Home Search Abstracts Browse Sessions Program Committee View Session E-mail Abstract Author

 

 




Session 172 Poster Abstracts
Health Services and Cost Effectiveness
Wednesday, 1:30 - 3:30 pm
Hall B


987    
Medicare’s HIV Patient Population: HAART Alters Historic Patterns
D Gilden, J Kubisiak, and Daniel Gilden*
JEN Associates, Cambridge, MA, USA

Background:  The U.S. government’s Medicare program covers physician and hospital costs for retirees and the long-term disabled. Advanced HIV has been a qualifying condition for Medicare, but survival was short before the advent of HAART, and Medicare’s HIV+ population had a limited, transient composition. The greater survival in the HAART era could change the composition and costs of this caseload.

Methods:  We examined the medical billing records of a 5% national Medicare sample spanning the years 1997 to 2001. We confirmed patients HIV status by flagging those with at least 3 HIV-related treatment procedures (excluding diagnostic tests) in a single year. The cohort was stratified by year and categorized by age, race, gender, and Medicare status. The population summaries were further stratified by the presence of major chronic diseases and HIV-related conditions. We did not have records for most medication costs since Medicare does not cover self-administered drugs.

Results:  Medicare’s total HIV-related costs exceeded $1.2 billion dollars over > 3.7 million months of Medicare eligibility. The confirmed HIV+ population grew from 42,540 in 1997 to 64,340 in 2001, while mortality fell 27%. Their HIV-related costs per month declined 22% and other Medicare costs remained constant. Prevalent HIV cases—those enrolled in previous years—grew from 68% of the 1998 caseload to 88% in 2001. The elderly represented 5.8% of the population over this period. In 2001, the prevalence of treated HIV in the U.S. elderly was 0.013%, compared with 0.008% in 1997. Hospitalization accounted for just over half of all Medicare payments. Renal disease proved to be the highest disease cost after HIV itself though it represented only 6.8% patient months. Renal expenses were concentrated in women and blacks. Blacks incurred 14-fold higher renal expenses than whites. Women had one-third more months with renal disease than men, but their treatment costs were 71% higher. In addition, diabetes was 2.4 times more frequent in those with renal disease and cardiovascular disease was 3.1 times more frequent.

Conclusions:  As HIV has become better managed, the Medicare’s HIV population has become larger and more permanent. Total costs have risen even as HIV payments declined. This population has taken on some of the characteristics of other aging populations:  end-organ and metabolic dysfunctions make up a rising proportion of payments.

Keywords: Treatment Costs; Population Epidemiology; Medicare