Search

Objective: To estimate and describe factors driving variation in spending for breast cancer patients within geographic region.Data Source: Surveillance, Epidemiology, and End Results (SEER)-Medicare database from 2009-2013.Study Design: The proportion of variation in monthly medical spending within geographic region attributed to patient and physician factors was estimated using multilevel regression models with individual patient and physician random effects. Using sequential models, we estimated the contribution of differences in patient and disease characteristics or use of cancer treatment modalities to patient-level and physician-level variance in spending. Services associated with high spending physicians were estimated using linear regression.Data Extraction Method: A total of 20 818 women with a breast cancer diagnosis in 2010-2011.Principal Findings: We observed substantial between-patient and between-provider variation in spending following diagnosis and at the end-of-life. Immediately following diagnosis, 48% of between-patient and 31% of between-physician variation were driven by differences in delivery of cancer treatment modalities to similar patients. At the end-of-life, patients of high spending physicians had twice as many inpatient days, double the chemotherapy spending, and slightly more hospice days.Conclusions: Similar patients receive very different treatments, which yield significant differences in spending. Efforts to reduce unwanted variation may need to target treatment choices within patient-doctor discussions. [ABSTRACT FROM AUTHOR]

Objective: To compare medication adherence, cost, and utilization in Medicare beneficiaries attributed to nurse practitioners (NP) and primary care physicians (PCP).Data: Medicare Part A, B, and D claims and beneficiary summary file data, years 2009-2013.Study Design: We used propensity score-weighted analyses combined with logistic regression and generalized estimating equations to test differences in good medication adherence (proportion of days covered (PDC >0.8); office-based and specialty care costs; and ER visits.Data Extraction: Beneficiaries with prescription claims for anti-diabetics, renin-angiotensin system antagonists (RASA), or statins.Principal Findings: There were no differences in good medication adherence (PDC >0.8) between NP and PCP attributed beneficiaries taking anti-diabetics or RASA. Beneficiaries taking statins had a slightly higher probability of good adherence when attributed to PCPs (74.6% vs 75.5%; P < 0.05). NP attributed beneficiaries had lower office-based and specialty care costs and were less likely to experience an ER visit across all three medication cohorts (P < 0.01).Conclusions: Examining the impact of NP and PCP provided care on outcomes beyond the primary care setting is important to the Medicare program in general but will also help practices seeking to meet benchmarks under alternative payment models that incentivize higher quality and lower costs. [ABSTRACT FROM AUTHOR]

Objective: To estimate health care utilization and costs incurred by homeless Veterans relative to nonhomeless Veterans and to examine the impact of a VA homelessness program on these outcomes.Data Sources/study Setting: Combined Department of Veterans Affairs (VA) administrative and Medicare claims data.Study Design: Observational study using longitudinal data from Veterans engaged with the VA system and enrolled in Medicare. Veterans with administrative evidence of homelessness at any point during 2006-2010 were matched on period of military service to Veterans with no evidence of homelessness.Principal Findings: Experience of homelessness was associated with 1.37 (95 percent CI = 1.34-1.40) and 0.16 (95 percent CI = 0.14-0.17) more outpatient encounters per quarter in VA and non-VA settings, respectively, and 1.31 (95 percent CI = 1.30-1.32) and 0.49 (95 percent CI = 0.48-0.49) more inpatient days per quarter in VA and non-VA hospitals, respectively. These were associated with higher costs. Relative to stably housed Veterans less than 65 years of age, those enrolled in a VA homelessness program had 94.4 percent (95 percent CI = 90.7 percent-98.1 percent) more VA outpatient visits but 5.5 percent (95 percent CI = 3.0 percent-7.9 percent) fewer Medicare outpatient visits.Conclusions: Homelessness was associated with an increase in VA and Medicare utilization and cost. A VA homelessness program decreased use of Medicare outpatient services. [ABSTRACT FROM AUTHOR]

Objective: To assess the issue of nonrepresentative sampling in Medicare Advantage (MA) risk adjustment.Data Sources: Medicare enrollment and claims data from 2008 to 2011.Data Extraction: Risk adjustment predictor variables were created from 2008 to 2010 Part A and B claims and the Medicare Beneficiary Summary File. Spending is based on 2009-2011 Part A and B, Durable Medical Equipment, and Home Health Agency claims files.Study Design: A propensity-score matched sample of Traditional Medicare (TM) beneficiaries who resembled MA enrollees was created. Risk adjustment formulas were estimated using multiple techniques, and performance was evaluated based on R2 , predictive ratios, and formula coefficients in the matched sample and a random sample of TM beneficiaries.Principal Findings: Matching improved balance on observables, but performance metrics were similar when comparing risk adjustment formula results fit on and evaluated in the matched sample versus fit on the random sample and evaluated in the matched sample.Conclusions: Fitting MA risk adjustment formulas on a random sample versus a matched sample yields little difference in MA plan payments. This does not rule out potential improvements via the matching method should reliable MA encounter data and additional variables become available for risk adjustment. [ABSTRACT FROM AUTHOR]

Objective: To identify hospital/county characteristics and sources of regional heterogeneity associated with readmission penalties.Data Sources/study Setting: Acute care hospitals under the Hospital Readmissions Reduction Program from fiscal years 2013 to 2018 were linked to data from the Annual Hospital Association, Centers for Medicare and Medicaid Services, Medicare claims, Hospital Compare, Nursing Home Compare, Area Resource File, Health Inequity Project, and Long-term Care Focus. The final sample contained 3,156 hospitals in 1,504 counties.Data Collection/extraction Methods: Data sources were combined using Medicare hospital identifiers or Federal Information Processing Standard codes.Study Design: A two-level hierarchical model with correlated random effects, also known as the Mundlak correction, was employed with hospitals nested within counties.Principal Findings: Over a third of the variation in readmission penalties was attributed to the county level. Patient sociodemographics and the surrounding access to and quality of care were significantly associated with penalties. Hospital measures of Medicare volume, percentage dual-eligible and Black patients, and patient experience were correlated with unobserved area-level factors that also impact penalties.Conclusions: As the readmission risk adjustment does not include any community-level characteristics or geographic controls, the resulting endogeneity bias has the potential to disparately penalize certain hospitals. [ABSTRACT FROM AUTHOR]

Objective: To assess frequency, type, and extent of behavioral health (BH) nonquantitative treatment limits (NQTLs) before and after implementation of the Mental Health Parity and Addiction Equity Act of 2008 (MHPAEA).Data Sources: Secondary administrative data for Optum carve-out and carve-in plans.Study Design: Cross-tabulations and "two-part" regression models were estimated to assess associations of parity period with NQTLs.Data Collection/extraction Methods: Optum provided four proprietary BH databases, including 2008-2013 data for 40 carve-out and 385 carve-in employers from Optum's claims processing databases and 2010 data from interviews conducted by Optum's parity compliance team with 49 carve-out employers.Principal Findings: Preparity, carve-out plans required preauthorization for in-network inpatient/intermediate care; otherwise coverage was denied. Postparity, 73 percent would review later by request and half charged no penalty for late authorization. Outpatient visit authorization requirements virtually disappeared. For carve-out out-of-network inpatient/intermediate care, and for carve-ins, plans changed penalties to match medical service policies, but this did not necessarily lead to fewer requirements or lower penalties.Conclusion: After 2011, MHPAEA was associated with the transformation of BH care management, including much less restrictive preauthorization requirements, especially for in-network care provided by carve-out plans. [ABSTRACT FROM AUTHOR]

Objective: To examine the relationship between cost and quality in Veterans Health Administration (VA) nursing homes (called Community Living Centers, CLCs) using longitudinal data.Data Sources/study Setting: One hundred and thirty CLCs over 13 quarters (from FY2009 to FY2012) were studied. Costs, resident days, and resident severity (RUGs score) were obtained from the VA Managerial Cost Accounting System. Clinical quality measures were obtained from the Minimum Data Set, and resident-centered care (RCC) was measured using the Artifacts of Culture Change Tool.Study Design: We used a generalized estimating equation model with facilities included as fixed effects to examine the relationship between total cost and quality after controlling for resident days and severity. The model included linear and squared terms for all independent variables and interactions with resident days.Principal Findings: With the exception of RCC, all other variables had a statistically significant relationship with total costs. For most poorer performing smaller facilities (lower size quartile), improvements in quality were associated with higher costs. For most larger facilities, improvements in quality were associated with lower costs.Conclusions: The relationship between cost and quality depends on facility size and current level of performance. [ABSTRACT FROM AUTHOR]

