Table of Contents s39

Table of Contents

1 1. Medicare Advantage: Overview for 2017

Medicare beneficiaries were given expanded access to a variety of health plan choices in addition to traditional fee-for-service Medicare under the Balanced Budget Act of 1997 (BBA). In addition to “original” fee-for-service Medicare, beneficiaries can select from (a) coordinated care plans which include health maintenance organizations (HMOs), provider sponsored organizations (PSOs), preferred provider organizations (PPOs), and plans of religious fraternal benefit (RFP) societies; (b) medical savings accounts; and (c) private fee-for-service plans.

BBA also required the development of a new payment methodology for the various managed care plans – a method of risk adjustment of payment rates that accounts for variations in per capita costs based on health status. Implementation of the risk-adjusted payment methodology began January 1, 2000. The BBA timelines for data collection indicated that initially risk adjustment of Medicare+Choice (now Medicare Advantage) payments should be based only on data from enrollees’ inpatient hospital stays, with later implementation of risk adjustment based on data from additional sites of care.

The Medicare Prescription Drug, Improvement and Modernization Act of 2003 (MMA) included a provision which changed Medicare+Choice into the Medicare Advantage program. The MMA links managed care rates and local fee-for-service costs, a link that had been broken by the BBA for Medicare+Choice plans.

Under section 1853(C)(1) of the Act, Medicare+Choice payments were based on the highest of three amounts: a “blended rate,” a “floor” amount, and a two percent increase over the prior year’s rate. The MMA mandated that for 2004, a fourth amount of 100 percent of projected fee-for-service Medicare be added to the prior Medicare+Choice payment methodology. For the years after 2004, the Secretary of Health and Human Services is required to recalculate 100 percent of the fee-for-service Medicare costs at least every three years – so that at least every three years the Medicare Advantage (MA) capitation rate will be the higher of the fee-for-service rate and the minimum increase rate. In addition, for 2004 and subsequent years, the MMA modified the minimum increase to be the larger of:

1) 102 percent of the previous year’s rate, or

2) An increase by the Medicare growth percentage over the previous year’s rate.

Section 5301 of the Deficit Reduction Act of 2005 (DRA) created a single rate book for calculating MA payments and applicable adjustments. Beginning in 2007, the statute requires that the previous year’s benchmarks be updated annually using the national per capita MA growth percentage. The DRA eliminated the two percent minimum update for all but the end stage renal disease (ESRD) MA rates.

In 2007, CMS moved to 100 percent of the MA payment rate based on risk-adjustment models – marking the completion of a seven-year transition from the prior demographic-based payment system to the risk-adjustment payment system. Risk adjustment ensures that plans are paid higher amounts for sicker beneficiaries. Beginning in 2009, CMS uses more recent fee-for- service diagnosis and claims data to update the relative risk factors in the CMS-HCC model, which is used to produce risk scores for all Medicare beneficiaries. The revised model incorporates updates to the community and institutional components, as well as a new enrollee segment. This model uses enrollees’ demographics and medical conditions collected into 70 hierarchical condition categories (HCCs) to predict their costliness.

Health plans that participate in the Medicare Advantage program receive monthly capitation payments for each Medicare enrollee. Each capitated payment is the product of two general parts:

 A base rate which reflects the payment if a Medicare Advantage enrollee has the health status of the national average beneficiary, and  A risk score which indicates how costly the enrollee is expected to be relative to the national average beneficiary.

The purpose of the risk scores is to adjust Medicare Advantage payments so that they accurately reflect how much each Medicare Advantage enrollee would be expected to cost.

CMS has implemented a number of changes in the Medicare Advantage payment methodology beginning in CY2012 as a result of payment changes enacted in the Affordable Care Act, including the following:

 a new blended benchmark as the county rate;  the new methodology used to derive the new blended benchmark county rates;  identify the qualifying bonus counties;  how to determine transitional phase-in periods; and  the applicability of the star system in the rebates.

For each county, CMS sets the Medicare Advantage benchmark. A Medicare Advantage plan’s payment from Medicare is based on how its bid compares with the local benchmark, which represents the maximum amount Medicare will pay to a plan in a given area on behalf of a Medicare Advantage enrollee. The plan’s bid reflects its costs to provide the Part A and Part B benefit package for a beneficiary of average health status and includes plan administrative cost and profit. The local Medicare Advantage benchmark represents a bidding target and is set using statutory formulas and adjusted for the plan’s quality ranking. If a plan’s bid is above the benchmark, then the plan receives a payment equal to the benchmark and the enrollees have to pay a premium, along with the Part B premium, that equals the difference between the bid and the benchmark. If a plan’s bid is at the benchmark, then the payment equals the benchmark. If a plan’s bid is below the benchmark, the plan receives a payment equal to its bid plus a rebate. The rebate is a fixed percentage – 50, 65, or 70 percent, depending on the plan’s quality ranking – of the difference between the plan’s bid and the benchmark. Once the rebate dollars are determined, the plan must return the rebate to its enrollees in the form of supplemental benefits or lower premiums. Although payments in 2017 will vary by plan based on its location and star rating, overall CMS estimates the net change to plan payments between 2016 and 2017 to be +3.05 percent. Key components of the 2017 Medicare Advantage rates include:

 Revision to the risk-adjustment model to better account for fully dual eligible beneficiaries.  Adjustment to certain measures within the Star Ratings program for socioeconomic status.  Update to the blended risk scores using 75 percent of the risk score calculated using data from the Risk Adjustment Processing System (RAPS) and 25 percent of the risk score calculated using encounter data.  Incentives to Medicare Advantage organizations to offer plans with lower maximum out- of-pocket limits.  Adjustment for coding intensity of 5.66 percent, which results in an incremental 0.25 percent reduction in payments.

Under the Affordable Care Act, beginning in 2012, the benchmarks are tied to a percentage of average fee-for-service spending. A county’s average fee-for-service spending relative to all other counties will determine whether the county’s benchmarks will be set at 95, 100, 107.5, or 115 percent of average fee-for-service spending. As a result, the benchmark will be lower than fee-for-service spending in relatively high spending areas and higher than fee-for- service spending in relatively low spending areas. CMS’ Office of the Actuary expects that under the revised methodology plans will receive smaller rebates and, in turn, have less to spend on additional benefits used to attract beneficiaries.

