University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Mortality in HIV-infected Ugandan adults receiving antiretroviral treatment and survival of their HIV-uninfected children: a prospective cohort study
Global Health Sciences Literature Digest
Published June 2, 2008
Journal Article

Mermin J, Were W, Ekwaru JP, Moore D, Downing R, Behumbiize P, et al. Mortality in HIV-infected Ugandan adults receiving antiretroviral treatment and survival of their HIV-uninfected children: a prospective cohort study. Lancet 2008 Mar 1;371(9614):752-9.


To measure the effect of cotrimoxazole prophylaxis alone and cotrimoxazole plus antiretroviral therapy (ART) on the number of hospitalizations in and survival of HIV-infected adults and their HIV-uninfected children


Rural Uganda

Study Design

Prospective cohorts conducted over three study periods


HIV-infected clients from an AIDS support organization and their household members


Hospitalizations and death among adults, orphanhood, death in HIV-negative children, and the number of persons that would need treatment to prevent the outcomes


This study is an analysis of data collected in a prospective cohort that included three different interventions and outcomes.

Study period 1, April 2001 through March 2003: HIV-infected clients from the Tororo Branch of The AIDS Support Organization (TASO) and their household members were enrolled. HIV testing and results disclosure were offered to all subjects not known to be infected at enrollment. Subjects were followed up with weekly household visits by study staff for a median of 5.5 months. Subjects with confirmed HIV-positive or with HIV-negative tests were included. All subjects were randomly assigned to receive hygiene education along with dilute chlorine to be used to make water potable or to receive education only.

Study period 2, September 2001 through March 2003: Study period 2 extended the safe water and education intervention to include cotrimoxazole prophylaxis for the HIV-infected subjects.

Study period 3, May 2003 through December, 2005: All clinically or immunologically eligible subjects (CD4 count <250 cells/mm3 or pre-2006 WHO clinical stage 3 or 4 or recurrent herpes zoster) who participated in the first two study periods and additional subjects recruited from TASO were provided with ART. All households were provided with insecticide treated bed nets. Subjects were randomly assigned to one of three ART monitoring strategies: 1) clinical monitoring plus quarterly CD4 cell counts and viral load measurements, 2) clinical monitoring and quarterly CD4 counts, or 3) clinical monitoring. Follow-up to enable comparison of these three strategies on the outcomes of interest is underway.

Patients received weekly visits from study staff who administered a questionnaire about adherence to medication, drug toxicity, or death of a household member since the previous visit. Subjects who were ill at the follow-up visit were encouraged to seek care at the clinic, which was available at no cost to the participant. Routine visits were not conducted after enrollment. The HIV-infected subjects were taken to the clinic if during the home visit they were found to have any specific symptoms or severe illness. Blood samples were collected quarterly at the home visit and used to measure CD4 counts and viral loads.

Statistical analysis: Hospitalization and mortality were compared between study period 3 (during which subjects used both cotrimoxazole and ART) and study period 2 (cotrimoxazole alone), and period 1 (neither treatment). The start of the study period for each person was the date they began medication. Adjustments were made using the most recent CD4 count obtained prior to starting each of the treatments. The independent effect of the medications on death was measured using the Cox proportional hazards model. The rate of hospitalization was measured using a Poisson regression model with a log-link function adjusted for age and CD4 count and quarterly seasonal periods. Generalized estimating equations methods with an exchangeable correlation structure for the analysis of repeated measures in the same participants were used. To determine the effect on the prevention of orphanhood, children with only one HIV-infected parent were included. The number of HIV-infected persons that would need to receive treatment to prevent an outcome only for the factors found to be associated in the multivariate analysis.


Enrolled in the first study period were 466 HIV-infected adults and 1481 HIV-negative household members. In the second study period were 399 HIV-infected adults who initiated prophylaxis with cotrimoxazole and 318 HIV-negative household members. For the third study period, there were 138 ART-eligible subjects and 318 HIV-negative household members from the previous study plus an additional 907 newly recruited ART-eligible participants and 3120 newly recruited uninfected household members. This resulted in 1045 participants who received ART and cotrimoxazole and 3438 HIV-negative household members.

Participants from the first two study periods and the newly enrolled subjects for study period 3 were similar in age and CD4 cell count, but the subjects who continued on into period 3 from the previous study period had lower median CD4 cell counts. Household members in the three study periods did not differ in demographic characteristics.

The median follow-up for study period 1 was 154 days, during cotrimoxazole therapy was 532 days, and during ART and cotrimoxazole was 749 days. The median number of HIV-negative household members was three, their median age was 11 years, and 51% were female.

