Setting directions for capacity building in primary health care: a survey of a research network

BACKGROUND: The South Australian Research Network 'SARNet' aims to build research capacity in primary health care, as part of a national government-funded strategy to integrate research into clinical practice. Internationally, research networks have been a fundamental part of research culture change, and a variety of network models exist. The 'SARNet' model uses a whole system, multidisciplinary approach to capacity building and supports individuals and groups. We undertook a descriptive baseline survey in order to understand the background and needs of SARNet members and to tailor network activities towards those needs. METHODS: A questionnaire survey, assessing members' professional background, research experience, and interest in research development and training, was sent to all members who joined the network in its first year. The visual 'research spider' tool was used to ascertain members' experience in ten core research skills, as well as their interest in developing these skills. Individuals were asked to classify themselves into one of four categories of researchers, based on previous research experience. These self-assessment categories ranged from non-participant to academic. RESULTS: Network membership was diverse. Of the 89 survey participants, 55% were general practitioners or allied health professionals. Overall, most survey respondents indicated little to moderate experience in 7 out of the 10 skills depicted in the 'research spider'. In comparison, respondents were generally highly interested in developing their research skills in all areas. Respondents' research skills correlated significantly with their self-assessed category of research participation (Spearman rank correlation, r = 0.82, p < 0.0005). Correlations between research category and publication record (Gamma association, γ = 0.53, p < 0.0005) or funding record (Gamma association, γ = 0.62, p < 0.0005) supported the internal validity of the survey instrument. CONCLUSION: Literature describing evaluation of the impact of networks is scarce. Our survey questionnaire could provide a useful instrument for evaluation of both networks and capacity building initiatives. The survey including the 'research spider' tool provided valuable information about members' needs and interest in strategies to develop their research skills. Initial needs analyses as well as on-going evaluation of network activities are important to include into the business plans of research networks, in order to ensure the network's effectiveness and support of its membership.