Background: The Effective Care Research Unit (ECRU), of the University of the Witwatersrand, is conducting a research project consisting of the provision of access to evidence based maternity care in the form of the Reproductive Health Library (RHL) and the effects that this has on the provision of care to women during childbirth. This is part of a larger research project aimed at improving care in general for women during childbirth. At ten state maternity services in and around Johannesburg the RHL, together with the required computers was provided, and training in both computer-using skills and the RHL was given to all maternity staff. The effect of access to such a programme as RHL on the provision of maternity care, is being assessed using a before and after design. This involves interviewing 200 postnatal women at each site and focus group discussions with staff before and after the provision of the RHL and computers. During the time the researcher was involved in the training of staff in how to use the computer and the RHL, many interesting comments were made by the staff. The staff received repeated training sessions at each site over the ten month study period. The comments were related to the problems the staff had encountered whilst learning to use the RHL and how this had influenced how they had used the RHL. We felt that it would be useful to explore these comments further, using in depth interviews.

Methods: Interviews were conducted by the researcher with staff members at each of the 10 sites that have been provided with the RHL and that have received training on the use of the RHL. Due to the high staff turn over at state maternity services in South Africa it was difficult to use a clear sampling method. We were able to interview only those staff members that were available (a convenience sample). The interviews followed a specified list of open-ended questions as well as space for free comments. Questions were designed to evaluate accessibility, content, format and utilisation of the RHL. Each interview was tape-recorded and then transcribed.

Results: Analysis of interview transcripts used methods of 'content analysis'. Transcripts were coded according to participant and study site characteristics; the researchers deemed these variables the most likely to influence respondents' answers. Categories for analysis were largely pre-defined and addressed each of the main issues - logistics/accessibility, format/presentation, contents, and other general comments. Responses within each category were analysed using constant comparison. During analysis, several key issues emerged from the main themes. These related issues were noted and summarised on transcript sheets and written up. Analysis of the interview transcripts is in process and final results will be presented at the Colloquium. Conclusion: The findings from the in depth interviews provide a practical evaluation of the RHL programme and its usefulness to maternity staff working in the clinical environment in over loaded and under resource settings. The interviews highlighted various problems experienced by new users of the RHL, which ought to be addressed in future issues of the Library.