Beginning in 2012, CMS will apply a Quality Bonus Payment (QBP) percentage in calculating the blended benchmark and will not cap the blended rate at the level of the pre- Affordable Care Act rate for plans with three to five stars on Medicare Compare.

Stars Rating QBP Percentage for QBP Percentage for QBP Percentage for 2012/2013 2014 2015-2017 Less than 3 stars 0 0 0 3 stars 3 3 0 3.5 3.5 3.5 0 4 stars 4 5 5 4.5 stars 4 5 5 5 stars 5 5 5

Beginning in calendar year 2015 payments, plans with less than four stars will no longer receive QBP percentage increases to the county rates and plans with four or more stars will receive QBP percentage increases to the county rates.

For 2017, CMS will update the MA coding adjustment factor to the statutory minimum of 5.66 percent. In order to properly pay health plans that enroll sicker-than-average beneficiaries, CMS adjusts payments to MA plans for the relative risk of their enrolled population. CMS uses claims data from fee-for-service to estimate the risk adjustment model and applies this model to diagnostic data submitted by MA plans. While this approach is a significant improvement over previous risk adjustment methodologies which relied largely on demographic data, it can result in higher payments for the same levels of risk when MA plans submit diagnoses more comprehensively than is done in fee-for-service. The higher level of reported diagnoses can arise for a variety of reasons including plans seeking to better understand the health status of their employees so they can provide better care to plans reporting more diagnoses for enrollees to generate higher revenue. While motivation behind the higher level of reported diagnoses can vary, the effect on the risk adjustment model is the same. That is, higher levels of diagnostic documentation generally lead to higher payments than would be the case if MA plans coded like fee-for-service providers.

As enrollment in MA grows, the imperative for CMS to adjust payments appropriately in response to greater coding intensity will grow. MA coding has the potential to threaten the solvency for the Hospital Insurance (HI) trust fund and lead to higher Part B premiums for all Medicare enrollees, not just MA enrollees. For example, a one percent increase in MA risk scores due to differential coding for 2017 would increase government payments to plans by about $2 billion. For this reason, CMS continues to monitor coding intensity closely and will utilize its authority to increase the coding intensity offset as appropriate. 2. Medicare Advantage Payment Rates: Background

Medicare managed healthcare options have been available to some Medicare beneficiaries since 1982 and Medicare has paid health plans a monthly per person county rate. The monthly per person, or “per capita,” county rates were determined under the Adjusted Average Per Capita Cost methodology (AAPCC).

AAPCCs are five-year moving averages of per beneficiary spending at the county level for fee-for-service Medicare. Because it was assumed that expenditures for enrollees in managed care would be lower due to care coordination efficiencies, the Medicare estimated fee-for-service costs were discounted by five percent. Separate AAPCCs were established for enrollees on the basis of age, sex, Medicaid eligibility, work status, and whether the individual was institutionalized. Medicare’s costs in paying claims for beneficiaries with end-stage renal disease (ESRD) were not considered in these county estimates. ESRD enrollees are handled on a statewide basis. The direct link between AAPCCs and payments created perceptions of geographic inequity as plans were more likely to serve counties with high AAPCCs and typically offered more comprehensive benefits than fee-for service Medicare or counties with lower rates.

BBA established a new payment methodology for HMOs and other plans under the Medicare+Choice program. This new methodology eliminated the direct link in the AAPCC method between managed care rates and local fee-for-service costs. Under the new methodology, the county rate was based on the greater of:

1) A blended local and national rate;

2) A minimum payment floor; or

3) A minimum two percent increase over the prior year’s rate, with the 1997 rate serving as the baseline.

MMA mandated that for 2004, a fourth amount of 100 percent of projected fee-for- service Medicare be added to the prior Medicare+Choice payment methodology. For the years after 2004, the Secretary of Health and Human Services is required to recalculate 100 percent of the fee-for-service Medicare costs at least every three years – so that at least every three years the MA capitation rate will be the higher of the fee-for-service rate and the minimum increase rate. In addition, for 2004 and subsequent years, MMA modified the minimum increase to be the larger of:

1) 102 percent of the previous year’s rate; or

2) An increase by the Medicare growth percentage over the previous year’s rate.

The county rates were used as the base rates for paying MA plans in 2004 and 2005. Beginning in 2006, however, the county rates were used to create benchmarks against which the plans will bid. Rates were set through competitive bid, rather than administered pricing. The benchmark for each plan was a weighted average of the county rates for the counties in the plan’s service area. The weights were the projected enrollment from the counties in the plan’s service area. The bid included plan administrative cost and profit. CMS based the Medicare payment for a private plan on the relationship between the bid and its benchmark.

Plan payment rates are now determined by the plan bid – the dollar amount the plan estimates will cover the Part A and Part B benefit package for a beneficiary of average health status – and the payment area’s benchmark – the maximum amount of Medicare payment set by law for a Medicare Advantage plan to provide Part A and Part B benefits. Plans with higher quality ratings are rewarded with a higher benchmark. If a plan’s bid is above the benchmark, its Medicare Advantage payment rate is equal to the benchmarks, and enrollees have to pay a premium equal to the difference.

If a plan’s bid is below the benchmark, its payment rate is its bid plus a percentage of the difference between the plan’s bid and the benchmark. The beneficiary pays no premium to the plan for the Part A and Part B benefits, but continues to be responsible for payment of the Medicare Part B premium and may pay premiums to the plan for additional benefits. The payment amount above the bid is referred to as the rebate and must be used by the plan to provide additional benefits to the enrollees in the form of lower cost sharing, lower premiums, or supplemental benefits.

