Author Affiliations: University of Washington (Drs Kerns, Pullmann, Walker, Lyon, and Bruns) and Public Health–Seattle and King County (Mr Cosgrove), Seattle.
Treating time-varying variables as fixed can result in a systematic bias referred to as “time-dependent bias,”1 “immortal time bias,”2 or “survival bias.”3 This is a common analytic error; in a review of publications from 9 top medical journals from 1998 to 2002, of 127 studies that used survival analysis with time-dependent variables, 41% (52 studies) were susceptible to time-dependent bias because they did not appropriately operationalize at least 1 time-varying variable in the model.1
Kerns SEU, Pullmann MD, Walker SC, Lyon AR, Cosgrove TJ, Bruns EJ. School-Based Health Center Use and High School Dropout Rates—Reply. Arch Pediatr Adolesc Med. 2012;166(7):675–677. doi:10.1001/archpediatrics.2012.553
Artificial Intelligence Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.