Expenditures for the Care of HIV-Infected Patients in the Era of Highly Active Antiretroviral Therapy [Bozzette SA, et al. NEJM 2001;344:817]: This is another report from the HIV Cost and Services Utilization Study Consortium (HCSUSC), which represents a random sampling of 2,864 patients representing the U.S. population with HIV infection. The present report concerns estimated costs based on patient interviews beginning in early 1996, with a three year follow-up. The results showed an average expenditure of $1,792/patient/month in 1996, which decreased to $1,359/patient/month in 1997. Estimated total annual expenditures were $20,300/patient in 1996 and $18,300/patient in 1998. The data for 1998 shows that about 55% of the total expenditures was for medications, 15% for outpatient services, and 30% for hospitalization. Trend data for this distribution changed substantially: overall expenses decreased by 16%, hospitalization expenditures decreased 43%, and pharmaceutical expenditures increased by 33%. It was noted that hospital expenditures reached a nadir in 1997, but then started to increase. Analysis by various categories showed substantial differences, with higher expenditures in patients with low CD4 counts, injection drug users, Medicaid recipients, patients in the northeast, and patients in a panel of <10 with HIV infection. Details are shown below:

Comment: It should be noted that the cost analysis here is based on patient interviews with arbitrary assignment of cost according to the services claimed. Thus, these are estimated costs based on average expenditures for services that patients remember. The authors point out that the decline they note is for average annual costs and should not be interpreted to indicate reduced lifetime costs. The estimated cost/year of life saved is estimated at $11,000 - $18,000/year, which is consistent with prior reports [Pharmacoeconomics 2000;17:325]. The authors used fixed prices so that the impact of escalating drug costs was not shown. The difference in cost by categories is not surprising, but certainly worth emphasizing. CD4 cell count has a profound impact; other significant factors were injection drug use, Medicaid as the source of insurance, and location in the northeast. Particularly prominent and worth emphasizing was the impact of panel size with a $5,000/year cost differential for a panel size <10 patients compared to a panel size >10 patients. The authors noted an increase in the more recent data for hospitalization, which they warn is a disturbing trend that possibly reflects drug failures due to resistance or intolerance.
posted 3/23/2001

back to news table of contents