Older driver estimates of driving exposure compared to in-vehicle data in the Candrive II study

Abstract

OBJECTIVE: Most studies on older adults' driving practices have relied on self-reported information. With technological advances it is now possible to objectively measure the everyday driving of older adults in their own vehicles over time. The purpose of this study was to examine the ability of older drivers to accurately estimate their kilometers driven over one year relative to objectively measured driving exposure. METHODS: A subsample (n = 159 of 928; 50.9% male) of Candrive II participants (age >/= 70 years of age) was used in these analyses based on strict criteria for data collected from questionnaires as well as an OttoView-CD Autonomous Data Logging Device installed in their vehicle, over the first year of the prospective cohort study. RESULTS: Although there was no significant difference overall between the self-reported and objectively measured distance categories, only moderate agreement was found (weighted kappa = 0.57; 95% confidence interval, 0.47-0.67). Almost half (45.3%) chose the wrong distance category, and some people misestimated their distance driven by up to 20,000 km. Those who misjudged in the low mileage group (</=5000 km) consistently underestimated, whereas the reverse was found for those in the high distance categories (>/= 20,000); that is, they always overestimated their driving distance. CONCLUSIONS: Although self-reported driving distance categories may be adequate for studies entailing broad group comparisons, caution should be used in interpreting results. Use of self-reported estimates for individual assessments should be discouraged.