Issue Brief. What s in the Stars? Quality Ratings of Medicare Advantage Plans, 2010

Transcription

1 Issue Brief What s in the Stars? Quality Ratings of Medicare Advantage Plans, 00 December 009

2 What s in the Stars? Quality Ratings of Medicare Advantage Plans, 00 The Centers for Medicare and Medicaid Services (CMS) rates the relative quality of the private that are offered to Medicare beneficiaries through the Medicare Advantage program. CMS rates Medicare Advantage on a one to five-star scale, with five stars representing the highest quality. The summary score provides an overall measure of a plan s quality, and is a cumulative indicator of the quality of care, access to care, responsiveness, and beneficiary satisfaction provided by the plan. Ratings are posted on the Medicare website to provide beneficiaries with additional information to help them choose among the Medicare Advantage offered in their area. In October 009, CMS updated the ratings for the 00 open enrollment period. The quality ratings are also used by CMS in its oversight of. In addition, health care reform legislation proposes using these quality ratings to help identify and reward high-quality. The five-star quality scores for Medicare Advantage are derived from four sources: () CMS administrative data on plan quality and member satisfaction, () the Consumer Assessment of Healthcare Providers and Systems (CAHPS ), () the Healthcare Effectiveness Data and Information Set (HEDIS ), and () the Health Outcomes Survey (HOS). The individual measures and the scores comprising the summary score are adjusted for skewness, to avoid a clustering of scores at either the high or low end. The individual measures are also adjusted for patient characteristics, to make the survey results more representative of all beneficiaries enrolled in Medicare Advantage. CMS does not publish quality ratings for a plan if it is missing too much data to calculate a score for a given measure or component of the summary score. Relatively newer are less likely than older to have ratings because of missing data. All offered for the first time in 009 or 00 do not have quality ratings for the 00 plan year, because the would be missing data from the earliest survey, HEDIS, which collected data from January to December 008. Very few offered for the first time in 008 were for 00. Plans with relatively low enrollment are also more likely to be missing data because contracts with less than,000 enrollees are not required to report HEDIS data to CMS. Private Fee-For-Service (PFFS) are more likely than other plan types to be missing data, and therefore not have quality ratings, because they are not required to report HEDIS data until 00, and many PFFS are relatively new. 6 This analysis uses this five-star rating system and additional information from the CMS Plan Directory and enrollment files 7 to examine the variation in quality ratings across the 00 Medicare Advantage, 8 based on the information available to beneficiaries on the Medicare Compare website. The brief also examines the variation in quality ratings across, based on whether they were fined or suspended by CMS due to marketing or other violations. 9 The analysis does not attempt to assess the validity of the quality ratings; instead, it examines the data posted by CMS to consider the implications for consumers and policymakers. The term plan is used in this analysis to mean a Medicare contract because Medicare Advantage organizations report the quality data at the contract level. Most of the findings for 00 are weighted by plan enrollment in each county in 009. All reported differences are significant at the 9 percent confidence level.

