Impact of peer education on HIV/AIDS behaviour change among secondary school youths: a static group comparison analysis of a peer education project in Rachuonyo County, Kenya

BACKGROUND: By the end of 2011, about 1·6 million Kenyans were living with HIV; this number is expected to reach 1·8 million by 2015, mainly due to new infections. The age bracket 15–24 years is a crucial period for interventions such as peer education to prevent new infections and save future generations from the scourge. The aim of this study was to determine the difference between peer education club members (beneficiaries) and non-members (non-beneficiaries) in the adoption of preventive behaviours including abstinence, faithfulness to a partner, condom use, and HIV testing. METHODS: The study covered eight public secondary schools in Rachuonyo County, where peer education clubs had been operational for 2 years. We applied a static group comparison design to guide the research process, and primary data were sourced from 260 beneficiaries and 212 non-beneficiaries. We used club membership and class registers to develop sampling frames for beneficiaries and non-beneficiaries, respectively. In order to select participants, we applied a systematic random sampling procedure, and to determine sample sizes we used Fisher's formula. The county education authority and school boards were informed and approved of the study. Participants provided their consent to ensure voluntary participation. Inclusion in the sample was based on the grade level and 1-year minimum membership to a peer education club. Quantitative analysis techniques included cross-tabulations with χ² statistic, beta coefficients (β), and odds ratios (Exp [β]). FINDINGS: The study found that 27 beneficiaries (10·4%) and eight non-beneficiaries (3·8%) were not sexually active and were therefore likely to be practising abstinence. Controlling for gender, age, religion, orphanhood status, and schooling consistency, the odds ratios indicated that beneficiaries were about 1·1 times more likely to embrace abstinence than non-beneficiaries (β=0·136; p=0·068; CI=90%). Among those who reported sexual activity, 166 beneficiaries (71·2%) and 110 non-beneficiaries (53·9%) reported sexual relations with only one partner over the preceding 18-month period. The beneficiaries were about 2·4 times more likely to practise faithfulness than non-beneficiaries (β=0·856; p=0·021; CI=95%). Up to 142 beneficiaries (60·9%) compared with 99 non-beneficiaries (48·5%) used a condom during the last sexual encounter. The beneficiaries were about twice as likely to use condoms consistently as compared with non-beneficiaries (β=0·737; p=0·037; CI=95%). Furthermore, beneficiaries were about 1·3 times more likely to have been tested for HIV than non-beneficiaries (β=0·269; p=0·051; CI=90%). INTERPRETATION: The peer education project encouraged abstinence, faithfulness to a partner, condom use, and HIV testing. Given the results of our study, we suggest that stakeholders should take the following steps: pursue the integration of peer education into school extracurricular activities, identify and train selected teachers as patrons of peer education clubs, develop peer education manuals for youths in primary and secondary tiers, and extend peer education to out-of-school youths. Even though the study controlled for the effect of background attributes, the results remain liable to confounding from social interaction between members of the two groups, given that beneficiaries and non-beneficiaries were sampled from the same schools. FUNDING: The Kenya Red Cross Society funded the study as part of their routine project evaluation.