Development of a valid office-based screening tool to identify unsafe older drivers

Background: The aging US population is increasing the prevalence of cognitively impaired drivers on the roads. A rapid, accurate, cost-effective and easily accessible screening method is needed to identify unsafe drivers before they cause accidents. This study is designed to determine which components of readily-available cognitive screening tools are accurate predictors of unsafe driving. Methods: Subjects completed a battery of four cognitive tests; Mini Mental State Exam(MMSE), Clock Drawing Tests 08:20 or 11:10(CDT), Trail-Making Tests A and B(TMT-A, TMT-B), followed by participation in the STISIM driving simulator. Logistic regression was used to obtain the Odds Ratio and 95% confidence interval for each test and test question based on the outcome of “safe” or “unsafe” driving. Results: Subjects who could not repeat the statement, “No ifs ands or buts,” were 8.78 times more likely to be unsafe drivers on the simulator. Subjects making intrusive marks on CDT(11:10), were 24.96 times more likely to be unsafe drivers. TMT-B showed subjects designated into the lowest tertile were 5.95 times more likely to be unsafe drivers. Additionally, subjects that committed two or more errors on TMT-B were 4.76 times more likely to be unsafe drivers. Conclusions: The simplicity of these tests enable a physician's staff to administer only a select few predictive questions to determine a patient's ability to continue driving safely. Upon failure of an office-based screening, physicians may refer patients for further evaluation by road test, the gold standard, before the decision is made to recommend or invoke driving cessation.