Abstinence-based programs fall into two categories: abstinence-only interventions present sexual abstinence as the exclusive means ofHIV prevention; abstinence-plus interventions emphasize abstinence as the most effective prevention strategy, but also actively promote condom use and safer sex. This thesis examines the effects of abstinence-based programs in two systematic reviews of randomized controlled trials. Both reviews were conducted under the auspices of the Cochrane Collaboration. The review of abstinence-only interventions identified 13 trials, which consistently found no significant program effects on biological or behavioral outcomes compared to various controls. The review of abstinence-plus programs contains 39 trials, of which 24 found a protective biological or behavioral intervention effect compared to various conu::ols. Next, this thesis investigates data related to program implementation in primary trials: that is, the design of active and control programs, their delivery by trial staff, their uptake by trial participants, and the contextual factors that affect implementation. These data can affect heterogeneity and comparability oftrial evidence in a systematic review. Furthermore, implementation information is critical for practitioners seeking to use evidence in practice. However, current guidelines do not prepare reviewers to identify, appraise, or use implementation data in analyses. This thesis relied on consensus methods and a literature search to generate a conceptual framework of program implementation with relevance to systematic reviews. This model is the first to incorporate implementation ofcontrol-group interventions, contextual factors, vulnerability of implementation data to bias, and variations in intended treatment design across trials. The framework has been received favorably by systematic reviewers, methodological experts, and review users. Finally, the thesis assesses the relevance of the framework to the systematic review of abstinence-programs. This application suggests ways that implementation data can inform reviewers' judgments about grouping trials, assessing heterogeneity, interpreting the generalizability of evidence, and suggesting directions for research and practice.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:487096 |
| Date | January 2007 |
| Creators | Underhill, Kristen |
| Publisher | University of Oxford |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0017 seconds