PubMed1998
OBJECTIVE: To consider evidence for the effectiveness of psychosocial interventions in reducing the risk of infection with HIV of injecting drug users. METHOD: We reviewed 19 published studies of the effectiveness of individual counselling, HIV testing, group interventions, street outreach, and a 'social' intervention designed to change norms for safer behaviours. RESULTS: Eight of 15 studies that examined behavioural outcomes provided evidence of the effectiveness of an experimental intervention, compared with a control or comparison group. In four of these studies, however, serious design limitations made results difficult to interpret. In the other four studies without design limitations, the success of the experimental interventions may have been due to their greater length and intensity as well as to having been conducted with stable and well-motivated populations. Nine of the 15 studies showed evidence of marked behaviorial changes in both experimental and comparison group(s), with the changes in many cases being sustained for upwards of 12 months. CONCLUSIONS: A close examination of the evidence and competing hypotheses for the pattern of results suggests that participating in evaluation research may itself be a valuable intervention. Implications for the development of interventions include the potential efficacy of health risk assessment. Implications for evaluation of interventions include the need for developing unobtrusive measures and for assessing the impact of behavioural assessments. Despite the large behavioural changes reported in most of the studies, a substantial proportion of subjects receiving interventions reported unacceptably high levels of risk behaviours. New, more potent interventions are needed, such as those designed to change the norms of entire communities of drug users concerning safer injection and safer sex.