Objective: To investigate whether enrollment in high-deductible health plans (HDHPs) led enrollees to choose lower-priced providers for office visits and laboratory tests.Study Setting: Claims data from more than 40 large employers.Study Design: We compared the change in price for office visits and laboratory tests for enrollees who switched to HDHPs versus enrollees who remained in traditional plans. We estimated separate models for enrollees who changed providers versus those who remained with the same provider to disentangle the effects of HDHPs on provider choice and negotiated prices.Data Collection: Claims data from 2004 to 2010 on 1.8 million enrollees.Principal Findings: After enrollment in HDHPs, 28 percent of enrollees changed physicians for office visits (compared to 19 percent in the Traditional Plan group, p < .01); however, this did not result in a statistically significant reduction in price for office visits. About 25 percent of enrollees changed providers for laboratory tests (compared to 23 percent in the Traditional Plan group, p < .01), resulting in savings of about $2.09 or a 12.8 percent reduction in price per laboratory test. We found that HDHPs had lower negotiated prices for office visits but not for laboratory tests.Conclusions: High-deductible health plan enrollment may shift enrollees to lower cost providers, resulting in modest savings. [ABSTRACT FROM AUTHOR]

Objective: To evaluate the impact of the 2006 Massachusetts health reform, the model for the Affordable Care Act, on short-term enrollment and utilization in the unsubsidized individual health insurance market.Data Source: Seven years of administrative and claims data from Harvard Pilgrim Health Care.Research Design: We employed pre-post survival analysis and an interrupted time series design to examine changes in enrollment length, utilization patterns, and use of elective procedures (discretionary inpatient surgeries and infertility treatment) among nonelderly adult enrollees before (n = 6,912) and after (n = 29,207) the MA reform.Principal Findings: The probability of short-term enrollment dropped immediately after the reform. Rates of inpatient encounters (HR = 0.83, 95 percent CI: 0.74, 0.93), emergency department encounters (HR = 0.85, 95 percent CI: 0.80, 0.91), and discretionary inpatient surgeries (HR = 0.66 95 percent CI: 0.45, 0.97) were lower in the postreform period, whereas the rate of ambulatory visits was somewhat higher (HR = 1.04, 95 percent CI: 1.00, 1.07). The rate of infertility treatment was higher after the reform (HR = 1.61, 95 percent CI: 1.33, 1.97), driven by women in individual (vs. family) plans. The reform was not associated with increased utilization among short-term enrollees.Conclusions: MA health reform was associated with a decrease in short-term enrollment and changes in utilization patterns indicative of reduced adverse selection in the unsubsidized individual market. Adverse selection may be a problem for specific, high-cost treatments. [ABSTRACT FROM AUTHOR]

Objectives: To conduct a systematic literature review of selected major provisions of the Affordable Care Act (ACA) pertaining to expanded health insurance coverage. We present and synthesize research findings from the last 5 years regarding both the immediate and long-term effects of the ACA. We conclude with a summary and offer a research agenda for future studies.Study Design: We identified relevant articles from peer-reviewed scholarly journals by performing a comprehensive search of major electronic databases. We also identified reports in the "gray literature" disseminated by government agencies and other organizations.Principal Findings: Overall, research shows that the ACA has substantially decreased the number of uninsured individuals through the dependent coverage provision, Medicaid expansion, health insurance exchanges, availability of subsidies, and other policy changes. Affordability of health insurance continues to be a concern for many people and disparities persist by geography, race/ethnicity, and income. Early evidence also indicates improvements in access to and affordability of health care. All of these changes are certain to ultimately impact state and federal budgets.Conclusions: The ACA will either directly or indirectly affect almost all Americans. As new and comprehensive data become available, more rigorous evaluations will provide further insights as to whether the ACA has been successful in achieving its goals. [ABSTRACT FROM AUTHOR]