Medicare payments are also based on enrolled beneficiaries’ demographics and health risk characteristics. Section 1853(a)(3) of the Social Security Act required the Secretary to develop and implement a new risk-adjustment methodology to be used to adjust the countywide rates to reflect the expected relative health status of each enrollee. The purpose of risk adjustment is to use health status indicators to improve the accuracy of payments and establish incentives for plans to enroll and treat less healthy Medicare beneficiaries. The base payment for an enrollee is the base rate for the enrollee’s county of residence, multiplied by the enrollee’s risk measure.

CMS selected the Principal Inpatient Diagnostic Cost Group (PIP-DCG) model as the risk-adjustment method to be implemented in 2000. This model recognized diagnoses for which inpatient care is most frequently appropriate and which are predictive of higher future costs. Until October 2002, plans submitted to CMS encounter data on enrollees’ inpatient hospitalizations only, which CMS processed through the PIP-DCG model to calculate plan payments.

To assist managed care organizations, CMS provided for a gradual phase-in of the effects of risk adjustment, initially adjusting only a portion of the total payment based on the new PIP- DCG methodology, with the remainder still adjusted under the pre-BBA AAPCC method based only on demographic information. This element of the risk-adjustment methodology provided a safeguard against disruptive changes in payments to the plans. For 2000, the transition schedule called for basing 90 percent of the prospective monthly payment per enrollee on the demographic-only method and 10 percent on the PIP-DCG risk-adjustment method. The demographic-only portion of the payment was adjusted for age, gender, Medicaid eligibility, institutional status, and working-aged status. The risk-adjusted portion of the payment, under the PIP-DCG model, was adjusted for age, gender, Medicaid eligibility, whether originally entitled to Medicare due to disability, and working-aged status, as well as health status.

The Balanced Budget Refinement Act of 1999 (BBRA) adopted CMS’ “phase-in” concept in statutory language, but altered CMS’ proposed timing by mandating that the 90/10 blend would continue through 2001. Most recently, the Benefits Improvement and Protection Act of 2000 (BIPA) again modified the phase-in schedule to further soften the financial impact of risk adjustment on Medicare managed care organizations. Under BIPA, the 90/10 blend continued through 2003. In 2004, the risk-adjusted portion of the payment increased to 30 percent, in 2005 to 50 percent, and to 75 percent in 2006. Beginning in 2007, the rates were fully risk adjusted.

Although the PIP-DCG model was an improvement over the previous demographic payment model based on the AAPCC, CMS recognized that its utility was limited by only using inpatient hospital data and providing additional payment for only a small number of high-cost, seriously ill beneficiaries. Subsequently, BIPA required the implementation of a model using not only diagnoses from inpatient hospital stays, but also from ambulatory settings beginning in 2004. To prepare for implementation in 2004 of a risk-adjustment model based on multiple sites of care, in 2001, CMS required plans to begin submitting data on enrollees’ hospital outpatient and physician office encounters. Although these encounters would have provided additional data for quality measures and re-estimation of payment weights for future risk-adjustment models, CMS was concerned about the associated administrative data burden. As a result, CMS selected a new risk-adjustment model beginning in 2004, which reduced the administrative burden on the managed care organizations significantly. Only a few data elements are required in addition to diagnoses. 3. The CMS-HCC Risk-Adjustment Model

The CMS-HCC model is a selected significant disease model because it incorporates a selected subset of ICD-9-CM diagnosis codes and places them into approximately 64 disease groups called Hierarchical Condition Categories (HCCs). Each disease group includes conditions that are related clinically and have similar cost implications.

The model is prospective in the sense that it uses diagnosis information from a base year to predict costs and adjust payments for the next year. Models of this type are largely driven by the costs associated with chronic diseases, and they capture the systematic risk (costs) associated with Medicare populations. In applying the CMS-HCC model to risk adjust payments for the MA program, the model is used to determine relative risk factors.

In order to use the model as an adjuster to a base rate, costs must be converted into relative cost factors – i.e., risk-adjustment factors. To create risk-adjustment factors, the dollar coefficients (i.e., the predicted expenditure estimates for the diagnostic and demographic characteristics in the model) are divided by the national average (mean) predicted expenditure for fee-for-service beneficiaries. The predicted national average was $5,129 for 2000, the calibration year of the model. These relative risk factors are used to adjust county ratebook amounts for the relative health status of the individual enrollee. The payment rate for an enrollee is the base rate for the enrollee’s county of residence, multiplied by the enrollee’s risk measure, expressed as the CMS-HCC weight.

The CMS-HCC model includes demographic variables for age and sex, Medicaid eligibility, and originally disabled status. A beneficiary having Medicaid status any one month in the data collection year triggers the Medicaid payment adjustment. There is also an adjustment for working-aged status.

The CMS-HCC model has a modification that distinguishes the community-dwelling Medicare population from the long-term institutionalized (defined as those in institutions more than 90 days) population. Institutional status is recognized in the payment year, rather than the prior year. Place of residence – community versus long-term institutionalized – has been found to be an appropriate predictor for high-cost enrollees.

Once identified as institutionalized, patients remain in this status until discharged home for more than 14 days. Note that the CMS-HCC model has been calibrated separately for the community-dwelling and long-term institutionalized populations, rather than using an add-on factor based on residence.

For the first time, for the 2010 plan year, CMS made a “coding pattern differences adjustment” to MA risk scores, reducing MA payments to account for differences in disease coding patterns between MA organization under Part C and original fee-for-service Medicare. CMS is required by law to adjust MA payments where it finds differences in coding patterns between MA plans and Part A and Part B providers. The MA coding pattern differences adjustment factor will adjust for the growth in MA risk scores that occurs above and beyond the average growth captured in the normalization factor. The adjustment was applied as a uniform 3.41 percentage reduction to all plans’ Part C risk scores in all years from 2010-2013. For 2014, the reduction was 4.91 percent and the reduction is set at 5.16 percent for 2015. For CY016, CMS finalized the proposed policy to make an adjustment reflective of the statutory minimum of 5.41 percent.

