from the plan]. If I needed any modifications like a wheelchair ramp, they would build a

wheelchair ramp for me.”

5.2.3

Medical and Specialty Services

The majority of focus group participants reported having a PCP and had been seeing their

current PCPs for at least a year, with some indicating they had the same PCP for more than 5

years (i.e., including prior to enrollment in the demonstration). The evaluation was particularly

interested in learning about beneficiary satisfaction with their providers due to the age of the

demonstration’s population (age 65 years or older). Some focus group participants were initially

concerned about possibly changing to a new PCP if she or he was not in the MMP network.

However, the MMPs’ single case contracting and the demonstration’s continuity of care

provision improved plans’ ability to contract with providers to serve demonstration enrollees.

One participant relayed how her PCP began participating, “When I started, they told me that my

PCP was not on their plan. But they would see if they could get him on the plan. They were able

to, so I was able to stay with him.”

When focus group participants were asked about their satisfaction with their current PCPs

and specialists, mixed feedback was reported by participants. Some participants reported positive

relationships with their current providers. Many of these participants felt that their doctors were

attentive and provided needed services.

I adore my provider [and] [plan name]. I think they’re amazing. I think they’re

totally wonderful. I am so impressed. I told them I would do commercials for

them.

I got one particular doctor … He’s my cancer doctor. I’ve been seeing him for the

last 7 years. He’s very strict. He’[s] the type of guy that if you got cancer, he’s

going to fight it for you. He’s there. The last 7 years, he’s been there for me.

However, a few participants felt that their doctors did not listen to them or they felt

rushed during visits. One participant explained,

[Doctor] has too many patients and when he comes in to see you, he only sees you

maybe about 5 minutes and he’s gone. And he drew my blood and then he

checked my heart and that’s it. [The care manager] kept telling him to give me a

yearly physical and he didn’t do it.

Beneficiaries, the State, and stakeholders raised several medical and specialty service

access issues. One stakeholder interviewed in fall 2017 pointed out that enrollees were facing

challenges accessing specialty services, such as podiatry. Several focus group participants also

described difficulty finding a PCP or securing appointments with various specialists. One

participant said, “And my experience is that when you go through the list [of available

physicians]—they’re not currently taking patients. So it does me no good.”

Likewise, one participant found it difficult to find a specialist. “I like that everything is

combined and that makes it simple for me. But I’m having a hard time finding the specialists that

are on that plan, that accept that plan. Ear, nose, and throat doctor, especially I’m having a real

hard time finding one.”

Another participant described her difficulty in accessing specialty services due to what

appears to be a lack of education on the part of either the referral provider or the specialist, “So

Dr. [NAME], she’s a gynecologist. She sent me to a specialist. So the specialist told me I needed

to have $125, and I told her I didn’t have it with me. So she told me when I get the $125, I could

come back.”

In 2015 and 2016, MMPs reported that nursing facility staff continued to advise enrollees

to disenroll from the demonstration in order to transition to their facilities. Despite State and

CMS education of nursing facility staff about the demonstration, in 2017, MMPs reported

continued provider resistance in this area (see Section 6, Stakeholder Engagement). Other

access issues identified early in the demonstration were mostly resolved by 2017 through the use

of single case contracting and continuity of care provisions.

5.2.4

Care Coordination Services

Most focus group participants in 2016, and nearly all focus group participants in 2017,

knew their care coordinators, known as care managers in South Carolina, and described regular

interactions with them. Participants appreciated that there was someone from the health plan who

they could personally contact and who knew their healthcare details and current life

circumstances. Participants in both years reported positive relationships and were pleased with

their care manager’s attentiveness and concern for their well-being.

She [the care manager] called me because, see, they try to keep up with your

circumstances, because what they do is they do home visits, too. Send someone

out to check your brain status about every two months.

She came to my house and sat on the front porch and talked about whatever we

want to talk about. And it’s basically your healthcare and they get an inventory of

how well you’re doing and what you’re doing and who takes care of you and do

you still bathe by yourself. That kind of stuff.

They came to my house once and they told me that if I felt that I needed

somebody to come in and help me with house cleaning or anything like this or if I

needed transportation to the doctors.

Although the majority of focus group participants reported having consistent care

managers, in 2017, a few said they received calls from multiple people at the plans and that their

care managers had changed.

… the caseworker changes every month, and nobody seems to know what’s going

on there. You can call and ask a question and it’s like, 30 minutes on the phone

trying to find it because nobody knows nothing. They come out to your house to

see if you’ve got what you need and stuff.

Many participants noted that their care manager will assist with administrative issues

such as incorrect billing or filing a complaint. One explained: “You know what I like about our

plan? You have a nurse that will call you once a month …. She will talk to you and let you know

that whatever the hospital or doctor charged, don’t worry about it. They got it covered.”

The importance of a face-to-face interaction was also noted by providers who RTI

interviewed. In 2017 an assisted living facility provider reported that it took some time for

residents to understand the purpose of the Healthy Connections Prime care manager: “When the

care manager asks for the resident they [resident] give us [staff] the phone back or hang up and

say, ‘I don’t know why they [are] contacting me.’” However, the provider explained that after

the care manager visited the enrollee, the enrollee was pleased with the care manager.

Demonstration enrollees also provided feedback on their providers’ communication

through the CAHPS survey. Most survey respondents in ATC and Select reported that their

doctor usually or always was informed and up-to-date about the care they received from

specialists (see Table 10). In 2017, 86 (ATC) and 87 percent (Select) of respondents reported

this rating which is comparable to national MA and MMP ratings. We provide MA benchmarks,

where available, understanding that MA and demonstration enrollees may have different health

and sociographic characteristics, and these differences could affect the results.

Table 10

Care coordination, 2017

CAHPS

survey item

National

distribution—

all MA

contracts

National

distribution—

all MMP

contracts

distribution—

MMP

contracts

Absolute

Total Care

Molina

Healthcare of

South

Carolina

Select

Healthcare of

South

Carolina

Percent

reporting that

in the past 6

months

personal doctor

“usually” or

“always” was

informed and

up-to-date

about care

received from

specialists

(n=103,052)

(n=6,942)

(n=459)

(n=180)

(n=150)

NA = not available due to too few respondents to this question; MA = Medicare Advantage; MMP = Medicare-

Medicaid Plan.

SOURCE: CAHPS data for 2017

5.2.5

Beneficiary Access to Care and Quality of Services

RTI focus groups and site visit interviews in 2017 indicate that Healthy Connections

Prime enrollees continue to encounter challenges in the areas of (1) access to providers,

(2) availability of aging-specific services, including vision, dental, hearing, and dementia care,

and (3) access to behavioral health services. Participants described having access issues due to

transportation services:

I don’t have a car, so I have to work with [transportation vendor]. And it’s just

impossible to find someone who does understand that you are not well, which is

why you’re going to the doctor. I don’t know what they have as a guideline for

the contractors. It is awful. You have to plan. If you’re sick on Monday, you can’t

go on Monday unless you call an ambulance. You have to wait 3 days.

Beneficiaries also noted major difficulties in accessing aging-specific services. In both

the 2016 and 2017 focus groups, participants reported receiving denials for needed items or

services such as incontinence supplies, nebulizers, and canes. Participants also expressed

frustration with the lack of benefits for hearing aids, as one participant noted “[the health plan]

don’t cover hearing, they don’t cover eyes, don’t cover dental, doesn’t pay for dentures.”

Another focus group participant described difficulties accessing dental providers: “I am not

happy with [my plan] … the lack of providers. I needed to have some dental work, and they sent

me to a children’s dentist.” Day centers for persons with dementia are available in some urban

centers, but few services exist in rural areas. However, one focus group participant expressed

appreciation for the day center service available for their parent: “My mom was able to go [to]

daycare. She gets to interact with other people. Be with people her age.”

