CASE-CROSSOVER AND CASE-TIME-CONTROL DESIGNS AS ALTERNATIVES IN PHARMACOEPIDEMIOLOGIC RESEARCH

Standard cohort and case-control designs are suited to the study of cu mulative effects of chronic exposures, but they are prone to confoundi ng by indication. Case-crossover and case-time-control studies are esp ecially useful for studying intermittent exposures with transient effe cts, and are less susceptible to confounding by indication. Each desig n has its strengths and weaknesses. Despite the increasing availabilit y of automated databases, cohort studies are usually time consuming an d expensive, and therefore not preferred for time-critical decisions. In case-control studies, the selection of appropriate controls can be difficult and time consuming, and sometimes impractical when the expos ure is rare. Case-crossover studies use the exposure history of each c ase as his or her own control to examine the effect of transient expos ures on acute events. It further allows to study the time relationship of immediate effects to the exposure. This design eliminates between- person confounding by constant characteristics, including chronic indi cations. Because exposure data for the case and control periods are pr ovided by the same person, the problems of differential recall may be reduced in many but not all case-crossover studies. Bias can result fr om temporal changes in prescribing or within-person confounding, inclu ding transient indication or changes in disease severity. The case-tim e-control design is an elaboration of the case-crossover design, which uses exposure history data from a traditional control group to estima te and adjust for the bias from temporal changes in prescribing. This paper will present a structured decision table of when to use which de sign in pharmacoepidemiologic research. In conclusion, case-crossover and case-time-control studies are the designs of choice when separatin g acute effects from chronic effects of transient exposures and if con founding by indication is an outstanding problem. (C) 1997 John Wiley & Sons, Ltd.