For CY2015, CMS used the 2013 CMS-HCC and 2014 CMS-HCC models, blending the resulting risk scores from each model at 67 percent and 33 percent, respectively. In the 2016 Rate Notice, CMS finalized the proposed policy to end the blend, and calculate risk scores entirely based on the 2014 CMS-HCC model. This revised model comprises both an update to the data years used to recalibrate the model, and a clinical update to the disease categories to reflect more recent clinical experience.

In 2017, CMS will implement an updated version of the CMS-HHC risk adjustment model. The updated model will be used in Part C payment for aged/disabled beneficiaries enrolled in MA plans. CMS believes that the revised CMS-HCC model will improve the predictive performance for aged and disabled full benefit dual, partial benefit dual, and non-dual beneficiaries in the community. The updated model results in more appropriate relative weights for the HCCs because the relative weights reflect the disease and expenditure patterns of each of the defined community segments. The revised segments are non-dual aged and non-dual disabled, full benefit dual aged and full benefit disabled, and partial benefit dual aged and partial benefit dual disabled beneficiaries. 4. Clinical Features of the CMS-HCC Model HCCs are disease groups broadly organized into body systems, somewhat analogous to the ICD-9-CM major diagnostic categories. Unlike the ICD-9-CM categories, however, the diagnoses within each disease group are related clinically and in terms of cost to the Medicare program.

While the PIP-DCG model placed an enrollee in only a single cost group based on his/her principal inpatient diagnosis with the greatest cost implications, the CMS-HCC model is structured so that each disease group contributes its incremental predicted cost to payment amounts. Conceptually, disease groups are not mutually exclusive because unrelated disease processes each contribute to the predicted costs of care. The CMS-HCC model uses diagnoses from physician visits and hospital inpatient and outpatient stays to assign each beneficiary to none, one, or more than one disease group. For example, an MA enrollee with heart disease, cerebrovascular disease, and cancer will reflect increments for each of these conditions. This makes the model additive – each additional diagnosis results in an increased payment.

In some cases, however, an additional diagnosis does not trigger an additional payment increment because a more severe diagnosis supersedes a less serious one. That is, the CMS-HCC model also can characterize a beneficiary’s illness level within a disease process. In some disease groups, the diagnoses are clinically related and ranked by severity (cost) in a hierarchy, since the more severe manifestations of a disease process principally define the impact of that disease group on cost.

An example of the above is the diabetes hierarchy. Diabetes diagnoses are organized into four severity groups, ranked from uncomplicated diabetes to diabetes with renal manifestations (highest cost implications). An enrollee may be coded with diagnoses in any or all of the four severity groups, but only the highest code in the hierarchy is used to increase payment for diabetes. There are similar hierarchies among cancer and cardiac diseases. In short, costs are additive across hierarchies and disease groups, but not within hierarchies.

As an example, if a beneficiary triggers Disease Groups 135 (Acute Renal Failure) and 136 (Chronic Kidney Disease (Stage 5)), then DG 136 will be dropped. In other words, payment will always be associated with the HCC in column 1, if a HCC in column 3 also occurs during the same collection period. Therefore, the organization’s payment will be based on HCC 135 rather than HCC 136. 5. Medicare Advantage Payment Methodology

Benchmark (Dffers for regional PPO and local  plans; includes any quality bonus) Compare standard bid and benchmark Standard plan bid 

Adjusted for case mix

Rebate = 0.5 to 0.7 Enrollee basic Plan bid Base rate = (varies by plan start ratings) Payment + premium + = below standard bid X to plan  (Always zero) benchmark (case mix-adjusted benchmark-actual bid) Base CMS-HCC X rate weight Enrollee basic Plan bid premium Adjustment (if any) to CMS below Base rate = (Equal to the payment for case mix factors Payment + + = benchmark benchmark  difference between of individual enrollees to to plan the bid and the maintain fixed enrollee premium benchmark)  

CMS-HCC CMS-HCC

  Patient characteristics: Diagnoses Sex Working aged status Age Medicaid status Disable status Source: MedPAC Payment Basics, October 2013 Table 1: CMS-HCC Model Relative Factors for Community and Institutional Beneficiaries

Community Community Community Community Community Community Variable Description Label Non Dual, Non Dual, FB Dual, FB Dual, PB Dual, PB Dual, Institutional Aged Disabled Aged Disabled Aged Disabled Female

0-34 Years - 0.244 - 0.318 - 0.344 1.031

35-44 Years - 0.303 - 0.306 - 0.383 0.999

45-54 Years - 0.322 - 0.338 - 0.374 1.007

55-59 Years - 0.350 - 0.388 - 0.371 0.986

60-64 Years - 0.411 - 0.449 - 0.395 1.028

65-69 Years 0.312 - 0.425 - 0.341 - 1.200

70-74 Years 0.374 - 0.511 - 0.406 - 1.092

75-79 Years 0.448 - 0.611 - 0.484 - 0.995

80-84 Years 0.537 - 0.739 - 0.552 - 0.860

85-89 Years 0.664 - 0.917 - 0.678 - 0.749

90-94 Years 0.797 - 1.037 - 0.817 - 0.626

95 Years or Over 0.816 - 1.094 - 0.913 - 0.456 Male 0-34 Years - 0.155 - 0.225 - 0.330 1.049

35-44 Years - 0.190 - 0.204 - 0.267 1.074

45-54 Years - 0.221 - 0.281 - 0.300 1.008

55-59 Years - 0.271 - 0.372 - 0.307 1.055

60-64 Years - 0.303 - 0.486 - 0.343 1.039

65-69 Years 0.300 - 0.492 - 0.334 - 1.269

70-74 Years 0.379 - 0.582 - 0.409 - 1.323

75-79 Years 0.466 - 0.692 - 0.491 - 1.331

80-84 Years 0.561 - 0.816 - 0.546 - 1.189

85-89 Years 0.694 - 1.009 - 0.679 - 1.129

90-94 Years 0.857 - 1.186 - 0.822 - 0.964

95 Years or Over 0.976 - 1.268 - 1.038 - 0.781

Medicaid and Originally Disabled

Medicaid ------0.062 Originally Disabled, 0.244 - 0.172 - 0.126 - - Female Originally Disabled, 0.152 - 0.192 - 0.105 - - Male