Although the demonstration was intended to integrate behavioral health services with

primary care and LTSS, State and stakeholders interviewees noted that little has changed due to

the scarcity of behavioral health providers and the focus of the State’s behavioral health services

on the younger population. State demonstration staff noted that although 15 percent of

demonstration enrollees have a behavioral health diagnosis, they believed there are more with

undiagnosed conditions. State staff have consistently indicated that access to behavioral health

care for older adults is a concern because a gap continues to exist between the services available

and the behavioral health needs of older adults. Reasons for this gap include the Department of

Mental Health’s (DMH’s) reported reluctance to treat beneficiaries who have a diagnosis of

dementia; and the fact that providers focused on disruptive behavior and under-treated conditions

such as depression. By fall 2017, the State and the plans continued to report low overall

utilization of mental health and behavioral health services. When behavioral health specialists

are not available, enrollees receive services through their PCP. Focus group participants spoke of

their experience with behavioral health services; one participant noted the lack of focus on

mental health services for older adults.

[Mental health services] helps whenever you have an illness, if you get agitated or

you get down and out. It’s just an amazing thing. And a lot of the mental health

people do not—in the mental health clinic, there are a lot of young people and

they don’t want to help you as you get older. But doctors should be able to refer

you or give you somebody that should be able to help you if you need it. I’m on

two medications, and I couldn’t do without them.

Although beneficiaries described challenges they still face in accessing care, aging-

specific services, and behavioral health services, several focus group participants expressed that

they had improved access to services such as medical care, durable medical equipment, and

transportation after enrollment in the demonstration.

Well, say for instance a colonoscopy, I had that last week. I’ve never had one. My

doctor pushed me into getting one. He said, “I want you to live for me. Give it a

try.” And I did. And I did have something wrong.

I had to get a scooter so that I could just go up and get my mail. [My new plan]

took over because I had been fighting [with my old health plan] for a year. They

took over and I had a scooter in my front yard in two months.

5.2.6

Person-centered Care and Patient Engagement

Focus group participants indicated high involvement in their own care. Most participants

reported that their PCPs and specialists listened to them and that they felt that they were part of a

team. One said, “I like most [that] I have a doctor that is including me in her team. I feel like

they’re [specialist and PCP] working together. So yes … they’re communicating about me.”

Only a few participants felt that they were not part of the decision-making process due to

a lack of communication regarding their own care or their loved one’s care. In 2017, for

example, one participant noted that the physician did not share information about her health

when she asked.

He did not listen to me; he patronized me. I am not a genius, but I am an

intelligent, informed individual. I can understand my care. If I can’t, I want to be

able to ask questions. But if I would ask him a question, he would say, “You don’t

need to know that.”

5.2.7

Personal Health Outcomes and Quality of Life

Most focus group participants in 2016 and 2017 said they had improved quality of life

and better health since enrolling in the demonstration. A few participants said their quality of life

had not changed and, noting their serious health conditions, such as a cancer diagnosis, they said

their health had not worsened. None of the participants said that their quality of life or health was

worse after enrolling in the demonstration.

Participants spoke about how their quality of life had changed after receiving services

such as pain management, supplemental services, and care coordination through one of the

Healthy Connections Prime plans.

For a while I didn’t [enjoy life] because I was in so much pain. But now that’s

why I can laugh and smile, because when I think about the accidents that turned

my whole life around, I had to adjust to it. I’ve got my life back…. I didn’t think I

could get it back to this full extent.

Her [mother’s] quality of life is better. The meals help her nutritionally-wise.

Keeps her up. Things like that, that she wasn’t getting that she gets now. They all

work together for the one common good, and that’s the personal health and well-

being of my mother.

Sometimes when people care about you, it makes your life a little better. When

they call me up and ask me do I need [services]—you see somebody out there

concerned, that you’re not left in the dark. You get someone who wanted to help

you. That made you feel a little more uplifted.

Some participants said they achieved a better quality of life by having consistent access

to health care or benefits provided by their plans.

I could take my medicine properly, because I used to cheat myself. Because it’s

free now … Before when I was working and I was taking medicine, I was taking

half because I couldn’t afford it.

In both 2016 and 2017, most participants emphasized that the demonstrations’ zero

copayments and no-cost medications reduced the stress in their lives because they did not have to

worry about out-of-pocket costs for their health care.

Because before [the demonstration] if you go [to the doctor] and … you got a

prescription [that costs] $200 … [where is] that $200 coming from? That could

put stress on you. Now if you get a prescription, I don’t care how much the

prescription costs. I just go into the pharmacy and get my medicine.

5.2.8

Experience of Special Populations

This section summarizes the beneficiary experience for Healthy Connections Prime

special populations, including individuals with LTSS or behavioral health needs, and

racial/ethnic or linguistic minorities. Because the Healthy Connections Prime demonstration

supports enrollees age 65 and older, many of whom have LTSS needs, the entire demonstration

population could be considered a “special population.” Thus, many findings already discussed in

this section showcase the demonstration experience of special populations. Please refer to

Table B-1 in Appendix B for a breakdown of enrollee demographic characteristics.

According to stakeholders, State, and MMP officials, there is not a large population of

linguistic minorities among Healthy Connections Prime enrollees. However, to support

communication with enrollees with special linguistic needs, plans and providers have written

materials and language call-in line services available in Spanish and other languages as needed.

All focus group participants were asked if they had experienced any improper or

discriminatory behavior by the State, MMPs, or providers. In 2017, two participants reported

feeling uncomfortable because of what they viewed as racist behavior by White providers. One

reported his specialist seemed reluctant to physically examine him. Another participant changed

providers as a result of her interactions. She explained:

They [physicians] are nice peoples. Well, they talk nice to you. Some peoples talk

mean to you, like that [previous specialist], she called me a girl all the time. She

come in the office, ‘How you doing, girl? How you feeling today, girl?’ And so I

didn’t like that because I’m 79 years old. How old do you be before you become a

woman? It sounds like a racist thing because I think that’s what they used to call

the women in slavery. They was women, but they still called them girl, and they

called the men boys.

Enrollees living in rural locations face service access challenges. Interviews and focus

group findings indicate that plan care managers and health navigators have been instrumental in

meeting the needs of beneficiaries, some of whom had not received services in several years

prior to enrolling in the demonstration. Focus group participants from rural communities

expressed satisfaction with in-home visits and assistance provided by care managers. To improve

access to care for enrollees in rural areas, Healthy Connections Prime staff and MMPs discussed

exploring new ways to reach these enrollees such as telehealth services through the South

Carolina Telehealth Alliance, which supports services such as remote patient monitoring and

video visits. South Carolina’s DMH has a tele-psychiatry program that comprises emergency

department consultations and community mental health services (South Carolina Department of

Mental Health [SCDMH], 2017). As pointed out by plans and the State and discussed previously

in this report, there continue to be complex challenges to expanding and integrating behavioral

health services to seniors in South Carolina.

5.2.9

Beneficiary Protections

This section describes the beneficiary protections available to demonstration enrollees

and enrollees’ awareness and use of those protections. It also includes a summary of grievance

(complaint) and appeals data received from (1) data reported by MMPs on complaints made

directly to them; (2) data reported on the Complaint Tracking Module (CTM) for complaints

received by SCDHHS and 1-800-Medicare; and (3) data reported by the Independent Review

Entity (IRE), which is a second-level review of appeals. Reporting periods vary across these

sources.

Complaint process and enrollee awareness of rights. MMPs are responsible for

addressing complaints that arise in their plans, and they must respond to complaints within

30 days. Enrollees are encouraged to address their concerns at the plan level first. However, they

may file complaints (grievances) through 1-800-Medicare, with SCDHHS, or seek assistance

from the Healthy Connections Prime Advocate to file a complaint (described below). All

complaints are forwarded to the contract management team (CMT) and entered into the CMS

CTM. The MMPs are required to maintain records of all appeals activities and notify CMS and

SCDHHS of all appeals (South Carolina three-way contract, 2014, pp. 128–30).

