University of California, San Francisco Logo

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

Home > Global Health Literature Digest > Zimbabwe HIV control study
Impact and process evaluation of integrated community and clinic-based HIV-1 control: a cluster-randomised trial in eastern Zimbabwe
Global Health Sciences Literature Digest
Published June 4, 2007
Journal Article

Gregson S, Adamson S, Papaya S, Mundondo J, Nyamukapa CA, Mason PR, Garnett GP, Chandiwana SK, Foster G, Anderson RM. Impact and process evaluation of integrated community and clinic-based HIV-1 control: a cluster-randomised trial in eastern Zimbabwe. PLoS Med 2007 Mar 27;4(3):e102.


To determine whether a combined community- and clinic-based HIV prevention program (improved STI treatment and counseling, targeting of female sex workers and clients, free condoms, enhanced peer education) is feasible and can reduce HIV incidence in a maturing epidemic in Zimbabwe.

Study Design

Cluster-randomized controlled trial, with six pairs of matched communities. Communities within each pair were randomly assigned by coin toss to intervention or control. Households with eligible individuals were identified using census data. HIV infection and questionnaire information were collected at baseline (1998-2000) and three years later (2001-2003).


Manicaland Province, Eastern Zimbabwe. Six pairs of matched communities were selected: small town, tea/coffee estate, forestry plantation, roadside trading settlement, and subsistence farming area (two pairs). Each community included a government or mission health center.


At baseline, 5,943 individuals in intervention and 6,037 in control communities were enrolled. Eligible adults in all households were recruited, except that only one randomly selected partner in each marital couple was enrolled. Study inclusion criteria were: 1. Males aged 17-54; 2. Females aged 15-44; 3. Having slept in a household in the community for at least four nights in the previous month; and 4. Having done so at the same time one year earlier. At follow-up, 2,664 and 2,564 in the intervention and control arms, respectively, were evaluated.


Standard government services available to both intervention and control communities included syndromic STI management, social marketing and condom distribution from health clinics and government outlets, home-based care, and limited IEC (information, education, and communication) activities. The intervention program included three key components:

  1. Peer education and condom distribution among commercial sex workers (CSW) and targeting of male clients at the workplace and in the community; to reduce unmarried women’s dependence on sex work, micro-credit income-generating programs were devised, but could not be implemented because of economic problems in the country;
  2. Improved management of STIs at local health centers by regular training, supervision, and monitoring of healthcare staff; capacity to provide STI counseling services was also strengthened;
  3. and

  4. Open days with HIV/AIDS IEC activities at health centers to promote safer sexual behaviors and increase uptake of STI treatment services. Peer education activities were held weekly at workplaces and community sites, such as beer halls and markets.

Intervention strategies were implemented by two local non-governmental organizations and the Zimbabwe Ministry of Health and Child Welfare.

Primary Outcomes

The primary outcome was HIV incidence over three years among those who were HIV-uninfected at baseline. Secondary outcomes included self-reported STIs and treatment effectiveness, sexual and health-seeking behavior change, and HIV/AIDS knowledge. Data on sexual partnerships and condom use were collected using a simple secret voting-box system, in which 75% of respondents were selected at random to participate. Process indicators included program coverage and quality.


The proportion of baseline respondents evaluated at follow-up was 55% in the intervention and 56% in the control communities; 23-26% of baseline participants who were still alive had moved, and among those remaining, 25-23% were unavailable for the follow-up survey. At baseline, HIV prevalence was 24% in intervention and 21% in control communities (p<0.001).The overall HIV-1 incidence rate was 1.77/100 per person years (ppy). Comparing intervention to control communities, the HIV incidence rate ratio (IRR) was 1.27 (95% CI 0.92–1.75). At follow-up, there was no difference in self-reported STIs or high-risk sexual behavior between trial-arm communities, although there was a reduction in STIs among men in the intervention sites. Males who attended program meetings had a lower HIV incidence (IRR 0.48, 95% CI, 0.24–0.98) and were less likely to report unprotected sex with casual partners (OR 0.45; 95% CI, 0.28–0.75). In contrast, young women in the intervention communities were more likely to have unprotected sex, and to have initiated sex. There were no differences in consistent condom use between intervention and control groups. More than 60,000 peer-education meetings were held, and 6.8 million condoms distributed. Coverage of training in syndromic STI management and counseling was high. Data reported from clinics on numbers of STI episodes treated revealed that STI cases fell from 51% to 66% in intervention communities, and by 7%- 52% in control communities.


