A Oakley1, D Fullerton, J Holland. 1. Social Science Research Unit, University of London, Institute of Education, UK.
Abstract
OBJECTIVES: The aim of this study was to identify and critically review behavioural interventions in the HIV prevention and sexual health fields. METHODS: Electronic and hand searches were conducted to retrieve relevant published and unpublished reports of outcome evaluations. A methodological review was carried out to identify those with sufficient methodological strengths to generate reliable conclusions as to effectiveness. Soundly designed studies were defined as those which met the four core criteria of employing control groups, providing pre- and post-intervention data and reporting on all targeted outcomes. RESULTS: A total of 68 separate reports of outcome evaluations were located. Only 18 were judged to be methodologically adequate. Major problems found were lack of a control group or non-equivalent/unbalanced control groups, small sample sizes, failures to report pre-intervention measures, short follow-up, and high attrition rates. Academic reviewers were more likely than the authors of papers to judge reports of evaluations as providing insufficient information to assess effectiveness. CONCLUSION: Evaluation design in this field needs to be improved. Recommendations include more use of randomized controlled trials and the raising of publication standards by journals.
OBJECTIVES: The aim of this study was to identify and critically review behavioural interventions in the HIV prevention and sexual health fields. METHODS: Electronic and hand searches were conducted to retrieve relevant published and unpublished reports of outcome evaluations. A methodological review was carried out to identify those with sufficient methodological strengths to generate reliable conclusions as to effectiveness. Soundly designed studies were defined as those which met the four core criteria of employing control groups, providing pre- and post-intervention data and reporting on all targeted outcomes. RESULTS: A total of 68 separate reports of outcome evaluations were located. Only 18 were judged to be methodologically adequate. Major problems found were lack of a control group or non-equivalent/unbalanced control groups, small sample sizes, failures to report pre-intervention measures, short follow-up, and high attrition rates. Academic reviewers were more likely than the authors of papers to judge reports of evaluations as providing insufficient information to assess effectiveness. CONCLUSION: Evaluation design in this field needs to be improved. Recommendations include more use of randomized controlled trials and the raising of publication standards by journals.
Authors: S Ma; N H T M Dukers; A van den Hoek; F Yuliang; C Zhiheng; F Jiangting; Z Lina; Z Xiuxing Journal: Sex Transm Infect Date: 2002-04 Impact factor: 3.519
Authors: Marta R Durantini; Dolores Albarracín; Amy L Mitchell; Allison N Earl; Jeffrey C Gillette Journal: Psychol Bull Date: 2006-03 Impact factor: 17.737