Systematic bias in outcome studies of persons with traumatic brain injury

Objective: (1) Examine systematic biases created by subjects lost at 1-year follow-up in samples of persons with traumatic brain injury; (2) identify potential threats to generalization of outcomes data. Design: A consecutive sample of admissions to acute rehabilitation studied 1 year following discharge. Setting: An inpatient brain injury rehabilitation unit in a large, academic medical center. Subjects: Eighty-eight patients with primary diagnosis of traumatic brain injury. Main Outcome Measures: Subjects were considered lost to follow-up when phone calls, mail, clinic visits, and assistance from family failed to allow contact 1 year after discharge from acute rehabilitation. Potential effects of the biased follow-up sample were examined for seven suboptimal outcomes. Results: A total of 38.6% of subjects were lost to follow-up. Subjects intoxicated at time of injury and those with history of substance abuse were more likely to be lost. Among subjects followed, the likelihood of working or being in school I year after discharge was significantly less for those intoxicated at time of injury and those with a history of substance abuse. Conclusions: Systematic bias in longitudinal studies may result from subjects with substance use problems being lost to follow-up. Population estimates for return to work or school will be overestimated if those lost who have substance use problems resemble those followed. (C) 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation.