Motivational interviewing for hiv-related behaviors among men who have sex with men

As of December 2010, there were 4627 HIV-positive persons in Norway, of which 67% were men and 33% were women. Immigrants infected prior to arrival in Norway and men who have sex with men comprise the largest proportion of newly diagnosed individuals with HIV infection. The situation is particularly worrisome for men who have sex with men, who show persistent high infection rates above 30% of all new HIV cases. Unprotected anal intercourse is the biggest risk factor for HIV infection among this group. Sexually transmitted infections (STIs), such as herpes and gonorrhea, increase the risk of becoming infected with HIV. At the same time, the use of alcohol and other substances increase the risk of STI/HIV transmission among men who have sex with men through the increased likelihood of unsafe sexual behaviors. The related problem behaviors unsafe sex and substance use, including alcohol, pose challenges in providing effective HIV reducing measures for men who have sex with men.

Motivational interviewing is a type of counselling used to motivate people to change an unwanted habit or behavior. The treatment approach, which can be individual or group-based, was originally developed for alcohol problems, but has gradually been applied to other health-related behaviors, such as substance use, unhealthy eating and physical inactivity. A core component of motivational interviewing is building clients' intrinsic motivation for behavior change that can promote better health.

Motivational interviewing is common both nationally and internationally, but there is limited evidence of the effectiveness of motivational interviewing on problem behaviors among MSM. The objective of this systematic review was to assess the effectiveness of motivational interviewing on HIV-related behaviors among men who have sex with men.

We conducted a systematic review according to the Norwegian Knowledge Centre's methods manual, which included a systematic search for literature in CENTRAL, EMBASE, ISI Web of Knowledge, MEDLINE, POPLINE, PsycINFO and Sociological Abstracts in October 2010. The search strategy was formed around the population of interest and type of intervention we wanted to investigate. Inclusion criteria were:

• Study Design: Randomized controlled trials (RCTs).
• Population: Men who have sex with men.
• Intervention: Motivational interviewing and variations of this intervention such as Brief Motivational Interviewing and Motivational Enhancement Therapy (MET).
• Comparison: No intervention, waiting list or other active intervention.
• Outcomes: Unprotected sex, STI/HIV infection, use of alcohol and other substances, changed motivation in relation to unsafe sex or use of alcohol and other substances.
• Language: All languages. 

Two reviewers performed independent assessments for inclusion of the identified studies as well as risk of bias assessment. We used an inclusion form and the Risk of Bias tool. We summarized the results in text and tables and used GRADE to grade the quality of the evidence.

The systematic search identified 275 relevant publications. We included nine studies, with a total of 5863 participants. All were randomized controlled trials (there was one cluster randomized trial). One study was from the Netherlands, while the remaining eight were from the United States. In the majority of studies, the intervention was motivational interviewing, and comparisons were either another type of active intervention, or no / minimal intervention. Sexual behaviors was the most frequently reported outcome measure. Across the included studies there was moderate or low risk of bias.

We extracted data for 32 primary outcomes (in a total of 77 effect estimates). Only one study reported a biological outcome (HIV infection): Overall, the risk of HIV infection was 15.7% lower in the intervention group than in the control group at 18 months follow-up. While the difference was not statistically significant at any follow-up times, it was clinically important. We found that the documentation for HIV infection was of moderate quality.

There were few statistically significant differences between the intervention group and the control group at study level. Only nine of the 77 effect estimates showed statistically significant differences between the groups (five were for alcohol use). We performed meta-analyses for three outcomes. The difference between the groups was not statistically significant regarding use of condoms at 2-4 months follow-up (2 studies, SMD= -0.06, 95% CI= -0.32, 0.20). We judged that the quality of the evidence was low (GRADE). The difference between the intervention group and the control group was not statistically significant for unprotected sex with non-primary partners at about 3 months follow-up (2 studies, RR= 1.04, 95% CI= 0.73, 1.47). We judged that the quality of the evidence was low (GRADE). The difference between the groups was not statistically significant for unprotected sex with a primary partner at about 3 months follow-up (2 studies, RR= 1.34, 95 % CI= 0.61, 2.95). We did not grade this outcome measure. Test for heterogeneity among the studies was not significant.

The findings in this systematic review indicate that the effectiveness of motivational interviewing on problem behaviors related to unsafe sex and alcohol use is largely similar to other active or minimal treatments for men who have sex with men. We found that there were few differences between the groups at study level. Most of these pertained to use of alcohol. None of the three meta-analyses we performed, which all regarded unsafe sex, showed significant differences between the groups.

The lack of a statistically significant differences between motivational interviewing and other active or minimal treatments for HIV-related behaviors among men who have sex with men supports findings from an earlier systematic review. Hettema and colleagues (1) synthesized the effectiveness of motivational interviewing across a range of problem behaviours and concluded that the effectiveness was greatest for alcohol and other drug-related behaviours. Our included studies suggested that motivational interviewing may be less effective with regards to sexual risk behaviours than alcohol use among men who have sex with men. Nonetheless, it is important to emphasize that the included studies showed some positive results: In one of the studies the number of HIV infections was nearly 16% lower in the first 18 months of follow-up among men who have sex with men who received motivational interviewing compared to those receiving minimal treatment.

We found moderate and low quality evidence to suggest that motivational interviewing is not likely to lead to changes regarding unsafe sexual behaviors compared to other active or minimal interventions. It seems the effectiveness of motivational interviewing on problem behaviors related to both unsafe sex and alcohol use is similar to other active or minimal intervention among men who have sex with men.