Review
The direct costs of HIV/AIDS care

https://doi.org/10.1016/S1473-3099(06)70413-3Get rights and content

Summary

We reviewed published studies reporting the direct medical costs of treating HIV-infected people in countries using highly active antiretroviral therapy (HAART). Of 543 potentially relevant studies, only nine provided adequate data to make a meaningful statement about costs. Within studies, people with more advanced disease incurred higher total costs. Valid comparisons of total direct medical costs between studies were not possible because of differences in the specific components included, the heterogeneous nature of study populations in terms of disease stage, the sources and methods used to estimate unit costs, and the level of aggregation at which results were reported. The advent of HAART has major implications for the cost of treating HIV-infected individuals. Although this information is important for planning purposes, only a small number of published studies provide useful estimates of the direct cost. A useful method of estimating resource use and costs is computer simulation.

Introduction

AIDS is characterised by the progressive destruction of a person's immune system and is the late and most serious stage of HIV infection. Over the past two decades, the epidemic has become a major challenge to health-care systems, with more than 20 million people dying from HIV/AIDS and a further 40·3 million people worldwide estimated to be infected in 2005.1 Recent advances in treatment, especially the use of potent combinations of nucleoside reverse transcriptase inhibitors, protease inhibitors, and non-nucleoside reverse transcriptase inhibitors (referred to as highly active antiretroviral therapy, or HAART), have resulted in dramatic reductions in the rates of HIV disease progression, opportunistic infections, hospital admissions, and deaths.2, 3, 4, 5 In developed countries, the remarkable success of HAART has also altered the use of health-care resources for HIV-infected patients, with the nature of treatment shifting from inpatient hospital admissions to outpatient care and greater use of medications in the outpatient setting.6 Furthermore, HIV prevalence is increasing in western countries as a consequence of falling death rates.1, 7, 8

These changes have occurred against a background of little empirical evidence about the direct costs of treating HIV-infected people in terms of per-person annual expenditures and lifetime costs,9, 10 the implications on the cost of using different treatment strategies,11, 12 or the extent to which costs vary by stage of HIV infection.13 Information on the direct costs of treatment is important since it provides a basis for health planners to allocate budgets or reimburse specific categories of expenditures. It also enables policymakers, when faced with changes in prevalence—eg, that currently occurring with the spread of HIV infection among injection drug users14—to make more informed choices between programmes. To optimally allocate scarce economic resources to prevention and treatment programmes, planners require accurate, up-to-date estimates of the direct costs of treatment.15, 16, 17 Although economic evaluation is an important approach for establishing priorities for health interventions,6, 18, 19 in practice this type of evaluation has been of limited value in HIV/AIDS because of the paucity of accurate cost data.9

The purpose of this review is to identify published estimates of the direct medical costs, including hospital inpatient, outpatient, and medication costs for treating HIV-infected individuals after HAART was introduced into routine clinical practice, to determine the extent to which these estimates can be validly compared, and to make such comparisons where possible. In cases where meaningful comparisons are not possible, we explain why this is the case. We also discuss the implications for future research in this area.

Section snippets

Methods

To be included, studies had to meet pre-specified criteria for information content. We identified studies that included an original estimate of the mean monthly or annual direct medical costs of treating an HIV-infected individual. We included studies from English language peer-reviewed published work that provided enough detail to allow meaningful comparisons by explicitly describing the study period, location and population (demographic and clinical distributions), types of treatments, data

Results

Of 543 titles initially identified, 92 were flagged as potentially relevant. Based on the abstracts, full versions of 33 were obtained. Another six relevant publications were cited within retrieved articles. After reviewing each study, nine were judged to meet the inclusion criteria.

Table 1 summarises the major characteristics of each study. The studies ranged in the number of included people from 7421 to 5708,22 and one study was considered population based23 because it included all patients

Discussion

We found substantial variation in studies reporting the total and component-specific direct medical costs of treating HIV-infected individuals in the HAART era. Valid comparisons of the estimates from the nine studies reviewed were not possible because of differences in the specific components included, the heterogeneous nature of the study populations in terms of disease stage, the sources and methods used to estimate unit costs, and the level of aggregation at which results were reported.

Search strategy and selection criteria

Relevant studies were identified from Medline, the Web of Science, and additional references cited within the articles retrieved. Keywords were “HIV” or “AIDS”, cross-referenced with “cost,” “expenditure,” or “utilisation.” Dates for inclusion were between January 1996 and June 2005. Abstracts were not considered.

References (36)

  • FJ Hellinger

    HIV patients in the HCUP database: a study of hospital utilization and costs

    Inquiry

    (2004)
  • L Crane et al.

    Hospital and outpatient health services utilization among HIV-infected patients in care in 1999

    J Acquir Immune Defic Syndr

    (2002)
  • HB Krentz et al.

    Impact of practice changes on an antiretroviral budget in an HIV care program

    Disease Management and Health Outcomes

    (2005)
  • MW Tyndall et al.

    Intensive injection cocaine use as the primary risk factor in the Vancouver HIV-1 epidemic

    AIDS

    (2003)
  • Priorities for action in managing the epidemics. HIV/AIDS in BC: 2003–2007

  • DR Holtgrave et al.

    Economic evaluation of HIV prevention programs

    Annu Rev Public Health

    (1996)
  • DR Holtgrave et al.

    Updates of cost of illness and quality of life estimates for use in economic evaluations of HIV prevention programs

    J Acquir Immune Defic Syndr Hum Retrovirol

    (1997)
  • MF Drummond et al.
  • Cited by (69)

    • A Cost-of-Illness Study of Patients with HIV/AIDS in Bogotá, Colombia

      2017, Value in Health Regional Issues
      Citation Excerpt :

      We observed no significant difference between CDC classifications and higher costs for CDC stage 1 or for CD4 cell count. This result contradicts international findings that suggest that individuals at more advanced stages of the disease incur higher costs [22,23,25]. The small sample size of our study (especially in CDC stages 1 and 2) could explain this nonsignificant result.

    • Estimation of the Direct Cost of HIV-Infected Patients in Greece on an Annual Basis

      2014, Value in Health Regional Issues
      Citation Excerpt :

      As a matter of fact, the aforestated studies used a different approach for the estimation of HIV infection’s direct cost. As a result, comparing the findings of the different studies seems quite challenging; however, some general conclusions can certainly be drawn [22,23]. The present survey seems to be in line with the European studies regarding the overall results; nonetheless, some differences can be distinguished.

    • What are the drivers of high-cost HIV patients?

      2021, International Journal of STD and AIDS
    View all citing articles on Scopus
    View full text