3 KEY FINDINGS Enrollment More than half (9 percent) of 00 Medicare Advantage received quality ratings; these account for 8 percent of all beneficiaries in Medicare Advantage, based on 009 enrollment. (Exhibit ) o Conversely, percent of, accounting for just percent of all Medicare Advantage enrollees, did not receive quality ratings. Nearly one in four Medicare Advantage enrollees ( percent) is in a plan with four or more stars. On average, Medicare Advantage received.7 stars. o One in five Medicare Advantage enrollees is in a plan with fewer than three stars. Geographic Differences The share of Medicare Advantage enrollees in that received four or more stars for 00 varies greatly by state, based on 009 enrollment. (Exhibit ) o In the majority of states ( states and DC), less than percent of Medicare Advantage enrollees are in with four or more stars. o More than half of Medicare Advantage enrollees in Massachusetts, Oregon, and Hawaii are enrolled in with four or more stars. California and Pennsylvania account for nearly half (9. percent) of all enrollees of Medicare Advantage with four or more quality stars, but only a quarter (6. percent) of all Medicare Advantage enrollees nationwide (data not shown). Tax Status Not-for-profit Medicare Advantage have significantly higher ratings than for-profit Medicare Advantage. (Exhibit ) o As previously noted, this analysis excludes without ratings in 00; 7 percent of non-profit (accounting for percent of all Medicare Advantage enrollees in non-profit ) and percent of for-profit (accounting for percent of all Medicare Advantage enrollees in for-profit ) did not receive quality ratings. Nearly one in four Medicare Advantage enrollees are in stars or better, 00 Distribution of Medicare Advantage by quality ratings: % 7% Among, not-for-profit Medicare Advantage have higher average ratings than for-profit, 00 Star Rating (out of ).... 6% % 9% % 9% % % % Total Number of Contracts = 9 Number of Rated Contracts = 6 NOTE: The term plan is used in this analysis to mean the Medicare contracting entity because plan ratings are aggregated at the contract level. The analysis of the distribution of by quality ratings is not weighted by enrollment. % % % % 7% 7%.87* Not-for-profit contracts Number of contracts: 86 Enrollees in contracts:.6 million Percent in contracts: 88% Exhibit or more stars Distribution of Medicare Advantage enrollees by quality ratings: Total Enrollment = 0. million Total Enrollment in Rated Contracts = 8.7 million Exhibit Share of Medicare Advantage enrollees in in 009 that received or more stars in 00 7% % 6% % 7% 6% 8% % % 9% % % Exhibit 7% 6% % 8% 9% % %.0* For-profit contracts million 8% 7 7% % % DC NOTE: The term plan is used in this analysis to mean the Medicare contracting entity because plan ratings are aggregated at the contract level. All states had at least plan that received star ratings. NOTE: Differences between tax status are statistically significant at the p<0.0 level. Asterisks (*) denote differences that are significantly different from the mean at the p<0.0 level. 7% 9% 0 with stars (7 states and DC) -% enrollees in star (7 states) 6- enrollees in star (6 states) 0-% enrollees in star (6 states) >% enrollees in star ( states) Stars. Stars Stars. Stars Stars. Stars Stars Missing.7

4 Plan Experience More experienced (with contracts beginning before 00) have higher ratings than first offered on or after January 00. (Exhibit ) o The majority (60 percent) of beneficiaries enrolled in relatively new Medicare Advantage (first offered on or after January 00) are in with quality ratings. It is not clear whether the difference between experienced and newer would narrow or widen if all were. The analysis can not be generalized to all first offered on or after 00 because, as previously noted, most first offered on or after 008 are not for 00. Plan Type PFFS and regional PPOs have below average ratings significantly lower than HMOs and local PPOs. However, more than half ( percent) of PFFS enrollees are in that did not receive quality ratings for 00, as compared to percent of local PPO enrollees, percent of regional PPO enrollees, and percent of HMO enrollees. As a result, findings may not be applicable to all PFFS plan enrollees in 00, particularly those in with contracts beginning on or after January 008. Over time, a larger share of PFFS will receive quality ratings, as reporting requirements phase in. (Exhibit ) o HMOs, local PPOs, one PFFS plan, and zero regional PPOs received four or more stars for 00. Among, older Medicare Advantage have higher average ratings than newer, 00 Star Rating (out of ).....7* Exhibit.00* Older : Newer : Contracts beginning before Contracts beginning after January, 00 January, 00 Number of contracts: 7 09 Enrollees in contracts: 6. million.6 million Percent in contracts: 0 6 NOTE: Differences between age of contract are statistically significant at the p<0.0 level. Asterisks (*) denote differences that are significantly different from the mean at the p<0.0 level. Among, Private Fee-For-Service and Regional Preferred Provider Organizations have below-average ratings, 00 Star Rating (out of ).....* Exhibit..*.69* HMO Local PPO PFFS Regional PPO Number of contracts: 6 68 Enrollees in contracts: 6. million 0.8 million. million 0. million Percent in contracts: 99% 88% % 98% NOTE: Asterisks (*) denote differences that are significantly different from the mean at the p<0.0 level. Differences between HMOs and PFFS and differences between HMOs and regional PPOs are statistically significant at the p<0.0 level; differences between HMOs and local PPOs are not statistically significant..7.7 Parent Organization quality ratings vary widely among the largest organizations offering Medicare Star Rating (out of ) Advantage that were, ranging from an average of.07 stars among. Kaiser Permanente, to an average of.07*.76 stars among Humana..8* (Exhibit 6)..0. o Among most of these large organizations, a relatively small share of Medicare. Advantage enrollees is in un Medicare Advantage. However, for two of the organizations, Coventry and. Aetna, more than half of all Medicare Kaiser Permanente Advantage enrollees (67 percent and 6 percent, respectively) are in without quality ratings for 00. Conversely, just percent of Humana enrollees and zero percent of Kaiser Permanente enrollees are in un. (See Table A) Exhibit 6 Among, quality ratings vary widely across organizations offering Medicare Advantage, 00 Medicare Advantage Organizations with the Highest Enrollment in Rated Plans.*.*.09*.9* Highmark HealthSpring Coventry Health Net Aetna EmblemHealth UnitedHealth Group.8* Wellpoint NOTE: Asterisks (*) denote differences that are significantly different from the mean at the p<0.0 level. CMS only covering % of Coventry s enrollees and 7% of Aetna s enrollees. Other organizations also have many that were not. No statistical comparisons were made between organizations..7.76* Humana

