6. Evidence-Based Policy: Randomised Controlled Trials’ Knowledge Claims to AIDS Policy

HIV and AIDS investment represents one of the biggest single issues here, with 16.11 billion US dollars being spent in 2011 (UNAIDS 2012) within international development. The nature of HIV interventions, in particular treatment roll-out, requires a gradual increase in this funding over time. Therefore, many donors are looking for cost-effectiveness in interventions to reach the widest beneficiary groups.

External validity concerns the extent to which the results of a study or experiment can be generalized to a wider population or a wider set of circumstances.

I am providing the first four references here not to take part in the ongoing controversy these articles are presenting on VMMC. I think they are useful as they represent the depth and breath of the debate and literature on this topic.

Arguably, in an ideal resource-rich context, this dilemma should not occur. One would also argue that an implicit choice is being made on the assumption that it is impossible to improve resource infra-structure in Africa and therefore saving more lives is targeted by trying to increase the coverage of treatment independent of the other available medical resources. However, this does not deal with the increased need for medical resources due to bringing more people into ART treatment if the predicted decline in costs is achieved from not spending the resources on laboratories.

In the case of VMMC debate, the discussion by Wamai et al. 2015 is interesting for its style of argument. Here, claims to effectiveness are based on the aggregate interpretation of all RCT and observational study results, when taken together. This leads the reader to assume some form of external validity, when in fact none has been demonstrated. This in turn is used to justify general policy prescriptions.

The issue here is also about the number of people who participate in these individual trials. While the cumulative number of people participated in 12 trials may appear large, each trial had different numbers of participants. These range from 400 participants in the Ipergay trial in France and Canada, to 2499 participants across 11 study sites in 6 countries (the USA, Ecuador, South Africa, Peru, Brazil and Thailand) in iPrEx. The Bangkok Tenofovir Study included 2413 participants focusing on injecting drug users while Botswana TDF2 study included 1219 sexually active male and female participants. There is the question here whether these numbers can simply be added together to create a total number of participants through which the efficacy of PrEP is demonstrated. Moreover, how individually or collectively do they relate to the larger population targeted by WHO guidelines? While one can imagine an argument that, for example, trial procedures create a commonality of adherence in support of a claim to global efficacy, this argument becomes redundant as such uniform adherence cannot be expected outside the trial context(s) as a longer-term outcome given the diversity of participants and their socio-cultural and political contexts.

The FEM-PrEP clinical trial was designed to assess whether a daily dose of the antiretroviral Truvada is safe and effective in preventing HIV infection among women at high risk of HIV exposure. In this Phase III, randomized, placebo-controlled trial, 2120 women were assigned to receive either Truvada or a placebo containing no active drug for approximately 12 months and came in for monthly clinic visits for approximately 14 months (Fem-PrEP 2016).