A review of the effectiveness and appropriateness of peer-delivered health promotion interventions for young people

This report reviews what is known about peer-delivered health promotion for young people. It is the fourth in a new EPI-Centre series of health promotion effectiveness reviews. Using peers to deliver health promotion interventions to young people is an increasingly fashionable strategy, especially in Europe and North America. It is based on the assumption that peers may be seen as more credible sources of information than adult, professionally trained, health educators, and may be particularly helpful in reaching ‘at risk’
young people. The aim of the review described in this report was to survey the available literature in order to examine critically the claim that the peer-delivered approach is a more appropriate and effective method of promoting young people’s health than more traditional approaches. The review looks across health topics; maps the available literature on peer-delivered health promotion; contains a methodological appraisal of the quality of studies which evaluate interventions; and identifies a number of recommendations which might usefully inform future research and practice in this area. A novel feature of the review, compared to others in the series, is that we have attempted a critical appraisal of studies describing processes involved in implementing interventions, as well as those designed to assess their impact on the target population. Literature searches were undertaken for all studies describing, discussing or evaluating peer-delivered interventions aimed at the primary prevention of disease or health promotion among young people aged 11 to 24 years. The review was restricted to studies in the English language, and excluded peer counselling or mediation interventions, as well as those where the principal medium of the intervention was video, theatre or newsletters. The searches produced 5124 citations, of which 523 met the inclusion criteria. Full reports for 462 of these were obtained within the timescale for the review. Most – 63% – of the studies were found using electronic databases; 24% were located using specialised bibliographic registers, and 13% as a result of hand searches, reference lists, or personal contacts. There were 316 reports of 271 separate interventions, of which 68% were carried in the USA and 15% in the UK. A smaller proportion of the UK studies compared to those from the USA described outcome evaluations. The most common focus of the 271 reports was drug use (alcohol, smoking, other drugs – 42%), followed by sexual health (28%). Most studies (79%) were carried out in educational settings, and used information only or skill development interventions (62%) deployed by young people of the same age or no more than a year older than the target group (73%). In only a minority of cases were the interventions based on needs expressed by the target group (23%) or developed on a partnership basis (37%). We examined in more detail the studies describing evaluations of interventions focused either on outcomes or processes or both together. This more detailed analysis was restricted to those outcome evaluations which were prospective experimental studies using what authors described as equivalent intervention and control groups and collecting pre- and post-intervention data. Process evaluations had to meet the criterion of being either formative, intermediate or summative evaluations. Forty nine outcome and 15 process evaluations were included in this more detailed review. The most common focus for the outcome evaluations was drugs (including alcohol and smoking – 53%) and for the process evaluations sexual health (56%). More outcome evaluations were carried out in educational settings than in process evaluations (93% versus 62%) and more of the outcome studies evaluated intervention employing skill development (78% versus 32%). More of the process than the outcome evaluations used peers close to the target population in age (45% versus 69%). Process evaluations were more likely to be based on expressed need (31% versus 8%) and to describe interventions developed on a partnership basis (69% versus 45%). Twelve of the 49 outcome evaluation studies were assessed as methodologically sound. Of these, seven described interventions which were effective for behavioural outcomes and three for proxy outcomes; one intervention was ineffective, and the impact of one was unclear. Five of the 12 studies directly compared peers and teachers as providers of health promotion interventions. Out of these five studies, two found peers to be more effective than teachers, two found them to be no more or less effective, and one concluded that neither peers nor teachers were effective. The critical appraisal of the process evaluations assessed them against seven quality criteria commonly advocated in the qualitative research literature as guides to reliability. The number meeting the various criteria ranged from 11 (for clearly stated aims and objectives) to three (data analysed by more than one researcher). Only two of the 15 studies met all seven criteria. Particular methodological problems were identified to be a lack of a clear description of the sample and a lack of a clear description of the methods used with only 47% of the process evaluations meeting these criteria. Overall, the review found some evidence to support the effectiveness of peer-delivered health promotion for young people. There were more sound outcome evaluations which demonstrated peer-delivered health promotion to be effective than ineffective. More than half of the sound studies showed a positive effect on at least one behavioural outcome. However, as in previous systematic reviews of health promotion, methodologically sound studies were disappointingly scarce. Because there were few sound studies, it was difficult to identify any specific characteristics of an effective model of peer-delivered health promotion. Many studies gave little information about such details as the attributes of the peer educators, method of recruitment, or kind and length of training. A further disappointment was that only three of the outcome evaluations described integral process evaluations. There is clearly room for both outcome and process studies to increase the extent to which they evaluate interventions based on information about what young people themselves say they need, and which are developed with some sort of partnership between the target group and the intervention providers. While many studies placed a good deal of emphasis on the importance of various theories, such as social influence or social learning theory, in developing effective interventions, the exact contribution made to identifying effective health promotion strategies for young people using theory in general, or theories in particular, is unclear. The studies reviewed in this report are not encouraging on the issue of peer-delivered health promotion reaching young people at enhanced risk of adverse health behaviours. Additionally, there is a significant gender issue, with young men being notably more reluctant to take on the role of peer educator. The current evidence-base for peer-delivered health promotion is
therefore limited. The intuitive appeal of the idea is not matched by much hard evidence. We suggest that greater care should be taken in future to develop and test interventions using sound methodological principles. However, our review findings did suggest several specific recommendations for future research and practice within peer delivered health promotion. Recommendations have also been made for the development of health promotion more generally for young people and for systematic review methodology.

• Population(s)
  • Other
• Prevention
  • Sexual risk behaviour
  • Education/media campaigns