Case-crossover and case-time-control studies: concepts, design, and analysis

Background  & Aim: Case-crossover  studies are the case-control  version of crossover studies. In these studies, cases and controls are the same subjects. The term crossover is applied for designs that all subjects pass through both treatment (exposure) and placebo (non-exposure) phases. In fact they are  crossover  of  subjects  between  periods  of  exposure  and  non-exposure.  This  design  is  most suitable for outcomes, which their induction time is short. Whatever the onset of outcome is more abrupt, it is more amenable to be studied in case-crossover  studies. Case-crossover  design can be implemented  in a number  of ways. In this article  some common  terms in the literature  of case- crossover  studies,  major  concerns  in  selection  of  controls,  different  ways  for  designing  case- crossover studies, some examples for analyzing these data according to different sampling designs and  also  design  and  analysis  of  case-time-control   designs  with  practical  examples  are  being explored.

[1]  R A Waugh,et al.  Effects of mental stress on myocardial ischemia during daily life. , 1997, JAMA.

[2]  M Maclure,et al.  Case–crossover and case–time–control designs as alternatives in pharmacoepidemiologic research , 1997, Pharmacoepidemiology and drug safety.

[3]  S Greenland,et al.  Confounding and Exposure Trends in Case‐Crossover and Case‐Time‐Control Designs , 1996, Epidemiology.

[4]  W. Navidi,et al.  Bidirectional case-crossover designs for exposures with time trends. , 1998, Biometrics.

[5]  R. Tibshirani,et al.  Association between cellular-telephone calls and motor vehicle collisions. , 1997, The New England journal of medicine.

[6]  M. Mittleman,et al.  Cautions about car telephones and collisions. , 1997, The New England journal of medicine.

[7]  S Suissa,et al.  The Case‐Time‐Control Design: Further Assumptions and Conditions , 1998, Epidemiology.

[8]  D. Trichopoulos,et al.  Transient Exposures and the Risk of Childhood Injury: A Case‐Crossover Study in Greece , 1998, Epidemiology.

[9]  J M Robins,et al.  Estimation of a common effect parameter from sparse follow-up data. , 1985, Biometrics.

[10]  S Suissa,et al.  THE CASE‐TIME-CONTROL DESIGN , 1995, Epidemiology.

[11]  A. Ahlbom,et al.  Does heavy physical exertion trigger myocardial infarction? A case-crossover analysis nested in a population-based case-referent study. , 2000, American journal of epidemiology.

[12]  R J Marshall,et al.  Analysis of case-crossover designs. , 1993, Statistics in medicine.

[13]  J. Robins,et al.  Control sampling strategies for case-crossover studies: an assessment of relative efficiency. , 1995, American journal of epidemiology.

[14]  A Comparison of Case‐Crossover and Case‐Control Designs in a Study of Risk Factors for Hemorrhagic Fever with Renal Syndrome , 1997, Epidemiology.

[15]  M. Hernán,et al.  Case-crossover and case-time-control designs in birth defects epidemiology. , 2003, American journal of epidemiology.

[16]  M. Maclure.  The case-crossover design: a method for studying transient effects on the risk of acute events. , 1991, American journal of epidemiology.

[17]  J. Muller,et al.  Triggering of acute myocardial infarction onset by episodes of anger. Determinants of Myocardial Infarction Onset Study Investigators. , 1995, Circulation.

[18]  M Maclure,et al.  Should we use a case-crossover design? , 2000, Annual review of public health.

[19]  S. Willich,et al.  Physical exertion as a trigger of acute myocardial infarction. Triggers and Mechanisms of Myocardial Infarction Study Group. , 1993, The New England journal of medicine.

[20]  N. Breslow,et al.  Statistical methods in cancer research. Vol. 1. The analysis of case-control studies. , 1981 .