Elsevier

The Lancet

Volume 371, Issue 9614, 1–7 March 2008, Pages 752-759
The Lancet

Articles
Mortality in HIV-infected Ugandan adults receiving antiretroviral treatment and survival of their HIV-uninfected children: a prospective cohort study

https://doi.org/10.1016/S0140-6736(08)60345-1Get rights and content

Summary

Background

Antiretroviral therapy (ART) is increasingly available in Africa, but physicians and clinical services are few. We therefore assessed the effect of a home-based ART programme in Uganda on mortality, hospital admissions, and orphanhood in people with HIV-1 and their household members.

Methods

In 2001, we enrolled and followed up 466 HIV-infected adults and 1481 HIV-uninfected household members in a prospective cohort study. After 5 months, we provided daily co-trimoxazole (160 mg trimethoprim and 800 mg sulfamethoxazole) prophylaxis to HIV-infected participants. Between May, 2003, and December, 2005, we followed up 138 infected adults who were eligible and 907 new HIV-infected participants and their HIV-negative household members in a study of ART (mainly stavudine, lamivudine, and nevirapine). Households were visited every week by lay providers, and no clinic visits were scheduled after enrolment. We compared rates of death, hospitalisation, and orphanhood during different study periods and calculated the number needed to treat to prevent an outcome.

Findings

233 (17%) of 1373 participants with HIV and 40 (1%) of 4601 HIV-uninfected household members died. During the first 16 weeks of ART and co-trimoxazole, mortality in HIV-infected participants was 55% lower than that during co-trimoxazole alone (14 vs 16 deaths per 100 person-years; adjusted hazard ratio 0·45, 95% CI 0·27–0·74, p=0·0018), and after 16 weeks, was reduced by 92% (3 vs 16 deaths per 100 person-years; 0·08, 0·06–0·13, p<0·0001). Compared with no intervention, ART and co-trimoxazole were associated with a 95% reduction in mortality in HIV-infected participants (5 vs 27 deaths per 100 person-years; 0·05, 0·03–0·08, p<0·0001), 81% reduction in mortality in their uninfected children younger than 10 years (0·2 vs 1·2 deaths per 100 person-years; 0·19, 0·06–0·59, p=0·004), and a 93% estimated reduction in orphanhood (0·9 vs 12·8 per 100 person-years of adults treated; 0·07, 0·04–0·13, p<0·0001).

Interpretation

Expansion of access to ART and co-trimoxazole prophylaxis could substantially reduce mortality and orphanhood among adults with HIV and their families living in resource-poor settings.

Introduction

Antiretroviral therapy (ART) is the most effective clinical intervention for reduction of mortality in people with HIV-1 infection. It is increasingly available in the developing countries where 90% of HIV-infected people live, including 63% in Africa.1 However, despite substantial efforts, most HIV-infected people in Africa who would benefit from ART do not have access to it.1 Availability in many areas is constrained by the high cost of medication,2, 3, 4 inadequate numbers of trained health-care providers,5, 6 poorly equipped clinics,7 and distance to health centres.8, 9 Effective ART programmes require high adherence to medication,10 attention to potential drug toxicity, and continuing diagnosis and treatment of opportunistic infections. Ideally, programmes in Africa would provide these services with limited use of physicians and minimum transportation requirements.5 Use of trained lay providers to deliver ART to HIV-infected people at their homes, collect standard health information, and refer patients for selected symptoms could potentially avoid adherence problems stemming from inadequate transportation to clinics and could reduce crowding at health centres.

Although several studies provide information about survival and changes in immunological and virological markers during ART in patients in Africa,11 an assessment of ART effectiveness requires a comparison group and a carefully followed up cohort because randomised trials would be unethical.12, 13 Data for effectiveness of highly active antiretroviral therapy (HAART) from developed countries are few because contemporaneous comparison groups were taking dual treatment at the time HAART became available. In Africa, initial introduction of ART has usually been as HAART. Insight into the effectiveness of HAART in Africa could enable improved decision making by individuals, governments, and donor agencies.

We analysed data from two prospective cohort studies in rural Uganda—the first of co-trimoxazole prophylaxis and the second of HAART. The main purpose of the randomised HAART study was to evaluate three different treatment monitoring strategies. Data aggregated across monitoring groups were used to assess the effect of a home-based ART programme on mortality, hospital admissions, and orphanhood in people with HIV and their families. We assessed the effect of adding HAART to co-trimoxazole prophylaxis and the effect of HAART and co-trimoxazole compared with the time before either intervention.

Section snippets

Participants and study design

HIV-infected participants (18 years or older) at the Tororo Branch of The AIDS Support Organisation (TASO), including those from Tororo, Busia, and Mbale districts in Uganda were enrolled. In study period 1, initiated in April, 2001, we followed up participants with home visits for a median of 5 months (figure 1). In study period 2, all participants with HIV without previous adverse reactions to sulfonamides were provided daily co-trimoxazole (160 mg of trimethoprim and 800 mg of

Results

Figure 1 shows the study profile. Age and median CD4-cell count differed across study periods (table 1). Age and sex were similar for participants who newly enrolled during period 3 and those who had also participated in period 2 (data not shown); however, median CD4-cell count for newly enrolled participants in period 3 was lower (122 cells per μL vs 143 cells per μL, p=0·02). Median follow-up before co-trimoxazole was 154 (IQR 147–161) days, during co-trimoxazole was 532 (488–542) days, and

Discussion

A home-based ART and co-trimoxazole programme was associated with a greater than 90% reduction in mortality in adults with HIV living in rural Uganda. These results were achieved even though no routine clinic visits were scheduled after initial enrolment, and home visits were provided by trained lay providers. Provision of ART to adults was also associated with a large reduction in mortality in their HIV-negative children, and with substantial reductions in the rate of orphanhood.

The overall

References (33)

  • G Pappas et al.

    Access to basic HIV/AIDS related clinical services in Uganda: results of a nationally-representative health facility survey

    AIDS Public Policy J

    (2006)
  • S Lwasa

    Planning for health infrastructure in Uganda: where is the need?

  • A Calmy et al.

    Simplifying and adapting antiretroviral treatment in resource-poor settings: a necessary step to scaling-up

    AIDS

    (2004)
  • DR Bangsberg et al.

    Non-adherence to highly active antiretroviral therapy predicts progression to AIDS

    AIDS

    (2001)
  • C Akileswaran et al.

    Lessons learned from use of highly active antiretroviral therapy in Africa

    Clin Infect Dis

    (2005)
  • E Losina et al.

    The independent effect of highly active therapy on severe opportunistic disease incidence and mortality in HIV-infected adults in Côte d'Ivoire

    Antivir Ther

    (2007)
  • Cited by (172)

    View all citing articles on Scopus
    View full text