Disease Coefficients Description Label

HCC1 HIV/AIDS 0.312 0.288 0.585 0.500 0.550 0.232 1.747 Septicemia, Sepsis, Systemic HCC2 Inflammatory Response 0.455 0.532 0.596 0.811 0.409 0.417 0.346 Syndrome/Shock HCC6 Opportunistic Infections 0.435 0.704 0.548 0.919 0.482 0.765 0.580 Metastatic Cancer and Acute HCC8 2.625 2.644 2.542 2.767 2.442 2.582 1.143 Leukemia Community Community Community Community Community Community Variable Description Label Non Dual, Non Dual, FB Dual, FB Dual, PB Dual, PB Dual, Institutional Aged Disabled Aged Disabled Aged Disabled Female

HCC9 Lung and Other Severe Cancers 0.970 0.927 0.973 1.025 0.955 0.879 0.727

HCC10 Lymphoma and Other Cancers 0.677 0.656 0.713 0.761 0.667 0.577 0.401 Colorectal, Bladder, and Other HCC11 0.301 0.352 0.332 0.361 0.325 0.400 0.293 Cancers Breast, Prostate, and Other Cancers HCC12 0.146 0.202 0.159 0.190 0.152 0.182 0.199 and Tumors HCC17 Diabetes with Acute Complications 0.318 0.371 0.346 0.431 0.354 0.423 0.441

HCC18 Diabetes with Chronic Complications 0.318 0.371 0.346 0.431 0.354 0.423 0.441 HCC19 Diabetes without Complication 0.104 0.128 0.097 0.160 0.098 0.136 0.160 HCC21 Protein-Calorie Malnutrition 0.545 0.753 0.752 0.845 0.562 0.709 0.260 HCC22 Morbid Obesity 0.273 0.227 0.410 0.373 0.244 0.242 0.511 Other Significant Endocrine and HCC23 0.228 0.444 0.228 0.353 0.193 0.351 0.337 Metabolic Disorders HCC27 End-Stage Liver Disease 0.962 1.110 1.242 1.349 0.889 0.963 0.962 HCC28 Cirrhosis of Liver 0.390 0.394 0.342 0.491 0.460 0.324 0.390

HCC29 Chronic Hepatitis 0.165 0.267 0.038 0.400 0.263 0.324 0.390

HCC33 Intestinal Obstruction/Perforation 0.246 0.524 0.369 0.503 0.324 0.510 0.335

HCC34 Chronic Pancreatitis 0.276 0.678 0.333 0.875 0.412 0.849 0.241 HCC35 Inflammatory Bowel Disease 0.294 0.483 0.334 0.613 0.209 0.496 0.244 Bone/Joint/Muscle HCC39 0.425 0.474 0.552 0.713 0.418 0.491 0.345 Infections/Necrosis Rheumatoid Arthritis and HCC40 Inflammatory Connective Tissue 0.423 0.377 0.370 0.345 0.390 0.290 0.329 Disease HCC46 Severe Hematological Disorders 1.388 3.188 1.219 4.256 1.226 3.529 0.680

HCC47 Disorders of Immunity 0.625 0.848 0.529 0.589 0.449 0.690 0.529 Coagulation Defects and Other HCC48 0.221 0.339 0.268 0.378 0.225 0.382 0.151 Specified Hematological Disorders HCC54 Drug/Alcohol Psychosis 0.383 0.569 0.706 0.919 0.388 0.613 0.102

HCC55 Drug/Alcohol Dependence 0.383 0.285 0.522 0.366 0.377 0.286 0.102

HCC57 Schizophrenia 0.608 0.395 0.612 0.432 0.547 0.366 0.271 Major Depressive, Bipolar, and HCC58 0.395 0.209 0.444 0.178 0.413 0.163 0.271 Paranoid Disorders HCC70 Quadriplegia 1.314 1.053 1.098 1.056 1.274 1.328 0.497 HCC71 Paraplegia 1.007 0.704 0.920 1.019 0.958 0.908 0.467 HCC72 Spinal Cord Disorders/Injuries 0.528 0.456 0.552 0.407 0.556 0.384 0.229 Amyotrophic Lateral Sclerosis and HCC73 0.970 1.082 1.230 1.219 0.570 0.814 0.224 Other Motor Neuron Disease HCC74 Cerebral Palsy 0.280 0.132 - - 0.158 0.052 - Myasthenia Gravis/Myoneural Disorders and Guillain-Barre HCC75 0.457 0.528 0.436 0.465 0.364 0.331 0.369 Syndrome/Inflammatory and Toxic Neuropathy HCC76 Muscular Dystrophy 0.505 0.457 0.553 0.512 0.429 0.168 0.104

HCC77 Multiple Sclerosis 0.441 0.540 0.687 0.794 0.407 0.459 - Community Community Community Community Community Community Variable Description Label Non Dual, Non Dual, FB Dual, FB Dual, PB Dual, PB Dual, Institutional Aged Disabled Aged Disabled Aged Disabled Female Parkinson’s and Huntington’s HCC78 0.674 0.585 0.751 0.516 0.629 0.394 0.145 Diseases HCC79 Seizure Disorders and Convulsions 0.309 0.227 0.357 0.195 0.349 0.245 0.088 Coma, Brain Compression/Anoxic HCC80 0.584 0.302 0.946 0.324 0.508 0.155 0.042 Damage Respirator HCC82 1.055 1.024 2.304 1.575 0.914 0.676 1.631 Dependence/Tracheostomy Status HCC83 Respiratory Arrest 0.658 0.781 1.033 0.484 0.704 0.429 0.727 Cardio-Respiratory Failure and HCC84 0.302 0.578 0.471 0.484 0.301 0.429 0.297 Shock HCC85 Congestive Heart Failure 0.323 0.412 0.355 0.415 0.320 0.367 0.191

