Effect of highly active antiretroviral therapy on outcomes in Veterans Affairs Medical Centers

Although the use of highly active antiretroviral therapy (HAART) has dramatically increased the pharmaceutical cost of treating HIV-infected individuals, the improved clinical outcomes of treatment, especially in patients with under 200 CD4 cells/μl have resulted in the decreased utilization of other healthcare resources [1–4]. However, no published studies have reported the economic impact of HAART using data subsequent to 1997. To provide insight into the longer term impact of HAART, we have evaluated inpatient mortality, the rates of acute inpatient hospitalization, and the costs of outpatient pharmaceutical agents for patients receiving treatment in Veterans Health Administration (VHA) Medical Centers in southern California from 1 October 1995 to 30 September 1999. We retrospectively obtained data from all four southern California VHA Medical Centers (located in Loma Linda, Long Beach, Los Angeles, and San Diego, California, USA) starting with the first quarter of the 1996 federal fiscal year (FY1996, beginning 1 October 1995) to the end of the fourth quarter of the 1999 federal fiscal year (FY1999, 30 September 1999). All data, including the costs of outpatient prescriptions (provided by the VHA for all patients under care), were obtained from each institution's standardized computerized database (the Immunology Case Registry). We employed the year-specific VHA Tortuously Liable billing rates for general medicine ward services in 1996, 1997, 1998, and 2000 to calculate the costs of acute hospitalization on the acute inpatient services (defined as the medical, surgical and neurological wards, and intensive care units) [5]. These daily costs were US$966–1409. As no specific directive was issued for 1999, we used the mean of the cost for 1998 and 2000 for that year. The number of HIV-infected patients receiving care increased from 865 in the first quarter of FY1996 to 1128 in the fourth quarter of FY1999. The proportion of patients who had developed advanced disease (fewer than 200 CD4 lymphocytes/μl or a clinical AIDS diagnosis) was 59 and 63% in the first and last study periods, respectively. Over the study period, total outpatient pharmaceutical expenditures for HIV-infected patients increased from US$434 888 to 1 572 527. Per-patient quarterly pharmaceutical costs increased by US$740, 1002, and 1117, respectively, for persons with more than 500, 200–499, and less than 200 CD4 lymphocytes/μl but no other AIDS defining diagnosis, and by US$299 for individuals with a clinical AIDS-defining diagnosis. On average, per-patient quarterly expenditures for antiretroviral agents increased from US$139 to 1148, whereas expenditures for other medications decreased from US$370 to 246. The use of protease inhibitors and non-nucleoside reverse transcriptase inhibitors constituted 47 and 11% of the total antiretroviral costs in FY1999. Acute inpatient hospital days totalled 1705 in the first quarter of FY1996 versus 930 days in the fourth quarter of FY1999. Correspondingly, the estimated quarterly costs of acute hospitalization decreased from US$1 647 030 to 1 310 370. The overall per-patient costs of outpatient pharmaceutical usage and acute hospitalization were US$506 and 1904 for the first quarter of 1996, versus US$1394 and 1162 for the last quarter of 1999 (Fig. 1).Fig. 1.: Each bar represents the per-quarter sum of the per-patient cost of outpatient pharmaceuticals (≤) and acute hospitalizations (′) at the participating medical centers.Finally, the combined number of inpatient deaths per quarter for the four sites decreased from 40 to 12 deaths between the first quarter of FY1996 and the first quarter of FY1997. Subsequently, no changes in the number of deaths were noted. Over the 4 year period beginning in October 1995, we found that a marked increase in per-patient antiretroviral pharmaceutical costs was accompanied by a sustained decrease in the days of acute inpatient hospitalization per 1000 patients (5159 in FY1996 versus 2354 in FY1999). Therefore, over a 4 year period, the use of HAART was associated with a relatively stable yearly per-patient outlay for the combined costs of acute hospitalization plus outpatient pharmaceutical use (Fig. 1). Of some concern, however, is the observation that despite increased pharmaceutical usage, the total mortality has not decreased since the first quarter of FY1997. Furthermore, we found an upward tendency of per-patient costs over the last 12 months of this study. Previous studies have shown that the use of HAART from 1995 to 1997 was accompanied by a decrease in overall medical expenditures that was largely attributable to decreased rates of hospitalization [2–4]. Projections based on such short-term data suggest that HAART is cost effective when compared with other medical interventions [6], as estimates of the cost per year of life saved range from US$10 000 to 91 000 [7–10]. The medical costs of HAART may be further offset by the societal benefit derived from the increased economic productivity of HIV-infected individuals [7]. The weaknesses of our study include its retrospective design, and our inability to link outcomes in individual patients to the use of specific treatment regimens, or to assess the benefits of more aggressive antiretroviral therapy by disease status. Also, these results pertain to a cross-sectional analysis rather than to a longitudinal study of a coherent population. Furthermore, we did not adjust our costs for the effect of inflation or capture all medical costs, such as those resulting from inpatient psychiatric or nursing home care, laboratory or radiological tests, other aspects of outpatient care, or medical care occurring outside the VHA. With regard to the latter concern, the use of non-VHA healthcare facilities accounted for only 2.6% of the costs of HIV care incurred by veterans in 1994 [11]. In addition, the precision of the VHA Tortuously Liable Billing Rate and its application to the costs of care for HIV-infected inpatients may be questioned. Finally, although the improved clinical outcomes that accompanied HAART in our facilities are consistent with the observations of other investigators, some of this improvement may be attributable to changes in the use of prophylaxis for opportunistic infection, changes in the proportion of patients who died while in the hospital, greater physician experience, and the increased use of dual nucleoside therapy, especially at the beginning of this observational study [1,12,13]. In summary, over a 4 year period, we demonstrated a sustained decrease in acute inpatient hospitalizations for HIV-infected patients that substantially counterbalanced the increased costs of pharmaceutical care resulting from the use of HAART. Nevertheless, further cost-effectiveness analyses will be important to assess the durability of the clinical and economic impact of HAART. The virological failure of up to 60% of treatment-experienced patients [14] and the increased recognition of the toxicities of antiretroviral therapy suggests that substantial additional medical costs may eventually accrue in the care of these patients. Matthew Bidwell Goetzae Anthony P. Morrealebf David C. Rhewae Stephen Bermancg Michael Ingdh Dian Eldridged Janice C. Justisc Ed Lotta

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.