REVIEW OF KANCARE: COST AND UTILIZATION November 2017
INTRODUCTION KanCare, the state of Kansas managed Medicaid program, will reach the end of its five-year demonstration period under a 1115 CMS waiver at the end of 2017. CMS recently granted a one-year extension for the program, which contracts with three managed care organizations (MCOs) to provide health care for the state s Medicaid enrollees and covers more than 415,000 Kansans. Numerous reports have been published on KanCare, but a concise picture of the program s performance is still needed. This effort summarizes the information in hundreds of pages of external quality reporting to draw out important findings about KanCare as it nears the end of its demonstration period. Leavitt Partners authored this secondary analysis report, with review from Kansas Health Institute (KHI), based on detailed reports and data provided by the Kansas Foundation for Medical Care (KFMC) as the Section 1115 demonstration evaluator. This effort highlights KanCare s observed performance relative to its stated goals and commitments over the four completed operating years (2013 2016) since the program began in 2013. This report, the first of three, focuses on whether KanCare controlled medical costs and improved beneficiaries health care utilization. The second report will focus on health care quality and access in the program, and the third will discuss the MCOs efforts related to coordinating care and improvement projects. KANCARE S GOALS Under KanCare, the state contracts with three health plans to provide Medicaid managed care services to beneficiaries Amerigroup (a subsidiary of Anthem), Sunflower State Health Plan (a subsidiary of Centene Corp.), and United Healthcare of the Midwest. KanCare requires the MCOs to report quality and performance measures, and places up to 2 percent of their revenue at risk if they fail to meet quality benchmarks for physical health, behavioral health, and long-term care. The Kansas Department of Health and Environment (KDHE) articulated four general goals that the state hopes to achieve through KanCare: (1) control Medicaid costs, (2) improve the quality of care, (3) provide integration and coordination of care, and (4) establish long-lasting reforms to sustain health improvements and provide a model for Medicaid reforms for other states. 1 KanCare included objectives to lower overall costs and promote healthy lifestyles (ostensibly preventing costly future health care). KDHE hypothesized that KanCare would reduce Medicaid cost growth by emphasizing health, wellness, prevention, and early detection as well as integration and coordination of care. 2 Slowing cost growth could occur as better care coordination and integration increase beneficiaries access to and utilization of preventive care, reducing the occurrence of costly medical events such as hospital stays and residence in long-term skilled nursing facilities (SNFs). DATA SOURCES This report draws from the external reviews of KanCare conducted quarterly and annually by KFMC, the state s contracted External Quality Review Organization and Section 1115 demonstration evaluator. KFMC s reports are based on patient and provider data collected by the health plans and reported to the state. KDHE s Division of Health Care Finance then submits these reports to CMS as specified by the terms of the Section 1115 waiver. The KFMC reports provide data on utilization, costs, quality metrics, and access measures for each completed year of KanCare, with some data running through the end of 2016. Leavitt Partners used these data to explore KanCare s effect on costs and utilization, drawing from the most recent reports that provided the data of interest in each category. Other data sources are cited as they appear in the report. PROGRAM COSTS AND UTILIZATION ACTUAL EXPENDITURE TRENDS Total program expenditures were $2.5 billion in 2013, rising to $2.9 billion in 2016 (in medical inflation-adjusted 2013 dollars), yielding an average compound annual growth rate (CAGR) of 7.45 percent (4.6 percent when adjusting for medical inflation) between 2013 and 2016. a By comparison, cost growth nationally has been higher over the same period; total Medicaid expenditures nationally grew at a CAGR of 9 percent (5.5 percent adjusted for inflation) from 2013 to 2016. a Medical inflation adjustment throughout this paper was performed using the medical care U.S. city average from the Bureau of Labor Statistics CPI- All Urban Consumers data (series ID CUUR0000SAM). Annual average indices were used to perform the adjustments. REVIEW OF KANCARE: COST AND UTILIZATION 2