Enrollees may contact the ombudsman program, known in South Carolina as the Prime

Advocate, at any time for assistance in making a complaint or filing an appeal. SCDHHS

contracted with the State’s LTC Ombudsman program to serve as the Healthy Connection Prime

Advocate (HCPA) program in March 2015. The State LTC Ombudsman reports to the South

Carolina Lieutenant Governor's Office on Aging, has a statewide network of 17 regional

ombudsmen, operates a toll-free telephone number with multiple language lines, and has

reporting systems in place to capture data on grievances and appeals for individuals receiving

long-term care services. This office also provides training and outreach to educate beneficiaries,

stakeholders, and the community about Medicare and Medicaid rules and enrollees’ rights.

Under the contract with SCDHHS, these services were expanded for HCPA, and two

Healthy Connections Prime Advocate positions were created to support demonstration enrollees.

Staff also provided education programs to communities and stakeholders to promote the

demonstration and they assist beneficiaries who are not yet enrolled with administrative issues

(e.g., Medicaid renewal). Together with the network of regional ombudsmen, the Healthy

Connections Prime Advocates promote demonstration enrollee self-advocacy, they engage

MMPs, or providers directly to resolve issues, and assist enrollees in filing complaints,

grievances, and/or appeals.

Focus group participants reported that they resolved issues at the plan level before they

were escalated to the formal grievance level. They usually discussed the issue with their care

manager or called the plan directly to resolve an issue. As one focus group participant explained:

“I sometimes get bills, but I’ll call the [Plan] and ask them why I’m getting it, and they call and

resolve it.” In both 2016 and 2017 few focus group participants had heard of the Prime Advocate

or knew there was a resource available if they had issues that were not resolved after contacting

their care manager or someone at the plan.

Complaint data. Plans track their grievances and appeals and report them in their

monthly reports to the CMT. In 2017, the top three reasons for grievances reported by MMPs

were inappropriate billing, poor customer service, and poor network access. When grievances are

referred to the Prime Advocate, staff investigate and work to resolve complaints made by or on

behalf of demonstration enrollees. They also document all cases and track grievance data to

identify trends and systemic problems. The Healthy Connections Prime Advocates participate in

CMT meetings and report all grievance and complaint data to the CMT. In RTI interviews each

year, Prime Advocate representatives reported that they were able to resolve cases that were

referred to them within a short time. They explained that it usually required a week to discuss the

issue with the pharmacy, hospital, plan, or other relevant entity, to hear the circumstances, and to

reach a resolution. Education for both the entity and the enrollee was a large part of their activity.

Nursing facilities statewide have high occupancy rates (approximately 95 percent); this is a

source of enrollee complaints at times when they were offered services at a nursing facility too

far from their home. The Prime Advocate staff work with the LTC Ombudsman staff to resolve

these more complex issues. At the Steering Committee meeting of February 2017, MMPs listed

the Prime Advocate as the top asset of the Healthy Connections Prime demonstration.

Overall, there is no consistent trend in the number of reported grievances per 1,000

enrollees through the course of the demonstration. The number of grievances per 1,000 enrollees

increased in the first 8 months of the demonstration, to a high of 21.5 complaints per 1,000

enrollees in quarter 3, 2015. This was followed by a general decrease to a low of 7.9 per 1,000 in

quarter 1, 2017. Subsequent quarters saw increases, to the third highest number of grievances per

1,000 enrollees of 15.7 in quarter 4, 2017. From February 2015 through December 2017, the

categories with the three highest number of complaints were: (1) benefits, access, and quality of

care; (2) enrollment/disenrollment; and (3) HIPAA and security concerns, customer service, and

availability of materials in alternative formats.

Appeals. MMPs are required to follow appeals processes that incorporate relevant

Medicare Advantage and Medicaid managed care requirements. Plans must respond to enrollee

appeals within 15 days. They must maintain records of all appeal activities and notify CMS and

SCDHHS of all internal appeals. Enrollees are required to be notified of all Medicare and

Medicaid appeal rights through a single notice (South Carolina three-way contract, 2014,

pp. 128–30).

Similar to the complaint data, over the course of the demonstration there is an

inconsistent trend in the number of plan reported appeals per 1,000 enrollees. The highest

number of appeals per 1,000 enrollees occurred in quarter 1, 2017, with 23.5 plan reported

appeals. Of the plan reported appeals, in each quarter, a majority (ranging from 62.2 percent to

100 percent) resulted in fully favorable outcomes for the enrollee. The number of appeals

reported to the IRE, a second-level review of appeals, increased over the 3 calendar years of the

demonstration, with two appeals reported in 2015, 32 in 2016, and 70 in 2017. In each year, the

majority of appeals was upheld, with less than 20 percent overturned in favor of beneficiaries.

The most common category of appeals referred to the IRE was for practitioner services.

Critical incidents and abuse. South Carolina requires MMPs to report critical incidents

and abuse of members receiving LTSS (CMS, 2018c). In 2015, the number of critical incidents

and abuse per 1,000 members ranged from 13.0 in quarter 2 to 17.7 in quarter 4. In 2016, the

number per 1,000 members receiving LTSS dropped to 4.6 in quarter 1 and 0 by quarter 4. In

2017, the number of reports of critical incidents and abuse per 1,000 members receiving LTSS

was 0.6 in quarter 1, 0 in quarter 2, 0.5 in quarter 3, and 0 in quarter 4.

Examples of practitioner services include, physician, chiropractic, dental, prosthetics/orthotics, and vision care.

[This page intentionally left blank.]

6. Stakeholder Engagement

6.1 Overview

This section describes the approach taken by South Carolina for engaging stakeholders,

the mechanisms for soliciting stakeholder feedback, and the impact of those efforts on the

demonstration in more detail.

The State has engaged stakeholders through a variety of activities and adapted its focus as

the demonstration has evolved. Early in the planning phase of Healthy Connections Prime, the

State conducted general presentations about the goals of the demonstration throughout the State

to provider associations, medical groups, and advocacy organizations to solicit feedback and

inform the design of the demonstration. In 2013, the State invited a broad range of providers and

advocates to participate in planning work groups, to update stakeholders about the progress of

the demonstration, solicit more feedback on design changes, educate the community, and

encourage participation in the demonstration.

After the three-way contracts were signed with the MMPs in fall 2014, the State’s focus

shifted to educating MMPs and stakeholders (e.g., HCBS providers, hospitals, nursing facilities,

etc.) and conducting targeted outreach to encourage more physicians and provider groups to

participate in the demonstration. At that time, physicians were reluctant to contract with MMPs

due to a general fear of managed care, a competitive provider environment in the State, and low

enrollment in the demonstration.

Highlights

• South Carolina’s Healthy Connections Prime staff cast a wide net to engage

stakeholders beginning in 2013. After the demonstration began in 2015, it adjusted its

focus to attract and engage more providers and to train MMPs.

• The demonstration has strategically used contractors, including the Prime Advocate,

to engage and educate stakeholders through a variety of venues and formats. Clearly

written and timely postings of policy changes, marketing materials with consistent

language, and provider website enhancements have enhanced stakeholder

engagement.

• Training has been an important aspect of South Carolina’s outreach to stakeholders. In

addition to initial demonstration training, the Office for the Study of Aging at the

University of South Carolina (OSA) has been an integral partner in using training as a

mechanism to resolve administrative modifications for stakeholder groups.

• As the demonstration has evolved, the State has engaged and redefined stakeholder

advisory groups as needed. By 2017, due to the uncertain future of the demonstration,

the State had narrowed its focus from a broad variety of stakeholders to engaging

partner organizations.

The State continued its relationship with the larger, stakeholder and advocate community

by electronically distributing salient demonstration news via the Prime Weekly Updates in 2014.

Subsequently, these transformed into monthly and then quarterly Stakeholder Updates. When the

demonstration began in early 2015, but was limited to opt-in enrollment, the State focused its

outreach to trainings of MMPs and their providers in processes that were new to them including

contracting with LTSS providers, HCBS service delivery, Dementia Dialogues, care giver

coaching, transition planning, and the use of Phoenix, the State’s case management system,

among other processes.

