As discussed previously, direct COI are the economic representation of resource consumption and have a natural monetary value. They can occur on the provider and the household side. This chapter analyses the direct COI in four steps: Provider cost before HAART (Sect. 2.1.1), provider cost in the HAART era (Sect. 2.1.2), lifetime provider costs (Sect. 2.1.3), and direct household cost (Sect. 2.1.4).
From the early beginning of treating HIV/AIDS, most health economic studies fo-cussed on the calculation of provider costs. During the first years there had been a clear dominance of research on hospital costs for patients with AIDS, in particular, in the United States of America.
Scitovsky et al. (1986) calculated the average cost per AIDS-related hospital admission as US$ 9,024 ranging from US$ 7,026 to US$ 23,425. A more comprehensive picture is presented by Scitovsky and Rice (1987), who estimated provider cost of the AIDS epidemic in the United States in 1985, 1986, and 1991, based on prevalence estimates provided by the Center for Disease Control (CDC). They predicted that the core provider costs of AIDS would rise from US$ 630 million in 1985 to US$ 1.1 billion in 1986 and to US$ 8.5 billion in 1991. The authors compared their estimates of the cost of AIDS in the USA with the estimates for end-stage renal disease (US$ 2.2 billion), traffic accidents (US$ 5.6 billion), lung cancer (US$ 2.7 billion), and breast cancer (US$ 2.2 billion). They concluded that the core provider costs of AIDS were relatively low in comparison with the provider costs of all illness as well as the costs of these other diseases. However, they also assessed the non-care costs (e.g., for research) to rise from US$ 319 million in 1985 to US$ 542 million in 1986 and to US$ 2.3 billion in 1991.
Hellinger(1988,1991,1992) developed a model that estimated the provider costs of AIDS using incidence-based measures derived from US data. His findings for inpatients were similar to those by Scitovsky and Rice (1987). He added the costs of outpatient antiretroviral medications and estimated an increase of provider costs from US$ 5.8 billion in 1991 to US$ 10.4 billion in 1994 and to US$ 15.2 billion in 1995.
Solomon and Hogan (1992) analyzed the Michigan Medicaid payment records for the period of 1985-1989. Payments data were merged with data from the Michigan Death Registry and the AIDS Surveillance Registry. The payments rose steadily with age through the 36-45-years-old group, declining slightly among older adults. Men received services with average cost of US$ 1,522 per month, with a median monthly cost of US$ 792. Women received services with average cost of US$ 777 per month, with a median monthly cost of US$ 119. On average, the HIV-related health care provided through the Michigan Medicaid Program cost about US$ 1,300 per month per person, an amount substantially lower than other estimates. The authors pointed out that there were several possible reasons for that result, for example, that their data included many persons who were still in the early stages of infection, and high costs at the end stages of the disease had been excluded by the protocol. The bulk of the cost associated with treating persons with HIV infection was for inpatient hospital services. Among men the next highest cost was on average for drugs. Among women, physician fees were the next largest expense. A possible reason for the observed gender discrepancy could have been differences between gay males in intravenous drug users of both sexes.
The study of Anderson and Mitchell (1997) examined whether the AIDS-specific home and community-based waiver program, which was implemented in Florida in 1990 as an alternative to institutional care, was effective in reducing Medicaid expenditures per beneficiary during its first 2 years of operation. Therefore, the authors used Medicaid claims data and country information to estimate the effect of the waiver on expenditures controlling for nonrandom program selection. Their results indicate that persons with AIDS who used waiver services incurred monthly expenditures that were on average 22-27% lower than otherwise similar nonpar-ticipants. The authors concluded that home and community-based care for AIDS patients resulted in lower expenditures per beneficiary.
The economic impacts of HIV/AIDS disease have also been analyzed in Europe. Beck (1995) studied the AIDS-related costs in a national AIDS referral center in London. He concluded that share of total drug cost increased between 1985 and 1989 from 5.0% to 30.0%. The median survival time from the date of the diagnosis of AIDS was 14.6 months before the introduction of Zidovudine (1987) and 21.0 months afterwards.
Based on the years 1990/1991, Papaevangelou et al. (1995) calculated the cost of HIV/AIDS inpatients and outpatients in two hospitals in Athens, Greece. The average annual cost per person was estimated to be US$ 8,428, consisting of costs for Zidovudine (US$ 1,343), outpatient care (US$ 1,122), and inpatient care (US$ 5,963). A quite similar temporal data background was used by Rovura and Leidl (1995) for collecting data to estimate individual healthcare costs of HIV/AIDS patients in Catalonia, Spain, according to the disease stage. Annual patient cost for early stage, intermediate, and AIDS, respectively, was US$ 648, US$ 9,707, and US$ 24,915, whereby the share of inpatient cost increased from 0.15% in the early share to 84% during AIDS.
Kyriopoulos et al. (1995) compared different estimates of HIV/AIDS healthcare expenditure in Greece based on study years 1987-1993. Average annual cost per
Table 2 Examples of provider cost-of-illness in the pre-HAART era [US$ p.a.]
Country Study year Total provider Annual costs per year provider costs [*1,000,000] per patient
Scitovsky et al. (1986) Scitovsky and Rice (1987)
Was this article helpful?