Of the 1373 HIV-infected participants, 233 (17%) and 40 (1%) of the 4601 household members died. To adjust for the interaction between the duration on ART and the effect of ART on mortality, ART data were divided into two time periods (first 16 weeks and after 16 weeks on ART) and analyzed as time-dependent covariates. Comparing ART plus cotrimoxazole to cotrimoxazole alone resulted in a 55% reduction in death during the first 16 weeks of treatment and a 92% reduction after the first 16 weeks on therapy. Compared to no treatment (study period 1), mortality was 95% less in the group treated with ART and cotrimoxazole.

For every 14 people treated with ART and cotrimoxazole for the first 16 weeks compared to co-trimoxazole alone, one year of life was saved and for treatment beyond 16 weeks, one year of life was saved for every eight people treated. The benefit associated with treatment held across all CD4 cell count strata, but the effect was more pronounced for those with lower CD4 cell counts; only two people would need to be treated for one year of life saved for those with CD4 counts, 50 cells/mm3 compared to 18 for those with CD4 counts over 200 cells/mm3.

There were 1935 children (age less than 18 years) who had only one living parent and the parent was HIV-infected and 191 had two parents living with HIV. Treatment of parents with ART and cotrimoxazole during the first 16 weeks reduced orphanhood by 78% when compared to treatment with cotrimoxazole alone and by 93% when treatment extended beyond 16 weeks.

There were 1676 hospital admissions among the HIV-infected participants. The frequency of admissions was reduced by 43% during the ART plus cotrimoxazole treatment period when compared to cotrimoxazole alone and by 64% when compared to no treatment.


This home-based treatment program substantially reduced mortality among HIV-infected adults and their HIV-uninfected children. Treatment also was associated with lower rates of hospitalization and orphanhood. Home-based provision of ART and cotrimoxazole with follow-up by trained lay-persons can substantially reduce mortality in HIV-infected adults and their HIV-uninfected children and reduce the number of orphans arising from the HIV epidemic in rural sub-Saharan Africa.

Quality Rating

This was a well-designed study in which the exposed and unexposed (HIV-positive and HIV-negative) persons were comparable. However, the study population is not clearly representative of the general population because it was conducted in a rural setting and the initial sample was obtained from HIV-infected persons using a community AIDS support program. The inclusion of household members of these initial participants adds strength to the study. Possible participation bias was not accounted for as the results did not include information on persons who declined enrollment in the study. The study relied on physical examination and laboratory tests to document both HIV infection and degree of immunocompromise, adding strength to the study. The study did not compare the characteristics of the infected and uninfected subjects. Mortality was determined through routine, structured interviews and hospital admissions were obtained by review of medical records. Follow-up was appropriate as were the statistical analyses, although information on censor date was not included. Overall this study is of good quality.

In Context

The reduced mortality associated with treatment in this home-based study is similar to that reported from facility-based studies in Europe,(1) Africa,(2,3,4,5,6,7) and Haiti.(8)

Programmatic Implications

The improved survival observed in this study provides evidence that HIV treatment can be provided without ongoing medical care by trained professionals in medical settings. For rural areas of Africa, this opens the door for expanded treatment to the large number of treatment-eligible persons who are not yet receiving ART.


  1. Sterne JA, Hernan MA, Ledergerber B, et al. Long-term effectiveness of potent antiretroviral therapy in preventing AIDS and death: a prospective cohort study. Lancet 2005; 366(9483):378-84.
  2. Wools-Kaloustian K, Kimaiyo S, Diero L, et al. Viability and effectiveness of large-scale HIV treatment initiatives in sub-Saharan Africa: experience from western Kenya. Aids 2006;20(1):41-8.
  3. Badri M, Bekker LG, Orrell C, Pitt J, Cilliers F, Wood R: Initiating highly active antiretroviral therapy in sub-Saharan Africa: an assessment of the revised World Health Organization scaling-up guidelines. Aids 2004;18(8):1159-68.
  4. Stringer JS, Zulu I, Levy J, et al. Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. Jama 2006; 296(7): 782-93.
  5. Spacek LA, Shihab HM, Kamya MR, et al. Response to antiretroviral therapy in HIV-infected patients attending a public, urban clinic in Kampala, Uganda. Clin Infect Dis 2006;42(2):252-9.
  6. Braitstein P, Brinkhof MW, Dabis F, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high-income countries. Lancet 2006;367(9513):817-24.
  7. Akileswaran C, Lurie MN, Flanigan TP, Mayer KH: Lessons learned from use of highly active antiretroviral therapy in Africa. Clin Infect Dis 2005; 41(3): 376-85.
  8. Severe P, Leger P, Charles M, et al. Antiretroviral therapy in a thousand patients with AIDS in Haiti. N Engl J Med 2005;353(22):2325-34.