HCC86 Acute Myocardial Infarction 0.233 0.306 0.473 0.618 0.280 0.438 0.497 Unstable Angina and Other Acute HCC87 0.218 0.306 0.336 0.618 0.280 0.438 0.497 Ischemic Heart Disease HCC88 Angina Pectoris 0.140 0.121 0.068 0.205 0.175 0.220 0.497

HCC96 Specified Heart Arrhythmias 0.268 0.284 0.369 0.377 0.283 0.258 0.224

HCC99 Cerebral Hemorrhage 0.263 0.282 0.474 0.690 0.278 0.280 0.114

HCC100 Ischemic or Unspecified Stroke 0.263 0.195 0.474 0.357 0.270 0.232 0.114

HCC103 Hemiplegia/Hemiparesis 0.538 0.324 0.548 0.435 0.603 0.401 0.031 Monoplegia, Other Paralytic HCC104 0.395 0.258 0.374 0.381 0.559 0.401 0.031 Syndromes Atherosclerosis of the Extremities HCC106 1.461 1.506 1.744 1.740 1.452 1.601 0.884 with Ulceration or Gangrene

HCC107 Vascular Disease with Complications 0.400 0.486 0.540 0.756 0.443 0.549 0.321

HCC108 Vascular Disease 0.298 0.333 0.324 0.319 0.316 0.326 0.094

HCC110 Cystic Fibrosis 0.620 2.538 0.985 3.365 0.358 2.861 0.305 Chronic Obstructive Pulmonary HCC111 0.328 0.262 0.422 0.354 0.358 0.293 0.305 Disease Fibrosis of Lung and Other Chronic HCC112 0.209 0.262 0.134 0.322 0.172 0.174 0.057 Lung Disorders Aspiration and Specified Bacterial HCC114 0.599 0.530 0.707 0.490 0.666 0.373 0.067 Pneumonias Pneumococcal Pneumonia, HCC115 0.221 0.128 0.162 0.049 0.302 0.220 0.067 Empyema, Lung Abscess Proliferative Diabetic Retinopathy HCC122 0.217 0.171 0.223 0.284 0.276 0.195 0.460 and Vitreous Hemorrhage HCC124 Exudative Macular Degeneration 0.499 0.385 0.278 0.090 0.336 0.115 0.228

HCC134 Dialysis Status 0.422 0.500 0.672 0.637 0.435 0.512 0.462

HCC135 Acute Renal Failure 0.422 0.500 0.672 0.637 0.435 0.512 0.462 HCC136 Chronic Kidney Disease, Stage 5 0.237 0.141 0.244 0.167 0.184 0.147 0.436 Chronic Kidney Disease, Severe HCC137 0.237 0.141 0.244 0.079 0.184 0.035 0.202 (Stage 4) Pressure Ulcer of Skin with Necrosis HCC157 2.163 2.203 2.879 2.626 2.274 2.655 0.924 Through to Muscle, Tendon, or Bone Pressure Ulcer of Skin with Full HCC158 1.204 1.393 1.576 1.559 1.074 1.237 0.295 Thickness Skin Loss Community Community Community Community Community Community Variable Description Label Non Dual, Non Dual, FB Dual, FB Dual, PB Dual, PB Dual, Institutional Aged Disabled Aged Disabled Aged Disabled Female Chronic Ulcer of Skin, Except HCC161 0.535 0.636 0.757 0.631 0.586 0.620 0.294 Pressure HCC162 Severe Skin Burn or Condition 0.321 0.348 0.003 0.537 0.525 0.119 0.076

HCC166 Severe Head Injury 0.584 0.302 0.946 0.324 1.065 0.155 0.042

HCC167 Major Head Injury 0.191 0.044 0.274 0.171 0.133 0.049 - Vertebral Fractures without Spinal HCC169 0.495 0.456 0.552 0.407 0.516 0.384 0.209 Cord Injury HCC170 Hip Fracture/Dislocation 0.418 0.513 0.520 0.668 0.383 0.484 - Traumatic Amputations and HCC173 0.266 0.340 0.412 0.383 0.233 0.232 0.267 Complications Complications of Specified HCC176 0.597 0.871 0.721 1.156 0.584 0.876 0.502 Implanted Device or Graft Major Organ Transplant or HCC186 1.000 0.618 0.816 1.075 0.795 0.655 0.962 Replacement Status Artificial Openings for Feeding or HCC188 0.571 0.785 0.775 0.870 0.579 0.867 0.500 Elimination Amputation Status, Lower HCC189 0.588 0.455 0.787 1.065 0.737 0.696 0.407 Limb/Amputation Complications

Disease Interactions Description Label

HCC47 CANCER Immune Disorders* Cancer Group 0.893 0.675 0.815 0.652 0.776 0.808 - HCC85 DIABETES Congestive Heart Failure* Diabetes MELLIT Group 0.154 0.096 0.205 0.160 0.178 0.139 0.154 Congestive Heart Failure* Chronic HCC85 COPDCF Obstructive Pulmonary Disease 0.190 0.174 0.240 0.217 0.186 0.181 0.164 Group Congestive Heart Failure* Renal HCC85 RENAL Group 0.270 0.493 0.271 0.711 0.299 0.609 - Cardiorespiratory Failure Group* RESPDEPANDARR Chronic Obstructive Pulmonary 0.336 0.256 0.564 0.524 0.460 0.449 0.423 E COPDCF Disease Group Congestive Heart Failure* Specified HCC85 HCC96 0.105 0.285 0.200 0.405 0.116 0.318 Heart Arrhythmias - SUBSTANCE Substance Abuse Group* ABUSE Psychiatric Group - 0.191 - 0.233 - 0.230 - PSYCHIATRIC

SEPSIS PRESSURE Sepsis* Pressure Ulcer ULCER ------0.252

SEPSIS ARTIF Sepsis* Artificial Openings for OPENINGS Feeding or Elimination ------0.568

ART OPENINGS Artificial Openings for Feeding or PRESSURE ULCER Elimination* Pressure Ulcer ------0.331

Chronic Obstructive Pulmonary COPDCF ASP SPEC Disease* Aspiration and Specified 0.254 BACT PNEUM Bacterial Pneumonias ------

