Performance on the Digital Clock Drawing Test Amongst Takers of the Driving Skills Test

As Japan's society ages, the number of traffic accidents involving older drivers is increasing. Japan's National Police Agency (NPA; Keisatsu-chō) conducts a training programme when drivers aged ≥ 70 years renew their driver's licence (kōreisha kōshū), and on 13 May 2022, a driving skills test (DST; untenginō kensa) using actual vehicles for drivers aged ≥ 75 years with records of 11 specific types of traffic violations was launched. The test-takers actually drive on test courses, such as those at driving schools, and need to score 70 points (or 80 points for Class 2 licences for drivers of taxis, buses and others) to pass the test and renew their licence. On-road driving tests are the gold standard for assessing the driving competency of older drivers, but desk-based screening using neuropsychological tests has also been considered to reduce the associated economic costs and ensure safe implementation of the driving test. Previous research has shown that using the clock drawing test (CDT) alone is not sufficiently effective, but it is recommended as part of a battery of tests. In Japan, since 2022, the CDT has been excluded from the cognitive test battery for older drivers aged ≥ 75 years of the kōreisha kōshū. Internationally, the CDT has been shown to be useful as a brief test of cognition that can be performed in a short time to screen for driving competency amongst older adults with cognitive decline to the dementia level [1], but not to the mild cognitive impairment level [2]. The Trail Making Test Parts A & B (TMT-A & B), which has high sensitivity and specificity, and the Maze test, which is simpler than the CDT, tend to be used for testing executive function [3]. However, in addition to executive function, the TMT and the Maze test also assess visual attention and processing speed, whereas the CDT also assesses verbal function, memory and visuospatial cognition and/or construction. Therefore, depending on how the test battery is assembled, the CDT can be utilised [4]. Compared with the other tests, scoring the CDT requires a greater level of technical skill. Therefore, to simplify the CDT implementation procedure and introduce automatic scoring, a touch screen-based, digital-CDT (dCDT) [5] was developed by the authors and former members of the developmental team with the cooperation of Arc Co. Ltd. (Nagoya, Japan). This study reports on a preliminary investigation aimed at establishing an automatic scoring system for the dCDT. The aims of this investigation were: (1) to administer the dCDT to DST-takers and assess the usability of the device in a group of people who have a history of traffic violations, and (2) to compare the results using the scoring system of Rouleau and colleagues [6] (Rouleau method), which is widely used in clinical practice in the health and welfare sector, and the scoring system used by the NPA until May 2022 as part of cognitive function testing in driver training for older drivers (NPA method) [7]. Our research plan, including this investigation, was performed in accordance with the ethical standards laid down in the Declaration of Helsinki, and it was approved by the Research Ethics Committee of the Institute of Innovation for Future Society, Nagoya University (approval no. 2024-1). From 16 to 18 July 2024, we asked Driver's Licence Testing Centre A to cooperate with our research and administered the dCDT using a touch-screen personal computer to 38 people who agreed to participate (mean age ± standard deviation, 77.42 ± 3.71 years; male/female, 29/7). The dCDT procedure was as described previously [5]. No age difference was seen between the nine test-takers (three females, six males) who had failed the DST by scoring < 70 points and the other test-taker group. In addition, no significant difference in dCDT results was seen between the groups using either the Rouleau or NPA method. While a strong correlation was observed between the dCDT results using both scoring methods, no correlation was found between either scoring method or age. The Rouleau method can be said to be broadly homologous to the NPA method. Table 1 shows the sensitivity and specificity for failure on the DST for each cut-off score using the Rouleau method. The low specificity may be due to the fact that the test targets people who have already committed specific traffic violations; however, amongst those who failed the DST, some had a high score of 9 points that were found, which reconfirms the fact that screening using the CDT alone is difficult. In addition, improvements such as making the screen more ecologically appropriate and changing to a step-by-step instruction method must be made so that the test subjects can more easily understand the additional procedures required for automatic scoring without any extra effort. We intend to establish a formula for the scoring algorithm after increasing the number of subjects and controlling for noise (e.g., hand tremors, dry fingers). This work was supported by the National Agency for Automotive Safety and Victims' Aid (NASVA). Satsuki Yamauchi (Nagoya University) and Hiroko Shinkai (Nagoya University) helped develop the prototype and first model of the dCDT device as members of the development team. Arc Co. Ltd. (Nagoya, Japan) provided the source code for the dCDT. Our research plan, including this investigation, was approved by the Research Ethics Committee of the Institute of Innovation for Future Society, Nagoya University (approval No. 2024-1). All participants provided consent to participate in the study. The authors declare no conflicts of interest. Raw data were generated at Nagoya University. Participants of this study did not agree for their data to be shared publicly, so supporting data is not available.