6.2 Organization and Support

6.2.1

Stakeholder Council

In May 2016, after passive enrollment began, the demonstration gathered momentum and

the State convened an Implementation Council (IC), comprised mainly of stakeholders who had

participated in the earlier planning workgroups. State contractors, demonstration stakeholders,

and MMPs also participated, and enrollees and caregivers were invited to participate. The State

had planned for the IC to continue, tasked with providing input on policies, assisting in

monitoring, and advising SCDHHS on the demonstration. However, in late 2016, the Medicaid

Director announced a suspension of passive enrollment and a discussion of the possibility of

transferring the demonstration population to a MLTSS model. The Healthy Connections Prime

team delayed reconvening the IC until there was more certainty about the future of the

demonstration beyond 2018. Consequently, during most of 2017, the State focused attention on

MMPs, rather than the broader stakeholder community. In addition to continued training, it held

two Steering Committee meetings in 2017 during which CMS, the State, and MMPs discussed

demonstration challenges and potential areas of improvement.

Targeted outreach. From the beginning of the demonstration, as the State identified

stakeholder concerns, it leveraged its contractors to conduct targeted outreach and training

focused on the needs of each stakeholder group (e.g., hospitals, nursing facilities, physicians, and

HCBS providers). The Office for the Study of Aging (OSA) at the University of South Carolina,

the State’s training contractor, planned to conduct a series of learning collaboratives in 2015 and

2016. Each collaborative training would provide an opportunity for the three MMPs and each

separate stakeholder group to discuss policy concerns and to clarify administrative and

procedural issues unique to the particular group (e.g., nursing facility referrals, transfers,

payment, etc. for the nursing facility learning collaborative). However, only a few learning

collaboratives occurred prior to the end of the State’s training contract in July 2017. Two were

well attended by HCBS providers and one with nursing facility providers; MMP representatives

attended all collaboratives.

The training conducted by OSA with nursing facility (NF) staff was most important

because the State and stakeholders reported that many NFs were somewhat hostile to the

demonstration, believing it would reduce the numbers of beneficiaries using their services. NF

staff reportedly provided incorrect information to enrollees including advising them to disenroll

in order to use skilled nursing facility or long-stay services. Despite their concerns, NF staff

participated in demonstration training conducted by OSA, which also made archived web-based

recordings available for repeated trainings for new NF staff. Throughout the demonstration, the

State and CMS provided nursing facilities with updated policy and procedure documents which

were also posted on the demonstration’s website and distributed via the demonstration’s

quarterly updates (SCDHHS, 2018d).

Another stakeholder group, HCBS providers, also benefited from targeted engagement by

the State and its contractors. This group had been contracted and paid by the State for many

years and according to State officials, HCBS providers were extremely hesitant to become

involved with managed care organizations, including MMPs. In addition to the learning

communities conducted by OSA, the State addressed HCBS’ primary concern by allowing their

billing to continue through Phoenix, the State’s case management system, although they were

contracted by the MMPs. This provision added an administrative burden to the State, but allayed

the fears of these small organizations and individuals who were concerned about developing new

billing mechanisms that might lead to payment delays. The State subsequently recoups these

funds from the MMPs.

In addition to trainings by OSA, SC Thrive, a beneficiary outreach organization

contracted by the State, conducted beneficiary outreach throughout the State, primarily to

educate beneficiaries about the demonstration and to conduct individual options counseling. In

many of the outreach venues, such as Senior Fairs, SC Thrive also provided education on the

demonstration to provider and advocacy groups. In 2017, MMP representatives joined SC Thrive

representatives in these outreach events.

The demonstration’s ombudsman, known in South Carolina, as the Prime Advocate (see

Section 5.2.9, Beneficiary Protections), is contracted by the State to provide stakeholder

outreach for the demonstration in addition to providing ombudsman services. The advocate’s

office attends outreach events for both beneficiary groups and providers with State health

insurance program (SHIP) and the Lieutenant Governor’s Office of Aging staff, and has

appeared on television panels and other events to promote and engage stakeholders.

To improve messaging to both physicians and enrollees regarding the new palliative care

benefit, in 2017 the State collaborated with the Center to Advance Palliative Care to develop

educational materials explaining the difference between the palliative benefit and other benefits,

such as hospice (see Section 1.2, Model Description and Goals).

To ensure that all demonstration materials provided to stakeholders had standardized

language and messaging, the State contracted Ikaso Consulting which designed user-friendly

quarterly updates, frequently asked questions (FAQs) and other materials targeted to

stakeholders, providers, and MMPs. In early 2017, Ikaso restructured the demonstration’s

website to provide resources and materials for stakeholders and a new section for providers

(SCDHHS, n.d.).

Enrollee advisory committees. In addition to seeking input from various stakeholders in

the planning and implementation of the demonstration, the three-way contract requires each

MMP to establish an enrollee advisory committee to obtain meaningful beneficiary input on

issues. The MMPs engaged enrollee advisory committees and have met quarterly throughout the

demonstration to hear their concerns about the demonstration and to solicit feedback on policy

refinements. In 2016 interviews, enrollee committee participants of all three plans reported a

positive experience, a wide range of topics that had been covered in committee meetings, and an

appreciation for having their voices heard. At a 2017 planning meeting with CMS and the State,

the MMPs listed the advisory committees as one of the three top successes of the demonstration

to date.

7. Financing and Payment

7.1 Rate Methodology

All Medicare and Medicaid-covered services are financed by capitated payments to the

MMPs; the Medicare and Medicaid contributions represent baseline spending, or the estimated

costs if the demonstration had not been implemented. Capitation payments are risk-adjusted,

using separate methodologies for Medicare Parts A and B services, Medicare Part D, and the

Medicaid components of the rate. The savings percentage is applied to baseline spending. This

section describes the rate methodology of the demonstration and findings relevant to early

implementation.

7.1.1

Rating Categories and Risk Adjustments

The Medicare baseline rates for Parts A and B services consist of a blend of Medicare

Advantage projected payments and Medicare standardized FFS rate for the county in which the

enrollee resides, weighted by the proportion of the target population that is expected to transition

from each program into Healthy Connections Prime. The Medicare Advantage portion is a

function of costs that would have occurred had the demonstration not taken effect, including

quality bonus payments for the relevant Medicare Advantage plans. The FFS county rates are

Highlights

• MMP officials reported financial losses as a result of lower than expected enrollment

due to passive enrollment delays and pauses, and unanticipated expenses. MMPs have

responded by limiting their flexible benefits and participating in efforts to increase

enrollment. In 2017, two of the three plans indicated that they would not be able to

continue in the demonstration past 2018 due to losses.

• The SCDHHS eligibility tracking system has struggled to efficiently identify enrollees

in community versus nursing facility settings. Errors in nursing facility systems have

led to charges to both the State and MMPs for the same enrollees, resulting in delayed

and inaccurate MMP payments, further contributing to MMPs’ financial struggles.

• A 13 percent decline in the per member per month rate for “community well”

Medicare-Medicaid beneficiaries for fiscal year 2017 exacerbated the financial

pressure on MMPs, raising concerns about the case-mix of the community population

and whether members are appropriately assigned to rate cells. Further, the decline in

the community rate highlighted MMPs’ concerns about their financial performance

under the demonstration.

• In July 2018, CMS announced the State and all three MMPs agreed to a two-year

demonstration extension to December 2020 as a result of three-way contract

amendment changes (e.g., adjustments to the savings percentage, enrollment methods,

and administrative reporting requirements).

generally reflected in the Medicare Advantage rates, but are adjusted for a geographic practice

cost index and the current hospital wage index. The Medicare Part D baseline is calculated using

the Part D national average monthly bid amount and is risk-adjusted using the existing Part D

prescription drug RxHCC model (South Carolina three-way contract, 2017, p. 196).

Average monthly low-income cost sharing subsidies and Federal reinsurance amounts are

estimated by CMS, and the total is added to the risk-adjusted rates. The Medicaid component is

risk-adjusted based on the rates applied to each of four enrollment categories that reflect

community or facility residence and HCBS waiver status. The structure of the per member per

month payment rate for these four categories (“the rate cell”) was developed to align payment

with risk while encouraging a rebalancing of nursing facility care to community-based care.

