This chapter illustrates the problems and the methodologic issues involved in collaborative reanalyses. When reanalyses involve the pooling of data from randomized clinical trials, cohort studies, or case-control studies, the term meta-analysis has often been used in recent years. An example of collaborative reanalysis of descriptive epidemiologic data is the comparison of prevalence data of Alzheimer's disease (AD) by the European Community Concerted Action Epidemiology and Prevention of Dementia (EURODEM) group. The objective of the study was to investigate geographic differences, age and sex patterns, and time trends in the prevalence of AD. The example of collaborative reanalysis of analytic epidemiologic data is the comparison and pooling of case-control studies of AD by the EURODEM group. The objective of the study was to establish the consistency of findings across studies, to investigate rare exposures, and to conduct subgroup analyses. The most important objective was to establish consistency across studies.