Eramudugolla Ranmalee, Price Jasmine, Chopra Sidhant, Li Xiaolan, Anstey Kaarin J
Centre for Research on Ageing, Health and Wellbeing, Australian National University, Canberra, Australian Capital Territory, Australia.
J Am Geriatr Soc. 2016 Dec;64(12):e253-e258. doi: 10.1111/jgs.14548. Epub 2016 Oct 22.
To design a low-cost simulator-based driving assessment for older adults and to compare its validity with that of an on-road driving assessment and other measures of older driver risk.
Cross-sectional observational study.
Canberra, Australia.
Older adult drivers (N = 47; aged 65-88, mean age 75.2).
Error rate on a simulated drive with environment and scoring procedure matched to those of an on-road test. Other measures included participant age, simulator sickness severity, neuropsychological measures, and driver screening measures. Outcome variables included occupational therapist (OT)-rated on-road errors, on-road safety rating, and safety category.
Participants' error rate on the simulated drive was significantly correlated with their OT-rated driving safety (correlation coefficient (r) = -0.398, P = .006), even after adjustment for age and simulator sickness (P = .009). The simulator error rate was a significant predictor of categorization as unsafe on the road (P = .02, sensitivity 69.2%, specificity 100%), with 13 (27%) drivers assessed as unsafe. Simulator error was also associated with other older driver safety screening measures such as useful field of view (r = 0.341, P = .02), DriveSafe (r = -0.455, P < .01), and visual motion sensitivity (r = 0.368, P = .01) but was not associated with memory (delayed word recall) or global cognition (Mini-Mental State Examination). Drivers made twice as many errors on the simulated assessment as during the on-road assessment (P < .001), with significant differences in the rate and type of errors between the two mediums.
A low-cost simulator-based assessment is valid as a screening instrument for identifying at-risk older drivers but not as an alternative to on-road evaluation when accurate data on competence or pattern of impairment is required for licensing decisions and training programs.
为老年人设计一种基于低成本模拟器的驾驶评估，并将其有效性与道路驾驶评估及其他老年驾驶员风险测量方法进行比较。
横断面观察性研究。
澳大利亚堪培拉。
老年驾驶员（N = 47；年龄65 - 88岁，平均年龄75.2岁）。
模拟驾驶中的错误率，模拟环境和评分程序与道路测试相匹配。其他测量指标包括参与者年龄、模拟器不适严重程度、神经心理学测量指标和驾驶员筛查指标。结局变量包括职业治疗师（OT）评定的道路错误、道路安全评级和安全类别。
即使在对年龄和模拟器不适进行校正后（P = 0.009），参与者在模拟驾驶中的错误率与OT评定的驾驶安全性仍显著相关（相关系数（r）= -0.398，P = 0.006）。模拟器错误率是道路上不安全分类的显著预测指标（P = 0.02，敏感性69.2%，特异性100%），有13名（27%）驾驶员被评定为不安全。模拟器错误也与其他老年驾驶员安全筛查指标相关，如有用视野（r = 0.341，P = 0.02）、DriveSafe（r = -0.455，P < 0.01）和视觉运动敏感性（r = 0.368，P = 0.01），但与记忆（延迟单词回忆）或整体认知（简易精神状态检查表）无关。驾驶员在模拟评估中的错误数量是道路评估中的两倍（P < 0.001），两种测试方式在错误率和错误类型上存在显著差异。
基于低成本模拟器进行的评估作为一种筛查工具，对于识别有风险的老年驾驶员是有效的，但当许可决策和培训项目需要关于能力或损伤模式的准确数据时，它不能替代道路评估。
