Marshall Joffe, MD, MPH, PhD
|Office Location||602, Blockley Hall|
|CCEB Appointment||Senior Scholar, Biostatistics|
|Primary Faculty Appointment||Professor of Biostatistics, University of Pennsylvania Perelman SOM|
Dr. Joffe's research centers on statistical methods for estimating the causal effects of time-varying treatments and exposures in epidemiology and medicine. Dr. Joffe is the Principal Investigator of a NIH-funded R01 on Analysis of Case-Control Follow-up Studies. Specific methodological interests include confounding by variables affected by treatment, the effects of noncompliance, sensitivity of inference to assumptions about temporal ordering of variables, confounding by indication, dealing with unmeasured confounders, and observational assessment of screening efficacy. Dr. Joffe is also involved in funded collaborative work in nephrology.
Joffe, M.M., Greenland, S.: Standardized estimates from categorical regression models. Statistics in Medicine 14: 2131-2141, 1995.
Joffe, M.M., Hoover, D.R., Jacobson, L.P., Kingsley, L., Chmiel, J.S., Visscher, B.R.: The effect of treatment with zidovudine on the incidence of Kaposi's sarcoma. Clinical Infectious Diseases 25: 1125-1133, 1997.
Joffe, M.M., Hoover, D.R., Jacobson, L.P., Kingsley, L., Chmiel, J.S., Visscher, B.R., Robins, J.M.: Estimating the effect of zidovudine on Kaposi's Sarcoma from observational data using a rank preserving structural failure-time model. Statistics in Medicine 17:1073-1102, 1998.
Joffe, M.M., Colditz, G.A.: Restriction as a method for reducing bias in the estimation of direct effects. Statistics in Medicine 17:2233-2249, 1998.
Joffe, M.M., Rosenbaum, P.R.: Propensity scores. American Journal of Epidemiology, (in press).