What works in positive prevention?

Two meta-analyses of HIV-prevention interventions specifically directed at people with HIV were published in 2006, finally amending the dearth of review-level research in the area – although they still concentrate entirely on US studies.

In the first,¹ Nicole Crepaz of the Centers for Disease Control and Prevention (CDC) sifted through 310 studies of prevention programmes for people with HIV and filtered out all but the twelve whose standard of evidence stood up to the most rigorous scientific scrutiny. In total, 4052 people with HIV participated in these twelve programmes.

Crepaz concluded that HIV-positive people responded at least as well as HIV-negative people, and possibly better, to prevention interventions. She found a significant average 43% reduction of sexual-risk incidents in participants in the twelve programmes (OR 0.57), and a 79% reduction in STI incidence (OR 0.21, though only a minority of programmes measured this, and the 95% confidence interval was very wide).

By comparing one study with another, Crepaz found that the most effective interventions were ones that:

• specifically taught people how to negotiate condom use and safer sex as their main focus
• included help for other aspects of living with HIV, such as disclosing status, medication adherence and maintaining self-esteem
• were intensive, that is, which involved at least ten sessions delivered over at least three months
• were delivered in a clinical setting or at a voluntary organisation that already provided services to people with HIV, rather than in outreach or community settings
• were at least partly delivered by professional counsellors
• were at least partly delivered on a one-to-one basis.

Crepaz found that the least effective programmes were those which simply delivered information on transmission risk and condom effectiveness.

In the second review,² Blair T Johnson of the University of Connecticut analysed 15 studies, including some that Crepaz had also analysed. Some of the studies actually covered more than one intervention, so Johnson was able to study a total of 19 different interventions.

Programmes used varied techniques, such as group therapy, support, role plays, videos, telephone support and one-to-one counselling, and lasted from 18 days to 45 weeks.

Johnson found that, on average, they produced an overall increase in condom use of 16% relative to baseline – not as good as Albarracin’s ‘active’ programmes, but as good as, or better, than her ‘passive’ ones. Johnson, like Albarracin, used this measure of effectiveness as it was the one used most frequently as the measure of success.

However, Johnson found no change whatsoever in the number of partners people had after intervention. Only 7 out of the 19 programmes measured if the number of partners changed, however, and Johnson suggests that this may have been confounded by people doing things not measured by researchers instead of reducing partners, such as serosorting. The most effective, a 2000 programme directed at HIV-positive teenagers in Los Angeles, increased their condom use by 82% compared with a control group. However, some were unsuccessful, and the only non-US programme, directed at HIV-positive people in Tanzania, actually reduced condom use by 25%. Johnson found that programmes that worked had three characteristics.

Firstly, they tended to be directed at younger people. Programmes where the average age of participants was 20 worked five times better than ones directed at 40 year olds. Johnson hypothesised that older people tend more often to be in long-term relationships where sexual habits are harder to shift, and says that better interventions for long-term couples where one partner has HIV need to be devised.

Secondly, the ones that worked were either motivational or taught people behavioural skills, and programmes which did both worked even better. ‘Motivational’ was the word Johnson used for programmes that provided things that improved participants’ overall quality of life such as increased social support or self-confidence.

Programmes that provided information on HIV risk alone had no effect; ones that added in either motivation or behavioural skills increased condom use by 12%; and ones providing all three things increased it by 33%.

Thirdly, and disappointingly, programmes directed at gay men did not work, by and large, and, conversely, ones that excluded gay men were effective, increasing condom use by 42%.

However, Johnson does not see this as evidence that gay men are uniquely deaf to safer-sex advice and support. He points out that not one single programme directed at gay men provided both ingredients proven to be necessary – they either provided greater social support or taught behavioural skills, but not both.

Johnson criticises the lack of scientific research into ways of helping HIV-positive people maintain safer sex and reduce HIV transmission. He comments:

“Perhaps the most surprising finding of this work is that more than two decades into the epidemic, there have been so few randomly-controlled trials of interventions that focus on people living with HIV, though there have been literally hundreds of studies conducted with uninfected populations. There is an urgent need for research in this area.”