Objective: To evaluate the effect of the Oregon and New Mexico Health Insurance Flexibility and Accountability (HIFA) demonstrations. HIFA is an optional state Medicaid expansion targeted at adults and children with incomes below 200 percent of the federal poverty level (FPL). The study has five research questions: What type of health insurance do HIFA enrollees self report in surveys? What are the demographic characteristics of these enrollees? What type of health insurance coverage, if any, did HIFA enrollees have just prior to enrollment in the HIFA program? Among those with prior coverage, what prompted participation in the HIFA program? What type of health insurance, if any, would HIFA enrollees have in the absence of HIFA?Methods: Data were collected via telephone interviews with a total of 406 enrollees from Oregon and 409 enrollees from New Mexico. The survey was conducted between July 7 and September 20, 2009, for both states. The sample frame for the survey was based on administrative records of adults enrolled in June 2009. After completion of the survey, active enrollment status as of the date the telephone interview was confirmed. Respondents no longer enrolled at the time of the survey (7 cases in NM and 14 in OR) were excluded from the analysis. The final sample size was 794 verified HIFA enrollees.Results: HIFA enrollees tended to be middle-aged, male, and relatively unhealthy. Employment status varied tremendously from the self-employed to retired to unable to work. HIFA enrollees were reasonably well educated with 80 percent having at least a high school education. Most HIFA enrollees (90 percent) reported being uninsured just prior to participation in HIFA. Of those who were uninsured, most reported having been uninsured for an extended time-a year or more. Most enrollees joined HIFA because they lacked access to health insurance or could not afford insurance on the private market. The overwhelming majority (76 percent) of respondents believed that they would be uninsured in the absence of HIFA, with few considering either an employer plan or private purchase to be a viable option. Over 90 percent of enrollees correctly indicated they had insurance coverage. However, characterization of the type of coverage was problematic, particularly in the absence of the program-specific name.Conclusions: HIFA enrolled a relatively sick, male, middle-aged population that tended to have been long-term uninsured--the kind of enrollees for which the programs were designed--with little apparent crowd-out of private insurance. The reported health status coupled with low incomes suggests that individual purchase is unlikely, a sentiment echoed by the respondents. In the absence of HIFA, most enrollees believed they would rejoin the ranks of the uninsured from where they came. [ABSTRACT FROM AUTHOR]

Objective To determine if providing dental insurance to older Americans would close the current gaps in dental use and expenditure between insured and uninsured older Americans. Data Sources/Study Setting We used data from the 2008 Health and Retirement Survey ( HRS) supplemented by data from the 2006 Medical Expenditure Panel Survey ( MEPS). Study Design We compared the simulated dental use and expenditures rates of newly insured persons against the corresponding rates for those previously insured. Data Collection/Extraction Methods The HRS is a nationally representative survey administered by the Institute for Social Research ( ISR). The MEPS is a nationally representative household survey sponsored by the Agency for Healthcare Research and Quality ( AHRQ). Principal Findings We found that expanding dental coverage to older uninsured Americans would close previous gaps in dental use and expense between uninsured and insured noninstitutionalized Americans 55 years and older. Conclusions Providing dental coverage to previously uninsured older adults would produce estimated monthly costs net of markups for administrative costs that comport closely to current market rates. Estimates also suggest that the total cost of providing dental coverage targeted specifically to nonusers of dental care may be less than similar costs for prior users. [ABSTRACT FROM AUTHOR]

Objective. The impact of consumer-driven health plans (CDHPs) has primarily been studied in a small number of large, self-insured employers, but this work may not generalize to the wide array of firms that make up the overall economy. The goal of our research is to examine effects of health savings accounts (HSAs) on total, medical, and pharmacy spending for a large number of small and midsized firms. Data Sources. Health plan administrative data from a national insurer were used to measure spending for 76,310 enrollees over 3 years in 709 employers. All employers began offering a HSA-eligible plan either on a full-replacement basis or alongside traditional plans in 2006 and 2007 after previously offering only traditional plans in 2005. Study Design. We employ difference-in-differences generalized linear regression models to examine the impact of switching to HSAs. Data Extraction Methods. Claims data were aggregated to enrollee-years. Principal Findings. For total spending, HSA enrollees spent roughly 5–7 percent less than non-HSA enrollees. For pharmacy spending, HSA enrollees spent 6–9 percent less than traditional plan enrollees. More of the spending decrease was observed in the first year of enrollment. Conclusions. Our findings are consistent with the notion that CDHP benefit designs affect decisions that are at the discretion of the consumer, such as whether to fill or refill a prescription, but have less effect on care decisions that are more at the discretion of the provider. [ABSTRACT FROM AUTHOR]

Objective. Recent public concern in response to states' intended repeal of Medicaid bed-hold policies and report of their association with higher hospitalization rates prompts examination of these policies in ensuring continuity of care within the broader context of Medicaid policies. Data Sources/Study Design. Minimum Data Set assessments of long-stay nursing home residents in April–June 2000 linked to Medicare claims enabled tracking residents' hospitalizations during the ensuing 5 months and determining hospital discharge destination. Multinomial multilevel models estimated the effect of state policies on discharge destination controlling for resident, hospitalization, nursing home, and market characteristics. Results. Among 77,955 hospitalizations, 5,797 (7.4 percent) were not discharged back to the baseline nursing home. Bed-hold policies were associated with lower odds of transfer to another nursing home (AOR=0.55, 95 percent CI 0.52–0.58) and higher odds of hospitalization (AOR=1.36), translating to 9.5 fewer nursing home transfers and 77.9 more hospitalizations per 1,000 residents annually, and costing Medicaid programs about $201,311. Higher Medicaid reimbursement rates were associated with lower odds of transfer. Conclusions. Bed-hold policies were associated with greater continuity of NH care; however, their high cost compared with their small impact on transfer but large impact on increased hospitalizations suggests that they may not be effective. [ABSTRACT FROM AUTHOR]

