In this application, we request support from the Agency for Healthcare Research and Quality for a national conference entitled, """"""""Techniques for Handling Bias in Health Services Outcomes Research"""""""". The conference is being planned and hosted by health services researchers in the Section of Health Services Research of Baylor College of Medicine and The Houston Center for Quality of Care and Utilization Studies (an HSR and D Center of Excellence in Health Services Research funded by the Department of Veterans Affairs). The conference is being supported in part by Baylor College of Medicine; Pfizer, Inc.; and the Veterans Affairs Health Services Research and Development Service. The conference is scheduled for November 7-9, 2000. The conference will focus on three main sources of bias - selection bias, information bias, and confounding. The conference will include four sessions. The first session will offer background talks about these types of bias. The second session will focus on methods for handling bias in health services outcomes research. Practical applications of the methods will constitute the third session. In the final session, invited health services researchers will present current research in which innovative techniques were used to handle bias. While health services researchers' concern for bias is not new, their skill in the application of the more advanced methods for adjusting for bias may need developing. Training in the new and more advanced techniques has not been readily available for health services researchers. In addition, the available training has often been designed for biostatisticians and psychometricians and not for the practicing researcher. This conference will bring together leading specialists in sampling techniques, missing data analyses, measurement theory, and modeling techniques, so health researchers from all disciplines can learn about new and advanced methods for handling bias in outcomes research.
