To evaluate HIV sexual risk reduction programmes conducted in the United States that have targeted adult heterosexual men.

Searches were conducted of the following databases for articles published in the English language: AIDSLINE (1980 to November 1998); PsycLIT (January 1981 to November 1998); and MEDLINE (1966 to December 1998). A manual search was conducted of Social and Behavioural Sciences Current Contents (1995 to November 1998) using the category "HIV and AIDS". Articles published as abstracts were excluded.

Study designs of evaluations included in the review

Inclusion criteria were not defined in terms of study design. Randomised controlled studies. Non randomised controlled studies, observational studies and pretest-post test studies were included.

Specific interventions included in the review

Prevention or intervention programmes that focused on reducing HIV heterosexual risk behaviour were eligible. The focus was on interventions with a psychological, behavioural, or educational component. Actual interventions studied included: information only (facts about HIV transmission and prevention strategies); condom skills/distribution (provided information plus direct practice of condom use or actual distribution of condoms); relational skills-focused (a variety of direct skills-based techniques including training and role play of safer sex negotiation, assertiveness and communication skills, and problem solving about high risk situations); HIV counselling and testing; enhanced HIV counselling and testing (combined standard counselling and testing with condom or relational skills training); individual risk counselling (tailored purely informational interventions to incorporate personal issues or concerns regarding HIV prevention); and community outreach programmes (e.g. needle exchange and condom distribution programmes, one-to-one street outreach) targeted populations who may not be motivated to participate in more formal individualised programmes. Interventions were conducted on a one-to-one basis and with groups.

Participants included in the review

Heterosexual adult North American men were the focus of the review. Studies that included both sexes and provided analysis broken down by gender were also eligible. Studies that included both heterosexually and homosexually active men were excluded when homosexual men comprised the majority and results were not analysed separately by gender of partner. The most frequently studied populations were intravenous drug users (IVD) and sexually transmitted disease (STD) clinics. Other populations included college students, HIV infected individuals, serodiscordant couples, those seeking HIV testing, and inner-city men. Participants ranged in age from 18 to 65 years with the majority between the ages of 20 and 45 years. Ethnicity of the most studied groups was, in order, African-American/Black, Caucasian and Latino.

Outcomes assessed in the review

Sexual behaviour outcomes or public health proxy of behaviour change had to be assessed. Studies evaluating only changes in knowledge, attitudes, or behavioural intention were excluded. A variety of actual outcomes were assessed including: condom usage or acquisition; partner reduction; increasing self-reported "caution"; and seeking HIV testing.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.

Some aspects of validity were mentioned briefly in the discussion but no formal validity assessment was undertaken.

The authors do not state how the data were extracted for the review, or how many of the reviewers performed the data extraction.

How were the studies combined?

Studies were grouped by intervention type and combined in a narrative review.

How were differences between studies investigated?

Differences between the studies were not discussed.

A total of twenty studies were included. It was not possible to calculate the number of patients.

Where more than two RCTs were reported per intervention type, only results from the RCTs are reported below. Details of results from all included studies were presented in the review.

1. Information only (2 studies, including one RCT and one non-randomised controlled study). Both studies found significant effects on improving HIV risk awareness, reflected in tests for HIV testing or reported increased cautiousness with sexual partners.

2. Condom skills/distribution (2 studies, including one RCT and one pretest-posttest study). Results were inconsistent with one study reporting significant benefit after the provision of access to condoms.

3. Relational skills-focused (9 studies with mixed gender populations, including 4 RCT and 5 non-randomised controlled studies). Six of the nine studies reported significant results.

4. Standard HIV counselling and testing (2 studies, including one retrospective case series with 1972 participants and one non-randomised, non-controlled pretest-posttest study). One study found a significant impact on behaviour among HIV+ men, reflected in STD infection rates.

5. Enhanced HIV counselling and testing (2 studies, including one observational study and one non-randomised controlled study). Results were inconsistent.

The observational study (1157 HIV + STD patients) found that men provided with service referrals posttest had significantly lower rates of gonorrhea than those who received no referral. Relative risk = 0.49 (95% CI: 0.26, 0.94).

The controlled study (1300 male and female intravenous drug users) found no differences in sexual risk behaviour between groups receiving HIV counselling and testing or enhanced counselling and testing.

6. Individual risk counselling (one observational study).

The study found that serodiscordant couples given couple and individual safer sex counselling reported increased sexual abstinence (P < 0.001) and consistent condom use (P < 0.001) compared to baseline. There was a high drop-out rate (44%).

7. Community outreach programmes (2 pretest-posttest studies targeting men and women intravenous drug users or sex partners of users using one-to-one field outreach with AIDS education). Both studies reported significant benefit from the intervention.

One study (1358 men) found significant reductions in a sexual risk behaviour index at 6 months follow-up (P < 0.001). Drop-out rate 25%.

One study (79 men) found a significant decrease in the number of men who ever engaged in vaginal, anal or oral sex without a condom (P < 0.01). High drop-out rate (56%).

Methodological limitations of the studies included: lack of randomised design; failure to provide statistical findings for variables assessed; excessive attrition; and unclear operationalisation of outcomes.

All intervention types (information-only; condom skills/distribution; behavioural skills-focused; HIV counselling and testing; individual risk counselling; street outreach) showed some efficacy, and there were no discernable differences by targeted population. Both group and individual formats were effective.

The aims were stated and inclusion criteria defined in terms of intervention, outcome, and participants. Inclusion criteria were not defined in terms of study design. The search included several relevant databases and journals though no attempt was made to locate unpublished material thus raising the possibility of publication bias. Methods used to select studies were not described. Study validity was not formally assessed though some aspects of validity were briefly mentioned in the discussion. Methods used to extract data were not described and only incomplete information on the included studies was presented in the text. Some studies reported very high drop-out rates which may have invalidated the results reported. Sample size was inconsistently reported and results were not consistently reported as values, point estimates and 95% confidence intervals of differences between groups, or as levels of significance. In RCTs, where results for only male participants were reported, it was not stated whether randomisation had been stratified according to sex. The number of outcomes per individual study was not reported and hence results assessed in terms of conventional levels of significance may be misleading. Given the heterogeneity among studies in terms of populations, interventions and outcomes, a narrative review was appropriate. Attention was not drawn to the better sources of evidence and where differences in outcomes were reported, potential causes of these differences were not discussed. In view of the above limitations, the conclusion should be interpreted with caution. Had the authors considered study validity or used some grading of the evidence, then perhaps the conclusions would have been clearer.

Practice: The authors state that HIV sexual risk reduction programmes are warranted and that sexual risk reduction interventions targeting heterosexual men can lead to decreases in HIV risk behaviour.

Research: The authors state that interventions are needed that are grounded in a good understanding of men's sexuality and focus on how to make safer sex work in the larger content of men's lives. Additional research is needed to increase the specific knowledge about the most efficacious types of HIV prevention formats, essential content and realistic outcomes.

NIMH, grant number #5P50 MH-43520; HIV Center for Clinical Behavioral Studies, and NIMH Training, grant number #T32-MH19139; Behavioral Science Research in HIV Infection.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.