Table 11 highlights the Medicaid rate cell structure.

Table 11

Rate cell Medicaid payments per member per month, demonstration year 1 and

demonstration year 2

Rate cell CY 2015 CY 2016 CY 2017

Community $80.44 $95.98 $83.21

Nursing Facility $4,688.42 $5,067.55 $5,253.91

HCBS Waiv

er $1,23

2.19 $1,200.09 $1,170.83

HCBS Waiver-Plus Rate $3,052.26 $3,259.20 $3,361.13

CY = calendar year.

NOTES: These rates reflect the application of the savings percentage but do not reflect the quality withhold

percentages for each year. Demonstration year 1 = February 1, 2015–December 31, 2016; demonstration year 2 =

January 1, 2017–December 31, 2017.

SOURCE: SCDHHS, 2015; Palmer, Howard & Laudenschlager, 2017. Milliman Client Report. Healthy

Connections Prime Calendar Year 2017 Amendment—DRAFT Medicaid Rate Component.

7.1.2

Savings Percentage

A saving percentage is applied to the Medicare Parts A and B and to the Medicaid

components of the monthly rate, but not to the Part D component (South Carolina three-way

contract, 2017, p. 197). These percentages are based on what CMS and SCDHHS expect to be a

reasonable amount of savings achieved by the plans over the course of the demonstration year,

relative to the cost of Medicare and Medicaid service delivery in the absence of the

demonstration. Table 12 shows the aggregate savings percentage for years 1, 2, and 3. Savings

percentages are not applied to the Part D component. CMS monitors Part D costs on an ongoing

basis, and material changes may be factored into future year savings percentages (South Carolina

three-way contract, 2017, p. 197).

One plan expressed that 4 percent savings in year 3 would be unfair and unrealistic, given

the extremely low enrollment in the first 14 months of the demonstration due to the delay in

passive enrollment. (For more on the delay in passive enrollment, please see Section 3.2.2,

Phases of Enrollment.) One MMP official expressed frustration that the savings percentage does

not account for the low enrollment numbers and passive enrollment’s uneven “…false starts, no

starts, delays, reinstatements, etc. ….” In response to MMP financial concerns, State officials

worked with the plans and CMS in fall 2017 and 2018, to identify possible solutions that would

enable the demonstration to continue through 2020 with all three plans. The savings percentage

reduction from 4 to 3 percent for calendar year 2018 and beyond was an important factor in these

discussions.

Table 12

Savings percentages by demonstration year

Demonstration year Period covered Savings percentage

Year 1 February 1, 2015–December 31, 2016 1

Year 2 January 1, 2017–December 31, 2017 2

Year 3 January 1, 2018–December 31, 2018 3*

NOTE: The July 2018 contract amendments formally reduced the savings percentage rate for demonstration year 3

from 4 to 3 percent.

SOURCE: https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-

Medicaid-Coordination-Office/FinancialAlignmentInitiative/Downloads/SCContractAmendment_07012018.pdf

7.1.3

Performance Incentives

Quality withholds. CMS and the State withhold a percentage of payment that MMPs are

able to earn back based on performance on specific quality measures. In demonstration year 1

(February 2015–December 2016), CMS and the State withheld one percent of the Medicare Parts

A and B and Medicaid components of the payment, but not the Part D component (South

Carolina three-way contract, 2014, p. 194). Plans are evaluated on specific quality measures in

order to earn back the withhold percentage. The withhold measures for demonstration year 1

included six implementation and process measures; in demonstration years 2 and 3, the withhold

percentage increased to two and three percent, respectively, and the quality withhold measures

expanded to include utilization measures such as all-cause 30-day readmission and follow-up

after hospitalization for mental illness. MMPs are repaid the withhold rate based on the percent

of the measures they passed. To receive a “pass” on the measure, the MMPs must meet the

performance benchmarks set by CMS and the State. To receive 100 percent of the withhold

amount, MMPs must pass at least 80 percent of the measures (CMS, 2018d). In July 2018,

contract amendments show that the quality withhold percentage of three percent will stay in

effect through 2020 (CMS, 2018a).

Rebalancing incentives. MMPs are offered rebalancing incentives to encourage the

placement of members in the community, and to discourage placement in an NF. The method to

accomplish this includes incentives and penalties where enhanced Medicaid payment is given for

90 days following a transition from a NF to the community, and a lower Medicaid payment is

given for 90 days following a transition from the community to a NF (South Carolina three-way

contract amendment, 2018, p. 4). SCDHHS provides an additional Medicaid incentive through

SCDHHS’s Money Follows the Person waiver called Home Again. State officials indicated that

as of the end of calendar year 2017, there have been no Healthy Connections Prime enrollees

who have been simultaneously enrolled in Money Follows the Person.

7.1.4

Medical Loss Ratio

MMPs are required each year to meet a target medical loss ratio (MLR) of 85 percent;

this is the minimum revenue that must be used on expenses directly for medical claims or care

coordination If the MLR is calculated as falling below the 85 percent threshold, then the MMP is

required to reimburse the State and CMS an amount equal to the difference between the

calculated MLR and 85 percent, multiplied by the coverage year revenue. The payment amount

to CMS and the State will be proportional between each payer’s contribution to the aggregate of

the MMPs Medicaid and Medicare revenue. MMPs are required to make this payment, or the

amount will be offset on future capitation payments (South Carolina three-way contract, 2014,

p. 190). As of quarter 2 of fiscal year 2018, the MMPs’ MLRs for demonstration years 1 and 2

had not yet been calculated.

7.2 Financial Impact

MMPs and State officials reported several challenges with respect to the rates and

structure of payment. Low beneficiary enrollment over the course of the demonstration

precluded engaging in alternative payment arrangements with providers, and contributed to

sustained financial losses for at least two of the three MMPs. Additionally, the State planned a

projected 13 percentage point drop in the Medicaid capitated rate to be retroactively applied to

the community cell rate for fiscal year 2017. This reduction was based on the 2-year historical

fee-for-service experience of beneficiaries who would have been eligible for the demonstration

during the baseline period (South Carolina three-way contract, 2017, p. 194). The reduction from

the previous year is, in part, reflective of a longer claims runout period and a reclassification of

beneficiaries from the community rate cell to the nursing facility rate cell during the baseline

period (Milliman, 2017, p. 9). This reduction in the community rate was another source of

frustration for the MMPs. Finally, the State’s eligibility tracking system has struggled to

efficiently identify eligible beneficiaries in community versus nursing facilities, as well assigning

eligible beneficiaries to the HCBS waiver payment cell. These complications have resulted in

delayed and inaccurate Medicaid payments to the MMPs, further contributing to the financial

struggle MMPs have had during the demonstration.

7.2.1

Early Implementation Experience

From the beginning of the demonstration, MMPs reported several financial implications

of low Healthy Connections Prime enrollment. MMPs reported that they had entered the

Financial Alignment Initiative on the assumption that over 50,000 beneficiaries would be eligible

to participate in the demonstration, whereas the adjusted number was 24,000 (see Section 3.2.1,

Eligibility). Furthermore, a number of delays in passive enrollment resulted in plans having even

fewer enrollees than they had estimated. During site visit interviews, MMPs frequently noted

that enrollment had not reached “critical mass,” thereby challenging the MMPs’ ability to

contract with at least one major health system, precluding implementation of alternative payment

methodologies with providers, and assessing the adequacy of capitated rates.

The State’s eligibility tracking system also has caused administrative and financial

concerns for the MMPs. Plans reported early in the demonstration that the State’s eligibility

tracking system was not accurately identifying beneficiaries who transition into a nursing facility

from the community. MMPs continued to receive a Medicaid community rate, beyond the

rebalancing period, for enrollees residing in nursing facilities. MMPs and the State have been

working to resolve this issue, but, as of the end of 2017, the backlog in payments have resulted in

millions of dollars of unpaid monthly payments to the MMPs. The State acknowledged this as an

issue with the “integrity of the data” from their eligibility system. State officials indicated they

had to manually determine Medicaid rates for up to 150 enrollees.

