The clinical effectiveness of individual behaviour change interventions to reduce risky sexual behaviour after a negative human immunodeficiency virus test in men who have sex with men: systematic and realist reviews and intervention development.

Background: Men who have sex with men (MSM) experience significant inequalities in health and well-being. They are the group in the UK at the highest risk of acquiring a human immunodeficiency virus (HIV) infection. Guidance relating to both HIV infection prevention, in general, and individual-level behaviour change interventions, in particular, is very limited.

Objectives: To conduct an evidence synthesis of the clinical effectiveness of behaviour change interventions to reduce risky sexual behaviour among MSM after a negative HIV infection test. To identify effective components within interventions in reducing HIV risk-related behaviours and develop a candidate intervention. To host expert events addressing the implementation and optimisation of a candidate intervention.

Data Sources: All major electronic databases (British Education Index, BioMed Central, Cumulative Index to Nursing and Allied Health Literature, EMBASE, Educational Resource Index and Abstracts, Health and Medical Complete, MEDLINE, PsycARTICLES, PsycINFO, PubMed and Social Science Citation Index) were searched between January 2000 and December 2014.

Review Methods: A systematic review of the clinical effectiveness of individual behaviour change interventions was conducted. Interventions were examined using the behaviour change technique (BCT) taxonomy, theory coding assessment, mode of delivery and proximity to HIV infection testing. Data were summarised in narrative review and, when appropriate, meta-analysis was carried out. Supplemental analyses for the development of the candidate intervention focused on post hoc realist review method, the assessment of the sequential delivery and content of intervention components, and the social and historical context of primary studies. Expert panels reviewed the candidate intervention for issues of implementation and optimisation.

Results: Overall, trials included in this review ( = 10) demonstrated that individual-level behaviour change interventions are effective in reducing key HIV infection risk-related behaviours. However, there was considerable clinical and methodological heterogeneity among the trials. Exploratory meta-analysis showed a statistically significant reduction in behaviours associated with high risk of HIV transmission (risk ratio 0.75, 95% confidence interval 0.62 to 0.91). Additional stratified analyses suggested that effectiveness may be enhanced through face-to-face contact immediately after testing, and that theory-based content and BCTs drawn from 'goals and planning' and 'identity' groups are important. All evidence collated in the review was synthesised to develop a candidate intervention. Experts highlighted overall acceptability of the intervention and outlined key ways that the candidate intervention could be optimised to enhance UK implementation.

Limitations: There was a limited number of primary studies. All were from outside the UK and were subject to considerable clinical, methodological and statistical heterogeneity. The findings of the meta-analysis must therefore be treated with caution. The lack of detailed intervention manuals limited the assessment of intervention content, delivery and fidelity.

Conclusions: Evidence regarding the effectiveness of behaviour change interventions suggests that they are effective in changing behaviour associated with HIV transmission. Exploratory stratified meta-analyses suggested that interventions should be delivered face to face and immediately after testing. There are uncertainties around the generalisability of these findings to the UK setting. However, UK experts found the intervention acceptable and provided ways of optimising the candidate intervention.

Future Work: There is a need for well-designed, UK-based trials of individual behaviour change interventions that clearly articulate intervention content and demonstrate intervention fidelity.

Study Registration: The study is registered as PROSPERO CRD42014009500.

Funding: The National Institute for Health Research Health Technology Assessment programme.