In an observational study, a researcher attempts to understand effects of an intervention upon people, without controlling which of them receive the intervention. Common in medical and social research, observational studies generally compare groups exposed to two or more contrasting conditions, restricting the focus of their comparisons not with experimental control but by statistical adjustment, and by artful choice of observations. In general, case-control studies, cohort studies and cross-sectional comparisons are all observational studies. This workshop surveys some effective strategies for planning and implementation of observational studies, highlighting established principles of design and analysis, current doctrines on the use of human subjects in research, and recent analytic advances from Statistics.
Designs and analyses of observational studies vary greatly, but some common threads join the most successful of them. After considering these, the workshop discusses relevant federal and institutional requirements for human subjects research. In particular, aspects relevant to securing IRB approval will be addressed. The workshop includes an overview of matching and adjustment techniques in an observational study, but devotes special attention to a statistical tool called the propensity score.

Instructors:

Ben Hansen is Assistant Professor in the Statistics Department and a Research Associate at the Institute for Social Research, University of Michigan. He has written and consulted on the use of propensity scores, optimal matching and sensitivity analysis in observational studies.

Chuck Kowalski is a Faculty Associate at CSCAR and Professor in the School of Dentistry at the University of Michigan. He joined the U-M in 1968, and has considerable statistical consulting experience in biomedical contexts, including work with pharmaceutical companies, the National Football League, the Nijmegen Growth Study, the Lancaster Cleft Palate Clinic, and the Frei University in Amsterdam. He served as Chair of the Health Sciences Institutional Review Board from 1997-2003. He is currently Co-chair of that committee, and has interests in research ethics, especially with respect to study design and risk/benefit analysis. He continues his work with the U-M PEERRS web-based system for education in the responsible conduct of research.

Ed Rothman is the Director of CSCAR, and has been a professor in the department of Statistics at the University of Michigan since 1969. He has extensive experience as a statistical consultant in studies both on and off campus.

Audience:

Social, biomedical and all interested researchers.

Date:

Tuesday, May 13, 2008

Time:

8:30 a.m. - 5:00 p.m.

Location:

Rackham Bldg, 3^rd floor, East Seminar Room

Fee:

Registration:

Call CSCAR at 734-764-7828. Enrollment is limited. Please make check payable to CSCAR-University of Michigan, or give the University of Michigan Project/Grant or short code to be billed. Send check to CSCAR, 3550 Rackham Bldg., University of Michigan, 915 E. Washington St., Ann Arbor MI 48109-1070.