Objective. To evaluate the cost-effectiveness of Project Dulce, a culturally specific diabetes case management and self-management training program, in four cohorts defined by insurance status. Data Sources/Study Setting. Clinical and cost data on 3,893 persons with diabetes participating in Project Dulce were used as inputs into a diabetes simulation model. Study Design. The Center for Outcomes Research Diabetes Model, a published, peer-reviewed and validated simulation model of diabetes, was used to evaluate life expectancy, quality-adjusted life expectancy (QALY), cumulative incidence of complications and direct medical costs over patient lifetimes (40-year time horizon) from a third-party payer perspective. Cohort characteristics, treatment effects, and case management costs were derived using a difference in difference design comparing data from the Project Dulce program to a cohort of historical controls. Long-term costs were derived from published U.S. sources. Costs and clinical benefits were discounted at 3.0 percent per annum. Sensitivity analyses were performed. Principal Findings. Incremental cost-effectiveness ratios of $10,141, $24,584, $44,941, and $69,587 per QALY gained were estimated for Project Dulce participants versus control in the uninsured, County Medical Services, Medi-Cal, and commercial insurance cohorts, respectively. Conclusions. The Project Dulce diabetes case management program was associated with cost-effective improvements in quality-adjusted life expectancy and decreased incidence of diabetes-related complications over patient lifetimes. Diabetes case management may be particularly cost effective for low-income populations. [ABSTRACT FROM AUTHOR]

Objective. To test the effect of a consumer-directed model (Cash and Counseling) of Medicaid personal care services (PCS) or home- and community-based waiver services (HCBS) on the cost of Medicaid services. Data Sources/Study Setting. Medicaid claims data were collected for all enrollees in the Cash and Counseling demonstration. Demonstration enrollees included those eligible for PCS (in Arkansas), those assessed to receive such services (in New Jersey), and recipients of Medicaid HCBS (in Florida). Enrollment occurred from December 1998 through April 2001. The follow-up period covered up to 24 months after enrollment. Study Design. Demonstration volunteers were randomly assigned to have the option to participate in Cash and Counseling (the treatment group), or to receive Medicaid services as usual from an agency (the control group). Ordinary least squares regressions were used to estimate the effect of the program on costs for Medicaid PCS/waiver services and other Medicaid services, while controlling for consumers' preenrollment characteristics and preenrollment Medicaid spending. Models were estimated separately for nonelderly and elderly adults in each state and for children in Florida. Data Extraction Methods. Each state supplied claims data for demonstration enrollees. Principal Findings. Largely because the program increased consumers' ability to get the authorized amount of paid care, expenditures for personal care/waiver services were higher for the treatment group than for the control group in each state and age group, except among the elderly in Florida. Higher costs for personal care/waiver services were partially offset by savings in other Medicaid services, particularly those related to long-term care. During year 1, total Medicaid costs were generally higher for the treatment group than for the control group, with treatment–control cost differences ranging from 1 percent (and statistically insignificant) for the elderly in Florida to 17 percent for the elderly in Arkansas. In year 2, these cost differences were generally greater than in year 1. Only in Arkansas did the treatment–control difference in total cost shrink over time—to less than 5 percent (and statistically insignificant) in year 2. Conclusions. Medicaid costs were generally higher under Cash and Counseling because those in the traditional system did not get the services they were entitled to. Compared with the treatment group, (1) control group members were less likely to receive any services at all (despite being authorized for them), and (2) service recipients received a lower proportion of the amount of care that was authorized. In addition, a flaw in Florida's reassessment procedures led to treatment group members receiving more generous benefit amounts than control group members. To keep total Medicaid costs per recipient at the level incurred under the traditional system, consumer-directed programs need to be carefully designed and closely monitored. [ABSTRACT FROM AUTHOR]

