Science councils in South Africa

Research evidence and policy: qualitative study in selected provinces in South Africa and Cameroon

BACKGROUND

The translation of research into policy and practice is enhanced by policymakers who can recognise and articulate their information needs and researchers that understand the policymakers' environment. As researchers, we sought to understand the policymaking process and how research evidence may contribute in South Africa and Cameroon.

METHODS

We conducted qualitative in-depth interviews in South Africa and focus group discussions in Cameroon with purposively sampled subnational (provincial and regional) government health programme managers. Audio recorded interviews were transcribed, thematically coded and analysed.

RESULTS

Participants in both countries described the complex, often lengthy nature of policymaking processes, which often include back-and-forth consultations with many diverse stakeholder groups. These processes may be influenced by political structures, relationships between national and subnational levels, funding and international stakeholder agendas. Research is not a main driver of policy, but rather current contextual realities, costs, logistics and people (clinicians, NGOs, funders) influence the policy, and research plays a part. Research evidence is frequently perceived as unavailable, inaccessible, ill-timed or not applicable. The reliability of research on the internet was questioned. Evidence-informed health decision-making (EIDM) is regarded as necessary in South Africa but is less well understood in Cameroon. Insufficient time and capacity were hindrances to EIDM in both countries. Good relationships between researchers and policymakers may facilitate EIDM. Researchers should have a good understanding of the policymaking environment if they want to influence it. Greater interaction between policymakers and researchers is perceived as beneficial when formulating research and policy questions as it raises researchers' awareness of implementation challenges and enables the design of tailored and focused strategies to respond to policymakers' needs.

CONCLUSIONS

Policymaking is complicated, lengthy and mostly done at national level. Provinces/regions are tasked with implementation, with more room for adaptation in South Africa than in Cameroon. Research evidence plays a role in policy but does not drive it and is seen as mostly unavailable. Researchers need a thorough understanding of the policy process and environment, how the health system operates, as well as the priorities of policymakers. This can inform effective dialogue between researchers and policymakers, and contribute to enhancing use of research evidence in decision-making.

Randomized controlled trials of HIV/AIDS prevention and treatment in Africa: results from the Cochrane HIV/AIDS Specialized Register

Introduction

To effectively address HIV/AIDS in Africa, evidence on preventing new infections and providing effective treatment is needed. Ideally, decisions on which interventions are effective should be based on evidence from randomized controlled trials (RCTs). Our previous research described African RCTs of HIV/AIDS reported between 1987 and 2003. This study updates that analysis with RCTs published between 2004 and 2008.

Objectives

To describe RCTs of HIV/AIDS conducted in Africa and reported between 2004 and 2008.

Methods

We searched the Cochrane HIV/AIDS Specialized Register in September 2009. Two researchers independently evaluated studies for inclusion and extracted data using standardized forms. Details included location of trials, interventions, methodological quality, location of principal investigators and funders.

Results

Our search identified 834 RCTs, with 68 conducted in Africa. Forty-three assessed prevention-interventions and 25 treatment-interventions. Fifteen of the 43 prevention RCTs focused on preventing mother-to-child HIV transmission. Thirteen of the 25 treatment trials focused on opportunistic infections. Trials were conducted in 16 countries with most in South Africa (20), Zambia (12) and Zimbabwe (9). The median sample size was 628 (range 33-9645). Methods used for the generation of the allocation sequence and allocation concealment were adequate in 38 and 32 trials, respectively, and 58 reports included a CONSORT recommended flow diagram. Twenty-nine principal investigators resided in the United States of America (USA) and 18 were from African countries. Trials were co-funded by different agencies with most of the funding obtained from USA governmental and non-governmental agencies. Nineteen pharmaceutical companies provided partial funding to 15 RCTs and African agencies co-funded 17 RCTs. Ethical approval was reported in 65 trials and informed consent in 61 trials.

Conclusion

Prevention trials dominate the trial landscape in Africa. Of note, few principal investigators and funders are from Africa. These findings mirror our previous work and continue to indicate a need for strengthening trial research capacity in Africa.