5 Quality ratings also vary among the largest organizations with Medicare Advantage enrollees, by specific type of plan. (See Table A) o Among HMOs, average quality ratings ranged from.07 among Kaiser Permanente to.0 among UnitedHealth Group. Other types of can not be directly compared across the largest organizations because many non-hmo were not due to missing data. Enforcement Actions Quality ratings are lower, on average, among that were fined or suspended by CMS between July 006 and July 009. (Exhibit 7) Changes from 009 to 00 The average number of stars for Medicare Advantage increased modestly between 009 and 00, from.0 to.7. (Exhibit 8) Among, average quality ratings for 00 are lower among Medicare Advantage that were fined or suspended between July 006 and July 009 Among by CMS, average quality rating is higher in 00 than 009 Star Rating (out of ). Star Rating (out of ).....* Exhibit 7.78* Exhibit 8.7* No enforcement action Fined Suspended marketing and enrollment activities Number of contracts: Enrollees in contracts: 6. million.8 million 0. million Percent in contracts: 8% 9 7% NOTE: Differences between no enforcement actions and fines and differences between no enforcement actions and suspensions are statistically significant at the p<0.0 level. Asterisks (*) denote differences that are significantly different from the mean at the p<0.0 level. Ratings were not available for terminated. SOURCE: Kaiser Family Foundation analysis of the 00 Medicare Health Plan Quality and Performance Ratings and CMS enforcement actions against Medicare Advantage organizations and prescription drug sponsors from July, 006 to July, IMPLICATIONS The analysis provides some insight into the information available to beneficiaries about plan quality based on the five-star system published by CMS. This data, used in conjunction with other plan information, provides information that may help consumers compare Medicare offered in their area. The analysis does not assess the validity of Number of contracts: 7 6 Enrollees in contracts: 7.6 million 8.7 million Percent in contracts: 8% 8% NOTE: Differences between years are statistically significant at the p<0.0 level. 009 plan ratings were weighted by county enrollment in July 008; 00 plan ratings were weighted by county enrollment in July 009. the quality ratings and the ratings may not include all of the factors that influence the quality of Medicare Advantage. The analysis only examines the summary scores for the, and the scores from the individual measures, which CMS makes available to beneficiaries, may be more relevant to some beneficiaries than the summary score. As the HEDIS reporting requirements are phased-in, consumers will have more information about PFFS currently lacking star ratings. Our analysis finds the vast majority of Medicare Advantage enrollees are in with quality ratings, even though almost half of the Medicare Advantage offered in 00 are not by CMS due to data limitations. Our analysis finds that quality ratings are closely tied to plan type, plan experience, and the governance structure of the plan (tax status). If the star ratings are used in payment policy for Medicare Advantage, non-profit and more experienced would be more likely to be rewarded. Almost one in four Medicare Advantage enrollees nationwide are in that received four or more stars, but enrollment in with four or more stars varies greatly by state. In most states, a small share of Medicare Advantage enrollees are in that received four or more stars, and in seven states and the District of Columbia, beneficiaries do not have the option to enroll in a plan with four or more stars. With one in five Medicare Advantage enrollees in with fewer than three stars, policymakers may want to focus greater oversight and attention on with relatively low quality ratings..7 This issue brief was prepared by Gretchen Jacobson, Anthony Damico, Tricia Neuman, and Jennifer Huang of the Kaiser Family Foundation. The Kaiser Family Foundation is not associated with Kaiser Permanente or Kaiser Industries.