Objective. To examine the effects of policy, health system, and sociodemographic characteristics on the likelihood that uninsured persons pay a lower price at their regular source of care, or that they are aware of lower priced providers in their community. Data Sources. The 2003 Community Tracking Study household survey, a nationally representative sample of the U.S. population and 60 randomly selected communities. Study Design. The survey asked uninsured persons if they paid full or reduced cost at their usual source of medical care, or if they were aware of providers in their community that charge less for uninsured people. We use binomial and multinomial logistic regression analysis to examine the effects of various policy, health system, and sociodemographic characteristics on use and awareness of lower priced providers. We focus especially on the effects of safety-net capacity, measured by safety-net hospitals, community health centers, physicians' charity care, and Community Access Program (CAP) grants. Principal findings. Less than half of the uninsured (47.5 percent) reported that they used or were aware of a lower priced provider in their community. Multivariate regression analysis shows that greater safety-net capacity is associated with a higher likelihood of having a lower priced provider as the regular source of care and greater awareness of lower priced providers. Lower incomes and racial/ethnic minorities also had a higher likelihood of having a lower priced provider, although health status did not have statistically significant effects. Conclusion. Although increased safety-net capacity may lead to more uninsured having a lower priced provider, many uninsured who live near safety-net providers are not aware of their presence. Greater outreach designed to increase awareness may be needed in order to increase the effectiveness of safety-net providers in improving access to care for the uninsured. [ABSTRACT FROM AUTHOR]

To evaluate the performance of different prospective risk adjustment models of outpatient, inpatient, and total expenditures of veterans who regularly use Veterans Affairs (VA) primary care.We utilized administrative, survey and expenditure data on 14,449 VA patients enrolled in a randomized trial that gave providers regular patient health assessments.This cohort study compared five administrative data-based, two self-report risk adjusters, and base year expenditures in prospective models.VA outpatient care and nonacute inpatient care expenditures were based on unit expenditures and utilization, while VA expenditures for acute inpatient care were calculated from a Medicare-based inpatient cost function. Risk adjusters for this sample were constructed from diagnosis, medication and self-report data collected during a clinical trial. Model performance was compared using adjustedR2 and predictive ratios.In all expenditure models, administrative-based measures performed better than self-reported measures, which performed better than age and gender. The Diagnosis Cost Groups (DCG) model explained total expenditure variation (R2=7.2 percent) better than other models. Prior outpatient expenditures predicted outpatient expenditures best by far (R2=42 percent). Models with multiple measures improved overall prediction, reduced over-prediction of low expenditure quintiles, and reduced under-prediction in the highest quintile of expenditures.Prediction of VA total expenditures was poor because expenditure variation reflected utilization variation, but not patient severity. Base year expenditures were the best predictor of outpatient expenditures and nearly the best for total expenditures. Models that combined two or more risk adjusters predicted expenditures better than single-measure models, but are more difficult and expensive to apply. [ABSTRACT FROM AUTHOR]

To discuss and quantify the incentives that Medicare managed care plans have to avoid (through selective enrollment or disenrollment) people who are at risk for very high costs, focusing on Medicare beneficiaries in the last year of life—a group that accounts for more than one-quarter of Medicare's annual expenditures. Medicare administrative claims for 1994 and 1995. We calculated the payment a plan would have received under three risk-adjustment systems for each beneficiary in our 1995 sample based on his or her age, gender, county of residence, original reason for Medicare entitlement, and principal inpatient diagnoses received during any hospital stays in 1994. We compared these amounts to the actual costs incurred by those beneficiaries. We then looked for clinical categories that were predictive of costs, including costs in a beneficiary's last year of life, not accounted for by the risk adjusters. The analyses were conducted using claims for a 5 percent random sample of Medicare beneficiaries who died in 1995 and a matched group of survivors. Medicare is currently implementing the Principal Inpatient Diagnostic Cost Groups (PIP-DCG) risk adjustment payment system to address the problem of risk selection in the Medicare+Choice program. We quantify the strong financial disincentives to enroll terminally ill beneficiaries that plans still have under this risk adjustment system. We also show that up to one-third of the selection observed between Medicare HMOs and the traditional fee-for-service system could be due to differential enrollment of decedents. A risk adjustment system that incorporated more of the available diagnostic information would attenuate this disincentive; however, plans could still use clinical information (not included in the risk adjustment scheme) to identify beneficiaries whose expected costs exceed expected payments. More disaggregated prospective risk adjustment methods and alternative payment systems that compensate plans for delivering care to certain classes of patients should be considered to ensure access to high-quality managed care for all beneficiaries. [ABSTRACT FROM AUTHOR]