In addition to these issues, nursing facilities have erroneously charged both the State and

the MMPs for the same nursing facility stay for some enrollees, in part because the State claims

system did not appropriately reject claims submitted by nursing facilities for MMP enrollees.

Thus, the State was also paying for nursing facility stays for MMP enrollees. A reconciliation

process was underway in calendar year 2017 and completed in summer 2018 (SCDHHS, 2018).

Future reports will include further discussion on this topic.

As described in Section 2.2.1, Provider Arrangements and Services, the demonstration

phased-in the transfer of HCBS services from the State to MMPs over a period of time. During

this transition, the State continued to pay for HCBS services and regularly recouped the amount

from the MMPs. This was a benefit to HCBS workers who preferred to continue to be paid

through the State’s Phoenix system as they had prior to the demonstration. This process has had

administrative challenges as it relates to other types of community-based LTSS. For example,

residential care facilities are carved out of the capitated rate paid to plans and are paid by the

State. While the State should recoup payments from MMPs for HCBS services, the State was

inappropriately charging MMPs close to 2 million dollars for residential care services. When this

discrepancy was discovered in July 2017, the State stopped recouping all HCBS service

payments from the MMPs, to identify and fix this eligibility and payment error.

To resolve these payment issues, the State contracted with Milliman to determine an

accurate amount to be reconciled, which was completed in mid-2018. Findings from the

Milliman analysis will be discussed in the second evaluation report.

7.2.2

Rate Methodology Design Implications

The design of the MMP rates had implications on the financial performance of the

MMPs. One MMP reported during the first year of the demonstration that the quality withhold

percentage should be smaller to reflect the operational costs of the demonstration. MMP officials

noted concerns regarding quality withholds as they relate to enrollment volume, which was

roughly 1,800 enrollees in the first 14 months. Interviewees also noted that start-up investment

varied in part based on MMP staffing models. For example, some MMPs leveraged staff from

other product lines to support the MMP product, whereas others hired a new cohort of staff to

administer the MMP product. As a result, some MMPs were more financially vulnerable to the

one percent-withhold of the rate.

Additionally, some MMP officials noted that without passive enrollment, the

denominators were too small to produce a reliable quality withhold performance measure,

placing at risk revenue that is needed to help cover other upfront investments. Given the start-up

and implementation costs, some plans suggested that the quality withhold should have been

delayed for the first year. The State and CMS worked to simplify the withhold reporting

requirements, but they maintained the withhold percentage. For calendar years 2015 and 2016,

three MMPs met the benchmark for a sufficient number of measures to receive the full withhold

amount. Performance on the quality withhold measures for calendar year 2017 had not been

determined at the time of this report.

The adequacy of the Medicaid capitated rates was also a concern for MMPs early on.

Limited enrollment during the early months of the demonstration, due to the delay in passive

enrollment, precluded plans from determining whether the capitated rates were adequate. There

was a small increase in the Medicaid rates in 2016; however, by late 2016 some MMPs began

reporting that the Medicaid capitated rates were insufficient given their low enrollment prior to

passive enrollment. At the time, there was hope that once passive enrollment began, the capitated

rates would be adequate as more members would be enrolled along with a greater variability in

case mix.

In calendar year 2017, the State sharply reduced the capitated Medicaid rates for the

community cell to which the vast majority of MMP enrollees are assigned (see Table 11). This

reduction was based in part on the expected baseline spending for the FFS dual eligible

population, absent the demonstration (SCDHHS, 2018), and corrections to the baseline data. The

reduction in the community cell rate further elicited specific concerns around member eligibility

assignment to the community vs waiver cells. Two plans in particular voiced concern that the

bulk of their members were assigned to the community cell. One plan posited that member risk

was not accurately determined when assigned to the community cell. Other MMP officials

reported that many of their community rate enrollees qualify for one of the three waivers

included in the demonstration, but because of backlogs and inefficiencies in the State’s waiver

eligibility process, member applications for waiver eligibility have been delayed. Therefore,

MMPs pushed back on the impending rate Medicaid decrease, which resulted in a slight

adjustment in the rate but still an overall decline from $96 to $83 per member per month. MMPs

reported that they cannot determine if the community cell rate is adequate because they are not

clear if their members are appropriately assigned.

7.2.3

Cost Experience

The financial implications of low enrollment and underpayment for enrollees in nursing

facilities are such that in 2017, two of the three plans identified key contractual conditions they

hoped would be met in order to continue in the demonstration beyond 2018. These conditions

included modifications to the capitated rates, specifically the Medicaid capitated rates, changes

to the savings percentage and quality withhold amount, reducing administrative burden, and

expanding eligibility for passive enrollment. Early in the demonstration, MMPs had noted that

there was a potential for cost savings through better care coordination resulting in utilization

reduction. However, MMPs reported that lower than expected enrollment, a reduction in the

community cell rate, and what they believe to be adverse selection, had resulted in financial

losses by the end of 2017. Plans indicated at that time that sustainability in the demonstration, as

it was originally designed, would be problematic.

One plan noted that the demonstration was always a “breakeven” proposition for them.

The decline in the community cell rate for fiscal year 2017 caused concerns that the

demonstration would be a financial loss, i.e., this plan could lose 1.2 million dollars in fiscal year

2018 on their demonstration product. Another plan expressed that “…in totality, we’re talking

about a 6–7 digit loss we’ve incurred since [the demonstration’s] inception ...” and that they are

working to improve efficiency and medically manage their population.

Two plans expressed that they believe that adverse selection was costing their plans more

than expected. MMP officials noted that members in the community cell have required intensive

services, including waiver-like services, to support them in the community and reduce transitions

to long-term care facilities. However, State officials have reported that plans are reluctant to

supply these services because of not wanting to financially sustain what MMPs see as

inefficiency in the State waiver eligibility determination system. Table 13 illustrates RTI’s

analysis of the distribution of the Hierarchical Condition Category (HCC) risk scores among

enrollees by each plan during the first demonstration year. The average HCC score across all

MMPs indicate their enrollees are predicted to have costs greater than the general Medicare FFS

population. Although we would not expect the more complex dually eligible population to be

comparable to the general Medicare FFS population, we provide these results to gauge member

medical acuity among the three MMPs. The average HCC ranged from 1.31 to 1.39, and the

median scores ranged from 1.03 to 1.11 suggesting little apparent differences in the medical risk

of enrollees across the three MMPs.

Table 13

Average, minimum, and maximum HCC score

by MMP during demonstration year 1 (February 1, 2015–December 31, 2016)

Plan

Number

enrollees

Average

HCC

score Min

Percentile

Max

10th 25th 50th 75th 90th

Absolute Total Care 3,774 1.39 0.27 0.55 0.80 1.11 1.62 2.57 9.75

Molina 2,904 1.31 0.27 0.54 0.74 1.03 1.50 2.40 9.63

Advicare 2,557 1.39 0.27 0.54 0.77 1.08 1.65 2.67 8.33

Select Health 5,664 1.37 0.27 0.57 0.79 1.06 1.60 2.56 10.31

HCC = Hierarchical Condition Category

NOTES: 118 enrollees have missing HCC scores and were not included in this calculation. Advicare withdrew from

the demonstration in 2016. HCC risk score is a measure of the predicted relative annual cost of a Medicare

beneficiary based on the age, gender, and the chronic conditions present in recent Medicare claims. Beneficiaries

with a score of 1 are predicted to have average cost in terms of annual Medicare expenditures. Beneficiaries with

HCC scores less than 1 are predicted to have below average costs, whereas beneficiaries with scores of 2 are

predicted to have twice the average annual costs.

At the time of this report, the State did not have information on whether the

demonstration was achieving Medicaid savings in the State. State officials reported that they

expect that average Medicaid spending per demonstration enrollee should be lower compared to

other Medicare-Medicaid dual eligible groups, in part because savings were built into the

capitated rates. State officials reported that formal cost reports and analyses are currently

underway.

