University of California, San Francisco Logo

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

Home > Global Health Literature Digest > Impact of Stepping Stones
Impact of Stepping Stones on incidence of HIV and HSV-2 and sexual behaviour in rural South Africa: cluster randomized controlled trial
Global Health Sciences Literature Digest
Published February 10, 2009
Journal Article

Jewkes R, Nduna M, Levin J, et al. Impact of Stepping Stones on incidence of HIV and HSV-2 and sexual behaviour in rural South Africa: cluster randomized controlled trial. BMJ 2008;337:1-11.

In Context

Few well controlled studies about the impact of a behavioral intervention on HIV and STI incidence or risk behaviors, or both, have found evidence of efficacy.(1,2,3) A possible limitation to previous studies is that such interventions must be in place for a substantial period of time before formal evaluation, because programs may improve over time.


To assess the effect of Stepping Stone, a behavioral intervention that has been used since 1995 and in more than 40 countries, on HIV and HSV-2 incidence and sexual risk behaviors


Eastern Cape, South Africa, a rural area comprised of subsistence farms, two moderately sized towns, seven small towns, and several villages

Study Design

A cluster-randomized controlled trial


Male and female residents of the selected areas, aged 15-26 years, who were primarily students


Main outcome measure was incidence of HIV. Secondary outcomes were HSV-2, unwanted pregnancy, sexual practices, depression, and substance abuse.


The 70 study clusters consisted of six townships and 64 villages. Each cluster was located 10 km from each other, had a junior or senior secondary school, and had a community that was willing to participate. The clusters were divided into seven strata: one comprising the townships and six comprising the villages grouped by proximity to selected roads. Study arm allocation was done using computer-generated randomization with equal number of villages assigned to each arm in each stratum.

Twenty men and 20 women were recruited from each cluster. Recruitment began with a community meeting called by the village chief. Study personnel attended the meeting and explained the study, then went to the local schools and recruited participants to attend a meeting. Approximately 60 potential participants attended each meeting. At the meetings, the group identified 40 of the participants to enroll in the study. Selection was based upon the group's decision regarding who was most likely to participate.

Stepping Stones is a 50-hour, participatory learning intervention to reduce incidence of HIV by building knowledge and skills in critical reflection, communications, and risk awareness. The intervention uses role-play and drama and brings into play the reality of participants day-to-day lives. The intervention is delivered to single-sex groups that are run in parallel. There are a total of thirteen 3-hour sessions complemented by three peer-group meetings and a final community meeting. Sessions were held primarily on school grounds after classes had ended. Topics included behaviors and factors that shape actions, sex and love, conception and contraception, risk taking and sexual problems, unwanted pregnancy, STIs and HIV, safer sex and condoms, gender-based violence, motivations for sexual behavior, dealing with grief and loss, and communication skills.

Clusters were randomized to this intervention or to a three-hour control intervention on HIV and safer sex.

Contact information, questionnaires and blood specimens were collected at baseline and at 12 and 24 months after the intervention. The questionnaire ascertained information on sexual behaviors (e.g. number of partners, condom use), history of sexual abuse and violence, drug use, depression, and unwanted pregnancy. Blood was tested for HIV and HSV-2 antibodies.

An intention-to-treat-analysis was conducted. HIV and HSV-2 incidence were measured separately for men and women. Analyses were conducted using generalized linear mixed models and generalized estimating equations.


There were 715 women and 694 men enrolled in the intervention group and 701 women and 666 men in the control arm. Twelve-month follow-up for these groups was 75.8% (women intervention), 75.3% (women control), 75.1% (men intervention), and 71.8% (men control). Follow-up at 24 months was 73.1% (intervention) and 76.0% (control) of women with baseline HIV results were retested and 69.5% (intervention) and 69.2% (control) of men were tested again for HIV.

Attendance at the intervention session was not complete; 16.8% of men and 12.5% of women did not attend any session. Only 27.5% of men and 25.4% of women attended all of the sessions. Characteristics of participants in the intervention and control arms were similar. HIV incidence was 3.46 per 100 person-years in the intervention group and 4.07 per 100 person-years in the control group and was markedly higher for women (5.65 per 100 person-years in the intervention group and 6.95 per 100 person-years for the control group). The incidence of HSV-2, however, was significantly lower in the intervention group (incidence rate ratio 0.67, 95% confidence interval 0.46 to 0.97, P=0.036).

Of the other outcomes of interest, none differed between the two arms among women at 24 months. For men in the intervention group, the proportion reporting inflicting intimate partner violence had declined significantly at 24 months.


The Stepping Stones intervention had no effect on HIV incidence but was associated with a decrease in HSV-2 incidence and perpetration of intimate partner violence by men. Stepping Stones had little effect on other risk behaviors.

Quality Rating

The study used a cluster randomization in which randomization was done blindly. Follow-up was adequate for this type of study and was equal between arms. Blinding of the assessor was not evident. The control condition was not described. The analysis was done using an intention-to-treat-approach. The study used biological outcomes, enhancing its strength. Overall, the study was of fair quality.

Programmatic Implications

This is an important study because it evaluated the effectiveness of a widely used intervention, something not routinely done. It is worth noting that the intervention evaluated has been in use since 1995 and has been applied in a large number of countries. The lack of impact of Stepping Stones on HIV incidence calls into question its continued use as an HIV prevention tool. The intervention requires multiple sessions; the large proportion of participants who did not complete the program highlights the need for single-session prevention programs. Of note, HSV-2 incidence declined after the intervention, possibly due to differences in its transmissibility, including in the types of exposure in which transmission can occur.

Also of note, the intervention reduced the proportion of men who reported perpetrating intimate partner violence. Intimate partner violence is a known risk factor for HIV;(4) as such, Stepping Stones may have an indirect effect on HIV through its reduction in intimate partner violence.


  1. Kamali A, Quigley M, Nakiyingi J, et al. Syndromic management of sexually-transmitted infections and behavior change interventions on transmission of HIV-1 in rural Uganda: a community randomized trial. Lancet 2003;361:645-52.
  2. Ross DA, Changalucha J, Obasi A, et al. Biological and behavioural impact of an adolescent sexual health intervention in Tanzania: a community randomised trial. AIDS 2007;21:1943-55.
  3. Pronyk P, Hargreaves JR, Kim JC, et al. Effect of a structural intervention for the prevention of intimate partner violence and HIV in rural South Africa: a cluster randomised trial. Lancet 2006;368:1973-83.
  4. Dunkle KL, Jewkes RK, Brown HC, Gray GE, McIntryre JA, Harlow SD. Gender-based violence, relationship power and risk of prevalent HIV infection among women attending antenatal clinics in Soweto, South Africa. Lancet 2004;363:1415-21.