ASP SPEC BACT Aspiration and Specified Bacterial PNEUM PRES ULC Pneumonias* Pressure Ulcer ------0.366

SEPSIS ASP SPEC Sepsis* Aspiration and Specified BACT PNEUM Bacterial Pneumonias ------0.321 Community Community Community Community Community Community Variable Description Label Non Dual, Non Dual, FB Dual, FB Dual, PB Dual, PB Dual, Institutional Aged Disabled Aged Disabled Aged Disabled Female

SCHIZOPHRENIA Schizophrenia* Chronic Obstructive COPDCF Pulmonary Disease ------0.363

SCHIZOPHRENIA Schizophrenia* Congestive Heart CHF Failure ------0.173

SCHIZOPHRENIA Schizophrenia* Seizure Disorders SEIZURES and Convulsions ------0.483

Disabled/Disease Interactions

DISABLED HCC85 Disabled, Congestive Heart Failure ------0.321 DISABLED Disabled, Pressure Ulcer PRESSURE ULCER ------0.608

Disabled, Chronic Ulcer of the DISABLED HCC161 Skin, Except Pressure Ulcer ------0.369 Disabled, Bone/Joint Muscle DISABLED HCC39 Infections/Necrosis ------0.567 DISABLED HCC77 Disabled, Multiple Sclerosis ------0.425 DISABLED HCC6 Disabled, Opportunistic Infections ------0.277

NOTES: The denominator is $9,185.29 In the “disease interactions” and “disabled interactions,” the variables are defined as follows: Immune Disorders = HCC 47 Cancer = HCCs 8-12 Congestive Heart Failure = HCC 85 Diabetes = HCCs 17-19 Chronic Obstructive Pulmonary Disease = HCCs 110-112 Renal = HCCs 134 – 137 Cardiorespiratory Failure = HCCs 82-84 Specified Heart Arrhythmias = HCC 96 Substance Abuse = HCCs 54-55 Psychiatric = HCCs 57-58 Sepsis = HCC 2 Pressure Ulcer = HCCs 157-158 Artificial Openings for Feeding or Elimination = HCC 188 Aspiration and Specified Bacterial Pneumonias = HCC 114 Schizophrenia = HCC 57 Seizure Disorders and Convulsions = HCC 79 Chronic Ulcer of Skin, except Pressure = HCC 161 Bone/Joint/Muscle Infections/Necrosis = HCC 39 Multiple Sclerosis = HCC 77 Opportunistic Infections = HCC 6 SOURCE: RTI International analysis of 2013-2014 Medicare 100% data and RTI International analysis of 2013-2014 Medicare 100% institutional sample. 6. Table 2: Disease Hierarchies for the Revised CMS-HCC Model

Hierarchical If the Disease Group is Listed in this column… …Then drop the Disease Condition Group(s) listed in this Category column (HCC) Hierarchical Condition Category (HCC) LABEL 8 Metastatic Cancer and Acute Leukemia 9,10,11,12 9 Lung and Other Severe Cancers 10,11,12 10 Lymphoma and Other Cancers 11,12 11 Colorectal, Bladder, and Other Cancers 12 17 Diabetes with Acute Complications 18,19 18 Diabetes with Chronic Complications 19 27 End-Stage Liver Disease 28,29,80 28 Cirrhosis of Liver 29 46 Severe Hematological Disorders 48 54 Drug/Alcohol Psychosis 55 57 Schizophrenia 58 70 Quadriplegia 71,72,103,104,169 71 Paraplegia 72,104,169 72 Spinal Cord Disorders/Injuries 169 82 Respirator Dependence/Tracheostomy Status 83,84 83 Respiratory Arrest 84 86 Acute Myocardial Infarction 87,88 87 Unstable Angina and Other Acute Ischemic Heart 88 Disease 99 Cerebral Hemorrhage 100 103 Hemiplegia/Hemiparesis 104 106 Atherosclerosis of the Extremities with Ulceration or 107,108,161,189 Gangrene 107 Vascular Disease with Complications 108 110 Cystic Fibrosis 111,112 111 Chronic Obstructive Pulmonary Disease 112 114 Aspiration and Specified Bacterial Pneumonias 115 134 Dialysis Status 135,136,137 135 Acute Renal Failure 136,137 136 Chronic Kidney Disease, Stage 5 137 157 Pressure Ulcer of Skin with Necrosis Through to 158,161 Muscle, Tendon, or Bone 158 Pressure Ulcer of Skin with Full Thickness Skin Loss 161 166 Severe Head Injury 80,167 7. Table 3: CMS-HCC Model Relative Factors for Aged and Disabled New Enrollees

Non-Medicaid & Medicaid & Medicaid & Non-Originally Non-Originally Non-Medicaid Originally Disabled Disabled & Originally Disabled Female Disabled 0-34 Years 0.664 0.985 - - 35-44 Years 0.936 1.221 - - 45-54 Years 1.035 1.337 - - 55-59 Years 1.004 1.342 - - 60-64 Years 1.122 1.438 - - 65 Years 0.522 1.059 1.130 1.566 66 Years 0.516 0.946 1.167 1.619 67 Years 0.544 0.946 1.167 1.619 68 Years 0.581 0.946 1.167 1.619 69 Years 0.605 0.946 1.167 1.619 70-74 Years 0.674 0.975 1.167 1.619 75-79 Years 0.892 1.092 1.167 1.619 80-84 Years 1.066 1.395 1.167 1.619 85-89 Years 1.324 1.458 1.167 1.619 90-94 Years 1.324 1.678 1.167 1.619 95 Years or Over 1.324 1.678 1.167 1.619 Male 0-34 Years 0.456 0.766 - - 35-44 Years 0.665 1.095 - - 45-54 Years 0.834 1.357 - - 55-59 Years 0.889 1.422 - - 60-64 Years 0.923 1.582 - - 65 Years 0.514 1.201 0.790 1.613 66 Years 0.533 1.208 0.957 1.613 67 Years 0.575 1.208 1.005 2.202 68 Years 0.641 1.208 1.074 2.202 69 Years 0.671 1.311 1.398 2.202 70-74 Years 0.776 1.311 1.398 2.202 75-79 Years 1.040 1.361 1.398 2.202 80-84 Years 1.270 1.603 1.398 2.202 85-89 Years 1.511 1.850 1.398 2.202 90-94 Years 1.511 1.850 1.398 2.202 95 Years or Over 1.511 1.850 1.398 2.202

NOTES: 1. The denominator is $9,185.29 2. For payment purposes, a new enrollee is a beneficiary who did not have 12 months of Part B eligibility in the data collection year. CMS-HCC new enrollee models are not based on diagnoses, but include factors for different age and gender combinations by Medicaid and the original reason for Medicare entitlement.