In July 2018, CMS announced changes to the South Carolina demonstration that may

allay some financial concerns of all three MMPs that will participate in the demonstration

beyond 2018. For example, one change includes allowing passive enrollment of eligible

beneficiaries who are enrolled in Medicare Advantage D-SNP plans. State officials noted this

would increase the eligible population and expand the risk pool to include more healthy

enrollees. Other amendments included reductions to shared savings percentages, and a reduction

in some reporting and other administrative requirements. These changes fall outside the reporting

period of this report and will be discussed in future reports.

8. Service Utilization

Highlights

• As measured across all eligible beneficiaries, the demonstration resulted in a 19.0

percent reduction in the probability of inpatient admissions, a 16.9 percent reduction in

the probability of skilled nursing facility admission, an 18.5 percent reduction in the

probability of overall ambulatory care-sensitive condition (ACSC) admissions, and a

19.8 percent reduction in the probability of chronic ACSC admissions. Conversely, the

demonstration resulted in a 27.2 percent increase in the probability of any new long-

stay nursing facility use.

• These reported changes, which although appear large in percentage terms in

comparison to the comparison group, are actually quite small difference-in- differences

(DinD) changes in absolute terms, reflecting usually less than a one percentage point

change (e.g., 0.77 increase/decrease) in the probability of service use. So although the

DinD change is statistically significant, the actual change is quite small.

• DinD results for those with serious and persistent mental illness (SPMI) on the

probability of inpatient or skilled nursing facility admission or emergency room visits

were in the same direction and to a similar degree as those for all eligible beneficiaries,

as well as for the number of evaluation and management visits.

The purpose of the analyses in this section is to understand the effects of the South

Carolina Healthy Connections Prime demonstration through demonstration year 1 (ending

calendar year 2016) using DinD regression analyses that control for differences in health and

other factors between the demonstration and comparison groups. The results of these analyses

represent impact estimates of the South Carolina demonstration on all demonstration eligible

beneficiaries, not just those enrolled in a Medicare-Medicaid Plan (MMP). In addition,

descriptive statistics on service utilization are provided for selected Medicare services in

Appendix C. Utilization data were analyzed for only three of the four MMPs in Healthy

Connections Prime; Advicare encounters were not included or analyzed because RTI

International deemed them incomplete.

Table 14 presents an overview of the results from analyses using Medicare and Minimum

Data Set (MDS) data through demonstration year 1. The relative direction of all statistically

Caution should be used when interpreting the results on long-stay NF stays. The demonstration was associated

with an increase in the probability of long-stay NF stays, relative to the comparison group. However, State-level

eligibility determination delays for long-stay nursing facility approval could have contributed to this finding.

During the period covered by this report, long-stay nursing facility residents were excluded from being eligible

to enroll into the demonstration. However, due to delays in determining nursing facility eligibility, individuals

newly requiring long-stay nursing facility services were enrolled in the demonstration, thus raising the

probability of a long-stay nursing facility stay in the demonstration group. Potentially, if beneficiaries with those

long-stay nursing facility stays had been more quickly identified, they would have been ineligible for

demonstration enrollment.

significant results at the p < 0.10 significance level (derived from 90 percent confidence

intervals) is shown.

The South Carolina demonstration had a statistically significant effect on eight utilization

and quality of care outcomes through demonstration year 1: the probability of an inpatient

admission, the probability of overall and chronic ambulatory care sensitive condition (ACSC)

admissions, the probability of an emergency room (ER) visit, monthly preventable ER visits, the

probability of a skilled nursing facility (SNF) admission, and monthly physician evaluation and

management (E&M) visits all were lower for the demonstration group relative to the comparison

group. The probability of any new long-stay nursing facility (NF) use, on the other hand, was

higher. There was no statistically significant difference between the demonstration and

comparison groups in all-cause 30-day readmissions or the probability of a 30-day follow-up

visit after mental health discharge.

For the SPMI population, the demonstration had a statistically significant effect for half

of the 10 outcome measures: the probability of an inpatient admission, all-cause 30-day

readmissions, the probability of an ER visit, the probability of a SNF admission, and monthly

E&M visits were all statistically lower for the demonstration group relative to the comparison

group.

Table 14

Summary of South Carolina demonstration impact estimates for demonstration period

(February 1, 2015–December 31, 2016)

(p < 0.1 significance level)

Measure

All demonstration

eligible beneficiaries

Demonstration eligible

beneficiaries with SPMI

Probability of inpatient admission Decreased Decreased

Probability of ambulatory care sensitive condition (ACSC)

admissions, overall

Decreased NS

Probability of ACSC admissions, chronic Decreased NS

All-cause 30-day readmissions NS Decreased

Probability of emergency room (ER) visit Decreased Decreased

Preventable ER visits Decreased NS

30-day follow-up after mental health discharges NS NS

Probability of skilled nursing facility (SNF) admission Decreased Decreased

Probability of any new long-stay nursing facility (NF) use Increased N/A

Physician evaluation and management (E&M) visits Decreased Decreased

N/A = not applicable; NS = not statistically significant; SPMI = serious and persistent mental illness.

SOURCE: RTI analysis of Medicare and Minimum Data Set data.

8.1 Overview of Benefits and Services

Under Healthy Connections Prime, eligible beneficiaries enroll in a managed care plan

that covers Medicare and Medicaid services, as well as new or expanded services available under

the demonstration. Healthy Connections Prime plans provide care coordination, a new palliative

care benefit, and may provide flexible benefits, known in South Carolina as “waiver-like”

services to enable enrollees to postpone institutionalization or prevent higher levels of care for

frail beneficiaries. In addition, each plan has supplemental benefits available to members

enrolled in their Healthy Connections Prime products that vary by plan.

8.2 Impact Analyses on the Demonstration Eligible Population

The population analyzed in this section includes all beneficiaries who met demonstration

eligibility criteria in South Carolina or in the comparison areas for South Carolina. For context,

in South Carolina, approximately 38 percent of eligible beneficiaries in demonstration year 1

whose utilization was analyzed were enrolled in Healthy Connections Prime. Appendix A

provides a description of the comparison group for South Carolina. Please see Section 3.2.1,

Eligibility, for details on demonstration eligibility. The subsection following this section presents

the results for demonstration eligible beneficiaries with SPMI.

Appendix B contains a description of the evaluation design, the comparison group

identification methodology, data used, measure definitions, and regression methodology used in

estimating demonstration impacts using a difference-in-differences approach. The regression

methodology accounts for differences between the demonstration and comparison groups over

the predemonstration period (February 1, 2013 to January 31, 2015) and the demonstration

period (February 1, 2015, to December 31, 2016) to provide estimates of demonstration impact.

Figures 1 and 2 display the South Carolina demonstration’s effect on key service

utilization measures for the demonstration group relative to the comparison group through

demonstration year 1. The demonstration decreased the probability of inpatient admissions by

0.77 percentage points (90 percent CI: –0.93, –0.61), decreased the probability of ER visits by

0.62 percentage points (90 percent CI: –0.80, –0.45), decreased the probability of SNF

admissions by 0.20 percentage points (90 percent CI: –0.26, –0.14), and decreased monthly

E&M visits by 0.0493 visits (90 percent CI: –0.0634, –0.0353). On the other hand, the

demonstration resulted in a 1.69 percentage point increase (90 percent CI: 1.29, 2.09) in the

probability of any new long-stay NF use during the first demonstration year.

Caution should be used when interpreting the results on long-stay NF stays. The

demonstration was associated with an increase in the probability of long-stay NF stays, relative

to the comparison group. However, State-level eligibility determination delays for long-stay

nursing facility approval could have contributed to this finding. During the period covered by

this report, long-stay nursing facility residents were excluded from being eligible to enroll into

the demonstration. However, due to delays in determining nursing facility eligibility, individuals

newly requiring long-stay nursing facility services were enrolled in the demonstration, thus

raising the probability of a long-stay nursing facility stay in the demonstration group.

Potentially, if beneficiaries with those long-stay nursing facility stays had been more quickly

identified, they would have been ineligible for demonstration enrollment.

