Declining mortality and morbidity from HIV in the US and in Europe is now well established since the appearance and widespread use of HAART in the treatment of people with HIV. Those of us involved in patient care in the pre-HAART era can attest to the frequent need for hospitalization and high utilization of other costly medical interventions that defined patient care at that time. Cost of care and outcome data are becoming increasingly commonplace at scientific meetings these last two years in an attempt, it seems, to provide a context in which to evaluate the high cost of multidrug therapy. No study has been able to describe and measure all of the relevant cost and outcome variables that would be necessary for a complete understanding of the costs and benefits of HIV therapy as it continues to evolve.

Nevertheless, with each attempt to study this issue, a better economic understanding emerges which is most certain to drive public health policy and corporate purchasing of healthcare. In the study by Lapins and colleagues, mortality rates and average monthly drug and non-drug costs for managed care members with HIV was examined from 1995-1997. Protease inhibitors and the use of triple combination therapy emerged during this time.

Clinical Partners (a national HIV case management company for managed care organizations) collected data on adult non-Medicaid/non-MediCal enrollees in Texas and California from January 1995 through December 1997. The costs of medical services (drug, non-drug and total) were determined from amounts paid on claims or by imputed costs (costs not adjusted for inflation) for similar services when the services were capitated, and calculated on a member-per-month basis. Non-drug costs included hospital, professional, laboratory and homecare services. The patient population consisted of 100% insured caucasian males, with no reported intravenous drug users.

The number of HIV members increased from 474 per month in January 1995 to 722 per month in December 1997, but the authors fail to identify the source of patient growth over this time (growing patient enrollment in health plans vs. declining mortality vs. increasing number of health plans under their management). Furthermore, there is no data regarding the changes in antiretroviral therapy over this time, so we are left to surmise that increases in drug costs resulted largely from appropriate and consistent use of combination antiretroviral therapy (perhaps an unsafe assumption). Despite these shortcomings, declines in mortality, non-drug costs and total cost were demonstrated over this time period, with an increase in drug costs:

HIV MPM Costs beginning 1995		HIV MPM Costs end 1997
Drug costs	$226	Drug Costs	$1072
Non-drug costs	$1365	Non-drug costs	$166
Total Costs	$1472	Total Costs	$1316
Mortality rate	4.8% (1 Q 95)	Mortality rate	0.25% (4 Q 97)
(Upon statistical analysis, the rising trend in drug costs and the declining trends in mortality and non-drug and total costs and significant r values)

No causal relationship can be made in this study between the use of combination antiretroviral therapy and declines in costs and mortality. However it is quite plausible, in this demographically homogeneous employed patient population, that improvement in patient outcomes is derived from the availability of new and potent drug combinations. In agreement with the authors, future examinations of the economics of HIV care should consider all components of care as well as indirect costs such as costs to community services, employers, friends, families, and economic benefits in terms of productive years gained.