6 See H.R. 96, Affordable Health Care for America Act of 009, as passed by the House of Representatives on November 7, 009, would provides bonus payments to high-quality qualifying in qualifying counties equal to an increase in the blended benchmark amount of.% in 0, % in 0 and % in 0 and subsequent years. A qualifying plan would be defined as a plan that had a quality ranking of stars or more in a preceding year, and a qualifying county would be defined as a county that () ranked within the lowest third of counties in the FFS costs for an area, for a year to be specified by the Secretary of HHS; and () had at least 0 percent of Medicare beneficiaries in the county enrolled in Medicare Advantage as of June of the specified year. The Secretary of HHS would be required to determine a methodology for computing quality performance scores that will be based on HEDIS, CAHPS, and other risk-adjusted quality measures, and would be required to collect outcome measures for Medicare Advantage plan enrollees. Quality bonus payments would be based on a blend of the quality performance scores and the outcomes of care. H.R. 90, Patient Protection and Affordable Care Act, as introduced on November 8, 009, would provide % bonuses to Medicare Advantage with or more stars, and % bonuses for with or more stars, beginning in 0. Plans would be eligible to receive a percent bonus if their rating is below stars and improves from the previous year, beginning in 0. New would be eligible for a % bonus, if they meet certain conditions. The Secretary of HHS would have the authority to design a different quality metric to rate the quality of the, and may make low enrollment not otherwise eligible for bonuses eligible based on the average regional or local plan quality. The CAHPS responses to all questions, except those regarding flu and pneumonia shots, are adjusted for patients age, education, mental and physical health status, eligibility for Medicaid, eligibility for Medicare Part D low-income subsidies, state of residence, and whether the survey was completed by a proxy. Data was collected for CAHPS and HOS between February and June 009, and April and August 008, respectively. See letter to all Medicare Advantage HMOs, PPOs, PFFS, 876 Cost contractors and Special Needs from the Center for Drug and Health Plan Choice, Centers for Medicare and Medicaid Services, 009 HEDIS, HOS, and CAHPS Measures for Reporting by Medicare managed care contractors, December 9, HMOs are allowed to perform medical record review for some HEDIS questions, whereas PPOs and PFFS can not, which would bias the results in favor of HMOs; however, it is not clear whether the questions subject to medical record review are included in the calculation of the Medicare Advantage quality ratings. 7 See the Centers for Medicare and Medicaid Services, Plan Directory for Medicare Advantage, Cost, PACE, and demonstration organizations, July 009, available at [ See also the Centers for Medicare and Medicaid Services, Medicare Advantage/Part D Contract and Enrollment Data, available at [ Last accessed November 7, The analysis excludes employer-direct contracts, Medical Savings Accounts (MSAs), 876 cost contracts, Health Care Prepayment Plans (HCPPs), demonstrations, Program of All Inclusive Care for the Elderly (PACE), Religious Fraternal Benefit (RFB), Special Needs Plans (SNPs), Provider Sponsored Organizations (PSOs), and Medicare Advantage only available in the U.S. territories. 9 Information about the that were terminated by CMS was not available for this analysis.

7 Table A. star quality ratings for Medicare Advantage in 009 and 00 Plans Enrollees in Mean Star rating Percent of enrollees in not Different from reference? Plans Enrollees in Mean Star rating Percent of enrollees in not Different from reference? Difference between 009 and 00? Overall 7 7,6,9.0 6% N/A 6 8,7,7.7 % N/A Yes Tax status For-profit 99,0,9.8* 7% Reference 60 6,0,778.0* % Reference Yes Non-profit 7,9,.67* % Yes 86,6,96.87* % Yes Yes Plan began before 00 No 7,67,97.8* % Reference 09,7,7.00* Reference Yes Yes 6 6,,76.6 Yes 7 6,67,.7* Yes Yes Plan type HMO/POS 07,866,08.* % Reference 6 6,66,8.* % Reference Yes Local PPO 8,66.9 6% No 68 78,7. % No Yes PFFS 8,07,90.* % Yes,06,0.* % Yes Yes Regional PPO 7,79.* % Yes 09,8.69* % Yes Yes Organizations with the largest share of Medicare Advantage enrollment in in 009 UnitedHealth Group 0,99,8.87* % N/A 6,7,06.9* % N/A No Humana,6,86.6* % N/A 9,,09.76* % N/A Yes Kaiser Permanente 6 89,9.99* N/A 6 86,96.07* N/A Yes Wellpoint 9,98.6* N/A 7,98.8* N/A Yes Highmark,9.9* 7% N/A 80,68.8* N/A Yes Health Net 9,9. 9% N/A 6,.* 7% N/A No HealthSpring,8.8* % N/A 6 79,798.0 % N/A Yes EmblemHealth 0,9.9* % N/A 66,87.09* N/A Yes Coventry Health Care 9, % N/A 0 6,6. 67% N/A No Aetna 8, % N/A,7.* 6% N/A No All other organizations 6,77,0.7* 9% N/A 96,08,88.* N/A Yes Any enforcement action July 006 July 009 No action,90,6.9* 6% Reference 7 6,,7.* % Reference Yes Suspension 8 07,68.* 6% Yes 6 9,8.7* 7% Yes Yes Fine,6,8.6* % Yes 9,78,0.78* Yes Yes Market share of largest plan in county Less than 06,,86.9* 7% Reference 9,8,0.0* 6% Reference Yes 0-9%,09,9.0 6% No 98,07,6. % No No 0-9% 9,88,6.06 % Yes 8,896,8. % Yes Yes or more,,69.8* 8% Yes 8,8,80.* 6% Yes Yes NOTE: Asterisks (*) denote values that are significantly different from the respective year s overall mean star rating, at the 9% confidence level. N/A denotes when the comparison to the reference group is not applicable; no organization was designated as the reference group and no statistical comparisons were made between organizations. 009 plan ratings were weighted by county enrollment in July 008; 00 plan ratings were weighted by county enrollment in July

