Objectives: We sought to determine whether the partial administration of the Cognitive Behavioral Driver's Inventory (CBDI) has a significant effect on its concurrent validity.
Method: Data were extracted from charts of clients with cerebrovascular accident or traumatic brain injury from three centers. The CBDI was administered either completely or partially (right and left perimetry or Wechsler Adult Intelligence Scale-Revised (WAIS-R; Wechsler, 1982; Picture Completion and Digit Symbol tests were not completed). Concurrent validity indicators were calculated for the CBDI and three different scenarios of partial administration of the CBDI.
Results: Only 52% of the road test failures were predicted correctly by the completely administered CBDI. Nonadministration of the WAIS-R rarely modified the CBDI results. Omission of perimetry scores tended to increase the sensitivity and decrease the specificity (not significantly).
Conclusion: The CBDI should be used as a complement, not a substitution, for a road test. Partially administrating the CBDI, specifically excluding perimetry measures, can affect its concurrent validity.