Figure 1. Actual and Projected Medicaid Expenditures, CY 2006 CY 2019 5.0 Expenditures in Billions of $ 4.5 4.0 3.5 3.0 2.5 2.8 4.0 3.8 3.5 4.6 4.3 3.8 2.0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Actual Current Estimate 2012 KanCare Projection 2012 Projection w/o KanCare Source: Presentation by Kansas Department of Health and Environment to KanCare Oversight Committee, Aug 22-23, 2017. The figure is a replication of the figure on slide 5 of the presentation. 3 Note: This figure shows all program costs, including non-demonstration costs, not adjusted for inflation. F igure 2. Real Average Monthly KanCare Expenditures per Enrollee, CY 2013 CY 2016 Expenditures per Enrollee, 2013 $ $700 $600 $500 $400 $300 $200 $100 $ - $625 $613 $563 $563 2013 2014 2015 2016 Source: Leavitt Partners analysis of KFMC reports. Note: Dollar values are adjusted for medical inflation to 2013 dollars. EXPENDITURES RELATIVE TO PROJECTIONS KanCare aims to slow the growth of the state s Medicaid costs. During the program s design phase, KDHE predicted it would save about $850 million in total state and federal Medicaid costs over five years, with $370 million in direct savings to the state. Figure 1 shows KDHE s 2011 projections of future state Medicaid expendit ures both with and without KanCare, compared against actual state Medicaid expenditures from 2006 onward. These figures are for the entire Medicaid program (not just KanCare), and are not adjusted for inflation. Comparing actual Medicaid expenditures to pre-kancare projections over the four-year period from 2013 to 2016, Medicaid expenditures were $1.7 billion lower than projected under the status quo (without KanCare), roughly double the anticipated savings. Annual spending was about $400 million less (12.2 percent) than the original KanCare projection of $3.8 billion in 2016. 3 Figure 2 shows KanCare s average per-member per-month (PMPM) expenditures, adjusted for medical inflation to 2013 dollars. PMPM expenditures fell from $625 in 2013 to about $563 in 2016 (from $7,500 annually to $6,756 annually). REVIEW OF KANCARE: COST AND UTILIZATION 3
MEDICAID POPULATION ( PERCENT OF 2016 MEMBER MONTHS) CHILDREN (64 PERCENT) AGED/BLIND/DISABLED (10 PERCENT) AGED/BLIND/DISABLED, MEDICARE-ELIGIBLE (2 PERCENT) DEVELOPMENTALLY DISABLED WAIVER (2 PERCENT) LONG-TERM CARE (6 PERCENT) MEDICALLY NEEDY (0.4 PERCENT) MEDICALLY NEEDY, MEDICARE-ELIGIBLE (0.4 PERCENT) WAIVER (1 PERCENT) OTHER ADULTS (14 PERCENT) DESCRIPTION Children in the foster care/adoption system, long-term care, MCHIP, and children in needy households. CHIP enrollees are not included in this analysis. All Medicaid enrollees who are not eligible for Medicare but are eligible for Supplemental Security Income (SSI) due to age, blindness, or disability, including spend-down enrollees who fit this description. Excludes enrollees receiving home- and community-based services under 1915(c) waivers. All Medicaid enrollees who are eligible for Medicare and for SSI because of age, blindness, or disability, including spend-down enrollees who fit this description. Excludes enrollees receiving home- and community-based services under 1915(c) waivers. Enrollees eligible for Medicaid in the state s developmental disabilities waiver program. Enrollees in long-term institutional care settings, such as intensive care facilities and nursing homes, as well as people in the frail elderly and physical disabilities 1915(c) waivers. Medically needy enrollees who are not also eligible for Medicare. Medically needy enrollees who are also eligible for Medicare. Medicaid enrollees in the autism treatment program, serious emotional disturbance program, and the home- and community-based services waivers for individuals with traumatic brain injuries and children in need of technology-assisted home care services. All Medicaid-eligible adults not in the above categories, including pregnant women and needy families. Figure 3. Real CY 2016 PMPM State Expenditures