Research Papers

Driving in Dementia Decision Tool: Knowledge Translation in Action

Version 1
Date added June 26, 2017
Downloaded 0 times/fois
Category 2017 CARSP XXVII Toronto
Tags Research and Evaluation, Session 3C
Author/Auteur Mark Rapoport
Stream/Volet Research and Evaluation

Slidedeck Presentation Only (no paper submitted)

3C_3_Rapoport

Abstract

While dementia poses a risk to safe driving, patients in earlier stages of the illness may be safe to drive for periods of time, and physicians are often uncertain when and how to report such patients to transportation authorities. The objective of this study was to use a knowledge translation intervention to aid physicians in deciding when to report older drivers with mild dementia or mild cognitive impairment (MCI) to transportation authorities. A parallel-group randomized controlled trial was conducted to assess a Driving in Dementia Decision Tool (DD-DT). The DD-DT was based on an algorithm derived from an earlier study, a new computerized clinical decision system, an educational package, and specialized reporting forms. Specialists and family physicians with expertise in dementia or care of the elderly were randomized to either use the DDD-DT or a control version. In the experimental DD-DT, participants received an algorithm-based recommendation on whether or not to report their patient; in the control version, they received a generic reminder about reporting legislation. Participants were stratified by gender and randomized. Quantitative analysis examined reporting and recommendation of a specialized road test. Qualitative interviews examined users' experience of the tool. Sixty-nine participating physicians were randomized; 20/35 randomized to the intervention group used the DD-DT with 114 patients, and 16/34 randomized to the control group used it with 103 patients. The proportion of patients reported to the authorities per protocol did not differ statistically in the intervention and in the control groups (50% vs. 49%; Z=-1.02, p=0.31). In a multivariate analysis controlling for physician gender, practice location, specialty, years in practice, risk-taking status, and patient age, two variables predicted algorithm-based reporting: caregiver concern (OR 5.8, 95% CI 2.5-13.6, p<0.0001) and abnormal clock drawing (OR 6.1, 95% CI 3.1-11.8, p<0.0001). Physicians' perceived stress from medical uncertainty was a marginal but non-significant predictor of all reports (OR 1.04, 95% CI 0.998-1.09, p=0.06). Group status likewise did not predict a secondary outcome of referral for specialized on-road testing, but caregiver concern and abnormal clock drawing were predictors, albeit with an effect of lesser magnitude. Interviews revealed frustration with the tool's recommendation that there was no expert consensus on reporting in some cases, and a perception that the tool would be most valuable for family physicians. The tool itself did not have the anticipated benefit of increasing reporting of higher-risk patients with dementia to the transportation authorities. However, we have confirmed in a real-world setting what our earlier research on hypothetical scenarios suggested: In-office measures of cognitive functioning and expressions of concern from caregivers to physicians does indeed raise concerns of these physicians to the point of reporting to transportation authorities. The intervention did not increase physician reporting of patients with MCI/mild dementia to transportation authorities, beyond the effects of caregiver concern and clock drawing abnormalities. Future directions will focus on assessing the tool in primary care settings.

Mark Rapoport