8 Table A. Star ratings of the organizations with the largest share of Medicare Advantage enrollment in in 009, by plan type, 009 and 00 HMOs Local PPOs PFFS Regional PPOs Plans Enrollees in Percent of Mean Enrollees enrollees in Plans star in not rating Mean star rating Percent of enrollees in not 009 vs. 00 UnitedHealth Group,89,00.9* % 0,9,6.0* % No Humana,60.7* 7,.06* Yes Kaiser Permanente 6 89,9.99^ 6 86,96.07* Yes Wellpoint 78,0. % 6 7,8.* No Highmark 76,00.0^ 69,.00^ N/A Health Net 8,0. 9,.^ No HealthSpring,8.8* % 6 79,798.0 Yes EmblemHealth 7,76.00^ %,8.* No Coventry Health Care 7 99,8.7^ % 8 7,. % No Aetna 7,9. % 0 9,6.* % No All Other Organizations 7,9,7.* % 6,7,67.* % Yes UnitedHealth Group 76,70.6* 7% 6,7.78* % Yes Humana 0 7,0.97* % 8,087.89* % No Wellpoint,.9* 8%,98.00^ % N/A Highmark 67,90.* 9,877.0^ N/A Health Net,089.00^ %,70.0^ % N/A HealthSpring N/A 0 N/A EmblemHealth,.0^ 0,.00^ N/A Coventry Health Care,76.00^ 6,.9 Yes Aetna 0,606.0^ 6,96. % No All Other Organizations 9 9,09.6 % 6,6.7* 9% Yes UnitedHealth Group 7,7.00^ %,809.0^ % N/A Humana 68,6.0^ 7,7.* N/A Wellpoint,96.00^ %,0.0^ 6% N/A Highmark 0 0 N/A 0 8,0.00^ % N/A Health Net 0 0 N/A N/A 0 N/A EmblemHealth 0 0 N/A Coventry Health Care 0 0 N/A N/A 0 N/A Aetna 0 0 N/A N/A 0 N/A All Other Organizations 7,886.0* 8%,80.7* 86% Yes UnitedHealth Group 60,88.8* % 9,9.89* % Yes Humana 7,669.0^ 80,76.0^ N/A Wellpoint 0,.0^ 6,89.* No Health Net 0 0 N/A N/A 0 N/A Aetna 0 0 N/A N/A 0 N/A All Other Organizations,8.00^ 7 7,87.* Yes NOTE: Asterisks (*) denote values that are significantly different from the respective year s overall mean star rating, at the 9% confidence level. Carets (^) denote values that were not tested due to no variance. N/A denotes when the statistical comparison is not applicable; no statistical comparisons were made between organizations. 009 plan ratings were weighted by county enrollment in July 008; 00 plan ratings were weighted by county enrollment in July 009. HealthSpring did not offer a local PPO in 009 and EmblemHealth will not offer a PFFS plan in 00. 7

9 The Henry J. Kaiser Family Foundation Headquarters 00 Sand Hill Road Menlo Park, CA 90 (60) Fax: (60) Washington Offices and Barbara Jordan Conference Center 0 G Street, NW Washington, DC 000 (0) 7-70 Fax: (0) This publication (#80) is available on the Kaiser Family Foundation s website at The Kaiser Family Foundation is a non-profit private operating foundation, based in Menlo Park, California, dedicated to producing and communicating the best possible analysis and information on health issues.