and Percent Changes Relative to CY 2013 4000 Developmentally Disabled 3000 Long-term Care Real 2016 Spending PMPM (2013 $) 2000 Other Waiver Programs Medically Needy 2016 Member Months 100,000 500,000 1,000,000 1000 Aged/ Blind/ Disabled Medically Needy, Medicare-eligible Other Adults 0 Aged/ Blind/ Disabled, Medicare-eligible Children -60-45 -30-15 0 15 30 Real Percent Change in Spending PMPM, 2013 2016 Source: Leavitt Partners analysis of KFMC reports. Note: This figure uses 2013 expenditures as a baseline for percent changes. Underlying dollar values are adjusted for medical inflation to 2013 dollars. REVIEW OF KANCARE: COST AND UTILIZATION 4
K ancare s performance on cost should also be evaluated for different subsets of Medicaid beneficiaries, given that the state pays MCOs a different per-enrollee rate for each eligibility category. Per-person Medicaid expenditures varied significantly across the various Medicaid populations (see Figure 3). PMPM expenditures, adjusted for medical inflation, decreased from 2013 to 2016 for medically needy dualeligibles (people who qualify for both Medicare and Medicaid benefits), the aged/blind/disabled, those under various waiver programs, and other adults. On the other hand, PMPM expenditures increased significantly over the same period for medically needy non-dual eligible adults, long-term care populations, children, and those covered under the developmental disability waiver. Based on this figure, much of the savings from KanCare seems to have come from decreasing PMPM costs for a few specific eligibility categories, offsetting the rising costs in other categories. U TILIZATION TRENDS AND COST Data on KanCare beneficiaries utilization of different types of health care services can help to determine whether the program is meeting its goals to promote healthy lifestyles and improve more effective use of health care services. KFMC reports provide utilization claims data based on date of service, reflecting payments MCOs made to providers. During the program s operation, utilization decreased for some expensive types of care and increased for preventive care. In particular, expenditures decreased for hospital inpatient services (-18 percent) and grew very slowly for outpatient emergency services (+1 percent) and nursing facility care (+3 percent). F igure 4. Drivers of Expenditure Growth by Utilization Category, CY 2012 CY 213 30 Primary Care Physician Percent Change in Utilization per 1,000, 2012 2013 20 10 0-10 Transportation Dental HCBS Vision Pharmacy Outpatient ER Beh Health Outpatient Non-ER Nursing Facility FQHC/RHC* 2013 Expenditures (Millions of 2013 $) 100 200 300 400 Change in Spending, 2012 2013 Decreased spending Increased spending -20 Inpatient -40-20 0 20 40 60 Percent Change in Unit Cost, 2012 2013 *One important factor contributing to the high percent increases in costs for primary care and FQHCs is the ACA primary care payment increase that occurred in 2013 and 2014. For Kansas, the Kaiser Family Foundation estimated this increase would be around 29 percent for primary care providers. Source: Leavitt Partners analysis of KFMC reports. Note: KFMC defines utilization per 1,000 as Units Reported/Member Months x 12,000, illustrating the services used per 1,000 beneficiaries in a 12-month period. There is no indication as to whether the definition of HCBS units or the population for which utilization is calculated have changed over time. Medical inflation in 2013 was about 2.5 percent on average, but these figures are not adjusted for inflation because it is not obvious which inflation factor should be applied to which category of expenditures. REVIEW OF KANCARE: COST AND UTILIZATION 5