SOURCE: RTI International analysis of 2013-2014 100 Percent Medicare data. 8. Table 4: Medicare Advantage Monthly Capitation Rates for 2017 for All Plans Except PACE Plans

Risk Parts A & B County Name 5% 3% 0% ESRD Bonus Rate Bonus Rate Bonus Rate Parts A & B ALACHUA $820.13 $820.13 $820.13 $6,861.75 BAKER $948.87 $934.64 $901.43 $6,861.75 BAY $875.85 $862.71 $832.06 $6,861.75 BRADFORD $858.25 $845.38 $815.34 $6,861.75 BREVARD $869.03 $856.62 $827.65 $6,861.75 BROWARD $922.60 $908.76 $876.47 $6,861.75 CALHOUN $809.45 $809.45 $807.82 $6,861.75 CHARLOTTE $879.13 $865.94 $835.17 $6,861.75 CITRUS $814.25 $814.25 $808.79 $6,861.75 CLAY $921.13 $907.31 $875.07 $6,861.75 COLLIER $846.12 $833.43 $803.81 $6,861.75 COLUMBIA $890.70 $890.70 $890.70 $6,861.75 MIAMI-DADE $1,023.12 $1,007.77 $971.96 $6,861.75 DESOTO $884.98 $871.71 $840.73 $6,861.75 DIXIE $929.24 $915.30 $882.78 $6,861.75 DUVAL $918.90 $905.12 $872.96 $6,861.75 ESCAMBIA $852.75 $840.99 $813.55 $6,861.75 FLAGLER $893.33 $877.14 $819.76 $6,861.75 FRANKLIN $845.23 $833.15 $804.98 $6,861.75 GADSDEN $892.08 $871.63 $821.62 $6,861.75 GILCHRIST $866.30 $866.30 $865.86 $6,861.75 GLADES $901.45 $887.93 $856.38 $6,861.75 GULF $886.38 $878.29 $848.59 $6,861.75 HAMILTON $833.03 $833.03 $833.03 $6,861.75 HARDEE $880.21 $867.33 $837.27 $6,861.75 HENDRY $868.12 $855.10 $824.71 $6,861.75 HERNANDO $889.68 $876.66 $846.28 $6,861.75 HIGHLANDS $868.66 $868.66 $862.94 $6,861.75 HILLSBOROUGH $882.92 $869.68 $838.77 $6,861.75 HOLMES $810.61 $810.61 $791.79 $6,861.75 INDIAN RIVER $866.10 $853.43 $823.86 $6,861.75 JACKSON $809.99 $809.99 $809.99 $6,861.75 JEFFERSON $890.87 $890.87 $870.97 $6,861.75 LAFAYETTE $864.22 $851.88 $823.07 $6,861.75 LAKE $893.43 $886.93 $856.94 $6,861.75 LEE $878.87 $865.69 $834.93 $6,861.75 LEON $859.14 $837.20 $786.02 $6,861.75 LEVY $860.95 $860.95 $860.95 $6,861.75 LIBERTY $805.07 $805.07 $805.07 $6,861.75 MADISON $806.01 $806.01 $806.01 $6,861.75 MANATEE $859.81 $846.91 $816.82 $6,861.75 MARION $860.50 $848.20 $819.52 $6,861.75 MARTIN $884.90 $871.95 $841.74 $6,861.75 MONROE $932.00 $918.02 $885.40 $6,861.75 NASSAU $854.18 $841.37 $811.47 $6,861.75 Risk Parts A & B County Name 5% 3% 0% ESRD Bonus Rate Bonus Rate Bonus Rate Parts A & B OKALOOSA $833.75 $833.75 $810.21 $6,861.75 OKEECHOBEE $975.26 $960.63 $926.50 $6,861.75 ORANGE $867.53 $854.52 $824.15 $6,861.75 OSCEOLA $866.45 $854.07 $825.19 $6,861.75 PALM BEACH $926.81 $912.91 $880.47 $6,861.75 PASCO $894.74 $894.07 $863.84 $6,861.75 PINELLAS $886.22 $872.93 $841.91 $6,861.75 POLK $882.26 $869.66 $840.25 $6,861.75 PUTNAM $865.77 $865.77 $865.77 $6,861.75 ST. JOHNS $890.94 $890.94 $861.19 $6,861.75 ST. LUCIE $864.27 $851.31 $821.06 $6,861.75 SANTA ROSA $844.73 $832.66 $804.50 $6,861.75 SARASOTA $875.22 $862.71 $833.54 $6,861.75 SEMINOLE $872.22 $859.14 $828.61 $6,861.75 SUMTER $815.71 $803.47 $774.92 $6,861.75 SUWANNEE $855.50 $855.50 $846.61 $6,861.75 TAYLOR $875.82 $863.30 $834.11 $6,861.75 UNION $951.78 $937.50 $904.19 $6,861.75 VOLUSIA $891.60 $868.08 $813.21 $6,861.75 WAKULLA $872.26 $849.25 $795.58 $6,861.75 WALTON $851.50 $844.45 $815.89 $6,861.75 WASHINGTON $842.90 $836.65 $808.36 $6,861.75

* Note: 1. ESRD rates are for dialysis only and are shown after the deduction of user fee of $5.25. 2. All rates include MIPPA IME deduction.