Figure 1

Demonstration effects on service utilization for eligible beneficiaries in South Carolina—

Difference-in-differences regression results for the demonstration period,

February 1, 2015–December 31, 2016

(90 percent confidence intervals)

E&M = evaluation and management; ER = emergency room; SNF = skilled nursing facility.

SOURCE: RTI International analysis of Medicare data.

Figure 2

Demonstration effects on long-stay nursing facility use for eligible beneficiaries in South

Carolina—Difference-in-differences regression results for the demonstration period,

February 1, 2015–December 31, 2016

(90 percent confidence interval)

NF = nursing facility.

SOURCE: RTI International analysis of Minimum Data Set data.

Table 15 provides estimates of the regression-adjusted mean values of the utilization

measures for the demonstration and comparison groups for the predemonstration and

demonstration periods for each service. The purpose of this table is to understand the magnitude

of the difference-in-differences (DinD) estimate relative to the adjusted mean outcome value in

each period. The values in the third and fourth columns represent the post-regression, mean

predicted value of the outcomes for each group in each period, based on the composition of a

reference population (the comparison group in the demonstration period). These values show

how different the two groups were in each period as well as the relative direction of any potential

effect in each group over time. In addition to the graphic representation above, the DinD estimate

is also provided, along with the p-value and the relative percent difference of the DinD estimate

compared to an average mean use rate for the comparison group during the first demonstration

period.

Table 15

Adjusted means and impact estimate for eligible beneficiaries in the demonstration and comparison groups in South Carolina

through December 31, 2016

Measure Group

Adjusted mean for

predemonstration

period

Adjusted mean for

demonstration

period

Relative

difference

(%)

Regression-adjusted

difference-in-differences

(90% confidence interval) p-value

Probability of inpatient admission Demonstration group 0.0473 0.0285 −19.0 −0.0077 <0.0001

Comparison group 0.0540 0.0405 (−0.0093, −0.0061)

Probability of ER visit Demonstration group 0.0663 0.0549 −9.0 −0.0062 <0.0001

Comparison group 0.0752 0.0691 (−0.0080, −0.0045)

Physician E&M visits Demonstration group 0.8588 0.7701 −5.4 −0.0493 <0.0001

Comparison group 0.9511 0.9063 (−0.0634, −0.0353)

Probability of SNF admission Demonstration group 0.0128 0.0067 −16.9 −0.0020 <0.0001

Comparison group 0.0179 0.0119 (−0.0026, −0.0014)

Probability of any new long-stay

NF use

Demonstration group 0.0275 0.0582 27.2 0.0169 <0.0001

Comparison group 0.0427 0.0621 (0.0129, 0.0209)

E&M = evaluation and management; ER = emergency room; NF = nursing facility; NS = not statistically significant; SNF = skilled nursing facility.

NOTES: Even though the comparison group was carefully developed to have similar characteristics to the demonstration group, there are always slight

differences in demographic, health, and area characteristics between the demonstration and comparison groups. The two types of results reported in this table

take these differences into account, but use different statistical methods to do so. Before calculating the mean values reported in the third and fourth columns in

this table, RTI adjusted the composition of the demonstration’s baseline and demonstration period groups and the comparison baseline period group to match the

characteristics of the comparison group in the demonstration period so that the means do not reflect any differences in the groups’ characteristics. The regression

DinD approach, results reported in the sixth column of this table, controls for these differences automatically, without changing the underlying characteristics of

the demonstration and comparison groups. Because of these differing methods, the difference-in-differences results obtained from the regression may differ

slightly from a similar calculation using the results in the adjusted mean columns. The relative percentage difference in the fifth column is calculated by dividing

the difference-in-differences value in column 6 by the value for the comparison group in the demonstration period in column 4.

SOURCE: RTI International analysis of Medicare and Minimum Data Set data.

To interpret the adjusted mean values in the third and fourth columns, as an example, the

adjusted mean of monthly inpatient admissions was lower for the demonstration group than for

the comparison group in both the predemonstration period and demonstration period. Similarly,

the adjusted mean of the probability of monthly ER visits was also lower for the demonstration

group than for the comparison group in both the predemonstration period and demonstration

period.

To help interpret the relative percentage difference reported in the fifth column, the DinD

estimate for monthly physician E&M visits of -0.0493 implies an annual relative percentage

decrease of 5.4 percent as a result of the demonstration. Similarly, the DinD estimate for the

probability of an SNF admission of -.0020 implies an annual relative percentage decrease of 16.9

percent as a result of the demonstration.

Figure 3 displays the South Carolina demonstration’s effects on RTI quality of care and

care coordination measures for the demonstration group relative to the comparison group through

demonstration year 1. The demonstration decreased monthly preventable ER visits by 0.0026

visits (90 percent CI: −0.0041, −0.0010). The demonstration also resulted in a 0.18 percentage

point decrease (90 percent CI: −0.0026, −0.0010) in the probability of an ACSC admission for

overall conditions and a 0.11 percentage point decrease (90 percent CI: −0.0018, −0.0005) for

ACSC chronic condition admissions. There was no statistically significant demonstration effect

on the probability of a 30-day follow-up after a mental health discharge or all-cause 30-day

readmissions through demonstration year 1.

Figure 3

Demonstration effects on RTI quality of care measures for eligible beneficiaries in South

Carolina—Difference-in-differences regression results for the demonstration period,

February 1, 2015–December 31, 2016

(90 percent confidence intervals)

ACSC = ambulatory care sensitive condition; ER = emergency room.

SOURCE: RTI International analysis of Medicare data.

Table 16 provides estimates for the regression-adjusted mean value for each of the

demonstration and comparison groups for the predemonstration and demonstration periods for

the RTI quality of care and care coordination measures. The purpose of this table is to

understand the magnitude of the DinD estimates for quality of care outcomes relative to the

adjusted mean values in each period. The values in the third and fourth column represent the

post-regression, mean predicted value of the outcomes for each group in each period, based on

the composition of a reference population (the comparison group in the demonstration period).

These values show how different the two groups were in each period and the relative direction of

any potential effect in each group over time. In addition to the graphic representation above, the

DinD estimate is also provided for reference, along with the p-value and the relative percent

change of the DinD estimate compared to an average mean use rate for the comparison group

during the first demonstration year.

To interpret the adjusted mean values in the third and fourth columns, as an example, the

adjusted mean of preventable ER visits was lower for the demonstration group than for the

comparison group in both the predemonstration period and the demonstration period.

To help interpret the relative percentage difference reported in the fifth column, the DinD

estimate for the probability of an ACSC admission (overall) implies an annual relative

percentage decrease of 18.5 percent as a result of the demonstration; and the DinD estimate for

the probability of an ACSC admission (chronic) implies an annual relative percentage decrease

of 19.8 percent as a result of the demonstration.

Table 16

Adjusted means and impact estimate for eligible beneficiaries in the demonstration and comparison groups in South Carolina

through the demonstration period February 1, 2015–December 31, 2016

Measure Group

Adjusted mean for

predemonstration

period

Adjusted mean for

demonstration

period

Relative

difference

(%)

Regression-adjusted

difference-in-

differences estimate

(90% confidence

interval)

p-value

Preventable ER visits Demonstration group 0.0395 0.0335 −6.0 −0.0026 0.0066

Comparison group 0.0468 0.0425 (−0.0041, −0.0010)

Probability of ACSC admission, overall Demonstration group 0.0113 0.0067 −18.5 −0.0018 0.0004

Comparison group 0.0131 0.0096 (−0.0026, −0.0010)

Probability of ACSC admission, chronic Demonstration group 0.0068 0.0038 −19.8 −0.0011 0.0037

Comparison group 0.0083 0.0058 (−0.0018, −0.0005)

Probability of 30-day follow-up after

mental health discharge

Demonstration group 0.2290 0.2248 NS −0.0567 0.2417

Comparison group 0.2269 0.2734 (−0.1364, 0.023)

All-cause 30-day readmissions Demonstration group 0.2824 0.3398 NS −0.0045 0.8153

Comparison group 0.3450 0.4197 (−0.0366, 0.0275)

ACSC = ambulatory care sensitive conditions; ER = emergency room.