As Figure 4 shows, utilization patterns shifted during the first year of KanCare toward community settings and away from costly hospital-based facilities and nursing facilities, a positive change for reducing the total cost of care. Increases in the utilization of primary care and federally qualified health centers (FQHCs)/rural health centers (RHCs) led, in part, to 78 percent increases in total (federal and state) expenditures on both categories (approximately $70 million and $17 million, respectively) in the first year of the program in 2013. Increased vision and dental utilization was also associated with expenditure increases in those categories. While utilization of non-emergency medical transportation (NEMT) increased by 15 percent, the MCOs paid lower per-unit costs for these services, leading to a 12 percent decline in NEMT expenditures. An increased utilization rate and unit cost for home and community-based services occurred alongside decreased expenditures for this category because fewer people used these services overall. Despite per-unit expenditure growth in some services, decreased utilization of those services in KanCare s first year led to net savings in those categories. For nursing facility days, outpatient emergency room services, and pharmacy, expenditures grew less than 3 percent despite per-unit costs in those categories rising 17 percent, 8 percent, and 6 percent, respectively. Hospital inpatient utilization and expenditures both declined significantly, resulting in over $79 million in savings on that category despite an 8 percent increase in the cost per inpatient day. One important factor to note affecting some utilization categories unit costs is an Affordable Care Act provision that used federal funds to increase Medicaid payments for primary care services in calendar years 2013 and 2014. According to a Kaiser Family Foundation report, these fees were projected to increase around 29 percent in 2013 for primary care providers in Kansas, including FQHCs, RHCs, and some behavioral health providers. 4 KFMC s 2016 annual report provides utilization data for 2015, again compared to 2012 (the year before the program started), and this data is presented in Table 1. By 2015, the number of primary care and FQHC/RHC claims Tab le 1. Utilization, 2012 and 2015 UTILIZATION PER 1,000 DIFFERENCE UTILIZATION PERCENT TYPE OF SERVICE UNITS REPORTED CY 2012 CY 2015 PER 1,000 DIFFERENCE HCBS Unit 3,058,464 5,183,500 2,125,036 69 percent Dental Claims 880 1,161 281 32 percent Transportation Claims 617 793 176 29 percent Primary Care Physician Claims 3,728 4,517 789 21 percent Vision Claims 326 351 26 8 percent Outpatient Non-ER Claims 1,794 1,945 151 8 percent Pharmacy Prescriptions 9,859 10,328 469 5 percent FQHC/RHC Claims 855 876 21 2 percent Nursing Facility Days 336,732 328,593-8,139-2 percent Outpatient ER Claims 762 746-16 -2 percent Behavioral Health Claims 5,151 4,990-161 -3 percent Inpatient Days 1,189 851-338 -28 percent Source: Leavitt Partners analysis of KFMC reports. Note: KFMC defines utilization per 1,000 as Units Reported/Member Months x 12,000, illustrating the services used per 1,000 beneficiaries in a 12-month period. There is no indication as to whether the definition of HCBS units or the population for which utilization is calculated have changed over time, so the comparison of HCBS utilization in 2012 and 2015 may not be entirely accurate. Dollar value comparisons after 2013 are not possible from the data in KFMC reports. REVIEW OF KANCARE: COST AND UTILIZATION 6
ECONOMIC IMPACT NEW FEDERAL DOLLARS IN KANSAS Kansas requires the MCOs to pay a privilege fee, which was 1 percent of premium revenue at the beginning of the program and was raised to 3.31 percent of premium revenue starting January 1, 2015, and will continue at 5.77 percent starting January 1, 2018. With these funds the state is able to draw down matching federal funds from CMS at the state s Federal Medical Assistance Percentage (just under 56 percent in 2016). As a result, new federal money has flowed into the Kansas economy since the beginning of KanCare. In addition to the direct increase in spending in the state economy, this money adds an indirect economic benefit to the state s economy as other industries in the state respond to the increased activity. Using economic impact analysis methods, we estimate that in 2016, the new federal spending in Kansas totaling about $58 million translates to total increases of economic output between $108 million and $112 million in that year alone. b The analysis also indicates that as part of this total effect, payroll in the state increased by between $27 million and $41 million, and more than 548 jobs were added in all industries in the state. had increased 21 percent and 2 percent from 2012. The number of pharmacy claims increased 5 percent from 2012 to 2015, and nursing facility, outpatient ER, and