The authors conclude that the intervention reduced high risk behavior, HIV, and STIs among men attending meetings, but did not translate into a reduction in population-level HIV incidence. The authors examined many possible reasons for this. There may have been inadequate statistical power to detect a difference, and/or changes in bridge populations may take longer to result in a change in the population. It is likely that the failure of the intervention to alter behaviors among women was due to the persistence of commercial sex among peer educators, which undermined their position as role models. Transactional sex continued out of economic necessity and the inability to implement the micro-credit programs.

Quality Rating

According to the Jadad criteria, this study received a good rating. The authors explained the randomization process, and there was a thorough description of loss to follow-up. However, some details were lacking. How initial clusters (communities) were selected was not described. Measures of effectiveness of STI treatment, recurrence, and incidence are based only on self-report, include GUD information among men and not women, were not validated by any diagnostic testing, and are highly non-specific. For HSV2, likely the most prevalent STI in the region, recurrence of symptoms does not necessarily indicate an incident infection, and cessation of symptoms does not mean appropriate treatment. Interpretation of changes in numbers of STIs treated in clinics is unclear, since the intervention was partly designed to increase uptake of services.

In Context

Community-based STI and peer interventions have been implemented and evaluated in a number of settings without any clear or consistent indication of effectiveness. Syndromic management of sexually transmitted infections (STIs) proved effective in an HIV epidemic in northwest Tanzania,(1) and peer education showed promise in early process evaluations.(2) However, subsequent RCTs of syndromic management (3) and mass treatment of STIs,(4) together with an information, education, and communication (IEC) behavior-change program,(5) were unable to demonstrate an impact in mature HIV epidemics.

Programmatic Implications

It is disappointing that this intervention was unable to demonstrate an effect. However, this may have been due to a number of factors, including the inability to address high-risk and transactional sex among women. HIV prevention efforts cannot ignore the importance of addressing the need of women for economic support. Because of the lack of specificity in STI treatment and diagnosis, particularly in identification and/or management of HSV2, results of this trial do not shed much light on the usefulness of improving STI services. However, it is consistent with previous community-based studies suggesting that in mature HIV epidemics, STI programs may not have much impact on HIV at the population level. The fact that this intervention didn’t work, despite three years of effort, 60,000 peer-encounters, and an unreported amount of resources, should be an incentive for researchers to address more critically the factors that support risk behavior, and to pay attention to HSV2 management.


  1. Grosskurth H, Mosha F, Todd J, Mwijarubi E, Klokke A, Senkoro K, et al. Impact of improved treatment of sexually transmitted diseases on HIV infection in rural Tanzania: randomised controlled trial. Lancet 1995 Aug 26;346(8974):530-6.
  2. Dube N, Wilson D. Peer education programs among HIV-vulnerable communities in Southern Africa. HIV/ AIDS management in southern Africa: priorities for the mining industry. Johannesburg Epidemiology Research Unit 1996; 107–110. (No abstract available.)
  3. Machekano R, McFarland W, Mbizvo MT, Bassett MT, Katzenstein D, Latif AS. Impact of HIV counselling and testing on HIV seroconversion and reported STD incidence among male factory workers in Harare, Zimbabwe. Cent Afr J Med 1998 Apr;44(4):98-102.
  4. Kamali A, Quigley M, Nakiyingi J, Kinsman J, Kengeya-Kayondo J, Gopal R, et al. Syndromic management of sexually-transmitted infections and behaviour change interventions on transmission of HIV-1 in rural Uganda: a community randomised trial. Lancet 2003 Feb 22;361(9358):645-52.
  5. Wawer MJ, Sewankambo NK, Serwadda D, Quinn TC, Paxton LA, Kiwanuka N, et al. . Control of sexually transmitted diseases for AIDS prevention in Uganda: a randomised community trial. Rakai Project Study Group. Lancet 1999 Feb 13;353(9152):525-35.