Abstract

Researchers are becoming increasingly interested in the evaluation of STI/HIV interventions that take place in “real world” programmatic or health systems settings. Likewise, programme and health service managers are facing ever growing demands to demonstrate results and achieve “value for money”. Monitoring and evaluation (M&E) systems are fundamental to both efforts. However, research and programme efforts are often not closely linked, with data being collected and analysed independently.

This presentation will examine M&E from a researchers' perspective. It will outline how a programme science perspective can contribute to the development of robust and useful M&E systems and processes. We will present two case studies. The first, the Integra project, evaluates the integration of HIV and SRH services in Kenya and Swaziland. It illustrates how research complemented routine M&E systems in a “real world” setting. This case demonstrates the benefits to programme managers of combining M&E data with economic research to enhance service performance. At the same time Integra researchers were able to use M&E data to examine the impact of “real world” issues like drug stock-outs on their research results. The second case study, the CHARME project, is an evaluation of the Avahan HIV prevention programme in India. This case illustrates the value of linking monitoring systems with research efforts to establish plausible evidence of the relationship between cost, programme activities and intensity and impact of an HIV prevention programme; where no control areas or “stepped wedge” evaluation design was feasible, due to rapid scale-up.

Finally, the presentation will identify a number of over-arching lessons learnt from these cases, and other experiences. It will highlight the importance of investing in a wide range of monitoring and evaluation activities, that go beyond Management Information Systems and qualitative peer evaluation. It will explore ways of enhancing and linking research and programme efforts, so that M&E can relate STI/HIV inputs, outputs and outcomes in a robust and logical way; but, at the same time, remains feasible and “value for money” in large scale “real world” settings.