inpatient utilization all decreased over that period. The category with by far the largest increase in utilization was home- and community-based services (HCBS), with a unit increase of 69 percent from 2012 to 2015. Additionally, sizable shifts in utilization from hospitals and institutional settings to ambulatory and community settings have occurred since KanCare was adopted, with significant reductions in total costs in those areas. These utilization changes could be one factor behind the lowerthan-forecasted program costs for KanCare. CONCLUSIONS This report focused on the economics of KanCare over the program s four completed operating years from 2013 to 2016, specifically its performance on slowing Medicaid cost growth and improving beneficiaries utilization patterns. In its proposal for the program, the state hypothesized that it could reduce costs by holding MCOs to outcomes and performance measures, as well as emphasizing health, wellness, and early detection. During KanCare s four completed years of operation, Medicaid program expenditures appear to have been lower than they would have been without the program. The state s actual costs under KanCare have been lower than projections made in 2012 for scenarios both with and without the program. Costs increased for some groups of members but decreased for others, and the total inflation-adjusted cost per enrollee has declined under KanCare from $7,500 annually in 2013 to $6,756 annually in 2016. During the KanCare years, the state beat the national cost trend; Kansas Medicaid expenditures grew at a lower rate than Medicaid expenditures nationally from 2013 to 2016. Additionally, the three KanCare MCOs have brought direct and indirect economic activity into the state by drawing down federal funds for Medicaid. Beneficiaries utilization patterns shifted during KanCare in ways that could explain these cost savings. While KanCare was in place, there was a decrease in utilization of hospital-based facilities and nursing facilities and an increase in community-based care. Additionally, expenditure growth decreased for some very expensive types of health care. Primary care visits also increased during the program, potentially reducing health care costs down the road by helping beneficiaries better prevent and manage disease. b Economic impact analysis uses multipliers to calculate the total effect of an externally induced shock to the economy, in this case an influx of outside money from CMS. The multipliers are the RIMS Type II multipliers for the state of Kansas for all industries, which were calculated using 2007 industry relationships and 2015 data. For calculation purposes, the direct spending increase is treated as if it all occurred in a single year in nominal dollars and without discounting, but the impacts reported should be interpreted as the total impact for the years 2013-2016 inclusive. It is not clear what industry(ies) should be treated as receiving the direct effect of the CMS funding, so we report ranges using the highest and lowest plausible industries. The lowest plausible multipliers are in the insurance carriers and other related activities industry, and the highest are in the ambulatory health care services industry. The true impact is likely somewhere in between the estimates based on these two industries. REVIEW OF KANCARE: COST AND UTILIZATION 7
BIBLIOGRAPHY 1. Kansas Department of Health and Environment Division of Health Care Finance. Quarterly Report to CMS Regarding Operation of 1115 Waiver Demonstration Program - Quarter Ending 3.31.13. 2. Kansas Department of Health and Environment Division of Health Care Finance. Annual Report to CMS Regarding Operation of 1115 Waiver Demonstration Program Year Ending 12.31.2013, p.2. 3. Kansas Department of Health and Environment. KanCare Update to Robert G. (Bob) Bethell KanCare Oversight [Internet]. 2017 Aug 22. Available from: http://www.kslegislature.org/li/b2017_18/committees/ctte_jt_robert_g_bob_bethell_joint_committee_1/ documents/testimony/20170822_39.pdf. 4. Social Determinants of Health: How Much Do We Understand? [Internet]. [cited 2017 Aug 17]. Available from: http://catalyst. nejm.org/social-determinants-of-health-how-much-understand/ REVIEW OF KANCARE: COST AND UTILIZATION 8
leavittpartners.com Copyright 2017