University of California, San Francisco Logo

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

Two-year virologic outcomes of an alternative AIDS care model: evaluation of a peer health worker and nurse-staffed community-based program in Uganda
Global Health Sciences Literature Digest
Published September 7, 2009
Journal Article

Chang LW, Alamo S, Guma S, et al. Two-year virologic outcomes of an alternative AIDS care model: evaluation of a peer health worker and nurse-staffed community-based program in Uganda. J Acquir Immune Defic Syndr 2009;50:276-82.

In Context

Provision of antiretroviral therapy (ART) in resource-constrained areas is challenging, due in large part to insufficient numbers of trained clinical personnel and other workers.(1,2) Evaluation of alternative models of care that use nurses or trained lay workers or both is necessary to determine if such programs provide adequate care.


To evaluate the long-term clinical outcomes of the Reach Out Mbuya Parish HIV/AIDS initiative, a community- and faith-based AIDS program


Urban slums of Kampala, Uganda

Study Design

Retrospective cohort study


All patients aged 13 years or older who had initiated ART through the Reach Out program from October 1, 2003 to July 31, 2004


Independent predictors and frequency of virologic failure, CD4 count, and survival


The Reach Out program provides HIV care that consists of free home-based care; clinical care that includes ART; and comprehensive social services, such as microfinancing, provision of school fees, and food aid. Many of the services, including home-based care, are provided by trained lay health workers, many of whom are HIV-infected. The first-line care at the ART clinic is provided by nurses, with supervision by clinical and medical officers.

Patients were followed through April 1, 2006 and at that date were classified as active if they had attended the clinic in the last 90 days, lost to follow-up if they had not attended the clinic in the last 90 days and were last known to be on ART, transferred if there was clear documentation of the patient changing care sites, stopped therapy if they had not received ART for more than one week without restarting ART and were followed by Reach Out staff. Patients were classified as having died if they were on a list of deceased patients maintained by Reach Out staff through active follow-up and verbal autopsies. The end time point for data collection and analyses was the date the patient died; the date of viral load testing, if active; or the date of the last clinic visit if lost, transferred, or stopped therapy. If a patient was classified as active but no viral load testing was available, then April 1, 2006 was designated as their study end date.

Adherence was determined by pill count. Viral load testing was done on all patients and CD4 tests predominately were available. Baseline CD4 results were those that were obtained within two months of starting ART or, if not available, the most recent results prior to ART.


Of the 393 patients who were eligible for ART, 360 (92%) initiated therapy and 258 (72%) were still active and on treatment as of April 1, 2006. Patients who remained on therapy through April 2006 were significantly more likely to have been pregnant and to have initiated ART with combivir/efavirenz than those who did not remain on therapy.

Viral load test results were available for 95% of the patients, and 86% of these had undetectable levels at some time between April 2006 and January 2007. Of the 233 active patients with CD4 results, 222 (95%) increased their counts from baseline. The independent predictors of virologic failure, defined as detectable viral load of >400 copies/mL, were any prior use of ART (odds ratio [OR] 7.88, 95% confidence interval [CI]: 2.03-30.5) and either no change or a decline in CD4 count (OR 7.12, 95% CI: 1.63-31.1).

Eighty-five percent of patients remained on therapy at six months, 77% at one year, and 73% at two years. Patients not remaining on ART included those who had died, were lost to follow-up, or met the criteria for having stopped therapy. Factors associated with not remaining on therapy were a baseline CD4 count of <100 cells/mm3 (P=0.02), to have initiated ART using D4T/3TC/efavirenz regimen (P<0.0004), and meeting the WHO criteria for clinical stage 4 disease at the time of starting ART (P=0.06) all as tested with the log-rank test. These factors also were determined to be independently associated with not sustaining therapy (RH 1.87, 95% CI: 1.14-3.07 and RH 3.51, 95% CI: 1.49-8.29, respectively).

At six months, 87% of patients were alive, 84% at one year, and 82% at two years. Death was associated with a baseline CD4 count of <100 cells/mm3 (P=0.01), to have initiated ART using D4T/3TC/efavirenz regimen (P<0.001), and meeting the WHO criteria for clinical stage 4 disease at the time of starting ART (P=0.0003). All examined factors were independently predictive of death: baseline D4T/3TC/efavirenz regimen (RH 3.75 (95% CI: 1.19-11.8), baseline CD4 count of <100 cells/mm3 (RH 2.44, 95% CI: 1.24-4.81), and WHO criteria for clinical stage 4 disease at the time of starting ART (RH 3.75, 95% CI: 1.19-11.8).


HIV care systems that rely on nurses and lay health workers can provide care that is associated with excellent clinical outcomes and offers an alternative model for AIDS care in resource-constrained areas.

Quality rating

The authors described this study as a retrospective cohort. This is appropriate in terms of the analysis applied. However, it was not designed as research and, as such, does not meet the criteria for an excellent observational study. This reflects the limitations of applying such standards to evaluation of a program. There is no information on the representativeness of the patients or the classification of patients (remained on ART or not) and outcomes was not complete. The laboratory outcomes are likely to be accurate even if samples were not obtained at exact times. The follow-up time is reasonable. Despite the limitations, this study provides excellent programmatic data and outcomes.

Programmatic implications

Although not designed as a research study, the evaluation of Reach Out provides data in support of care for HIV-infected persons that do not rely heavily on professionals. Patients enrolled in this program had clinical results that were similar to those reported from other settings.(3,4) Two areas of concern highlighted by this paper are the association of prior ART use and WHO clinical stage 4 disease with worse outcomes. If poor outcomes from prior ART use reflect antiretroviral resistance, then surveillance for drug-resistant strains and enhanced efforts to improve adherence may be in order. Waiting to begin ART until stage 4 disease occurs is not currently the norm in developed countries and the worse outcomes associated with this practice highlight the need to begin treatment at less advanced stages of disease. The increased number of persons in need of ART will place demands on an already stressed system. Thus, the findings from this study may be of even greater importance by demonstrating an alternative care model with good outcomes.


  1. Van DammeW, Kober K, Laga M. The real challenges for scaling up ART in sub-Saharan Africa. AIDS 2006;20:653-6. Abstract not available.
  2. Colebunders R, Ronald A, Katabira E, et al. Rolling out antiretrovirals in Africa: there are still challenges ahead. Clin Infect Dis 2005;41:386-9.
  3. Bekker LG, Myer L, Orrell C, et al. Rapid scale-up of a community-based HIV treatment service: programme performance over 3 consecutive years in Guguletu, South Africa. S Afr Med J 2006;96:315-20.
  4. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions. Ann Intern Med 1999;131:81-7.