<Back    TITLE PAGE    Forward>

Executive Summary

NOTE: This project is a continuation of Evaluating Drivers Licensed with Medical Conditions in Utah, 1992-1996, which was completed in June 1999 (DOT HS 809 023, available from the National Technical Information Service, Springfield, Virginia 22161). In order that this report be comprehensible as a stand-alone document, some of the introductory and explanatory materials from that report are repeated here.

The Utah Driver License Division operates a specialized licensing program for drivers who have medical conditions. The program was developed by the Division under the guidance of the Utah Medical Advisory Board, a group of physician advisors. The intent of the board was to create a program that improved public safety while imposing on drivers the fewest possible restrictions consistent with that goal. The program guidelines describe the physical, mental and emotional capabilities appropriate for various types of driving and thus determine license eligibility. The major function of the program is to identify drivers whose ability to drive may be impaired by their medical conditions, and then to control the impact of these limitations on traffic safety. This is accomplished by means of regulating how and when they may legally drive (that is, the imposition of restrictions on their driving privileges). Restrictions on licenses may include speed, area and/or time of day limitations. The program is based on self-reporting of medical conditions by applicants for driver's licenses. Drivers are first categorized by their specific medical conditions (termed functional ability categories by the program) and then by their ability to drive [termed functional ability level (1-12)]. The imposition of a restriction, and the kind of restriction, is indicated by the functional ability level. Drivers who are licensed with medical conditions may receive a full unrestricted or restricted license depending on their functional ability level.

The functional ability or medical condition categories include:

  1. diabetes mellitus and other metabolic conditions
  2. cardiovascular conditions
  3. pulmonary conditions
  4. neurological conditions
  5. epilepsy and other episodic conditions
  6. learning/memory/communications
  7. psychiatric or emotional conditions
  8. alcohol and other drugs
  9. visual acuity
  10. musculoskeletal abnormalities/chronic medical debilities
  11. functional motor ability
  12. hearing

The Utah Crash Outcome Data Evaluation System project (CODES) was funded to evaluate the effect of the existing medical condition licensing program on public safety. The project was funded by the National Highway Traffic Safety Administration (NHTSA), with the support of the Utah Driver License Division in the Utah Department of Public Safety, and the Utah Department of Transportation.

In order to measure the effectiveness of this public safety program, we compared the citation, total crash and at-fault crash rates of drivers licensed with medical conditions to those of similar drivers matched on age group, gender and county of residence. Comparison drivers were obtained randomly from the general driving population and rates of adverse driving events were examined over a five-year period, 1992-1996. A two-to-one matching strategy was used. Sampling was performed with replacement.

Analyses were conducted for each functional ability category (medical condition) by each functional ability level. Analyses for drivers licensed with two medical conditions were conducted separately, by restriction status. We used probabilistic linkage to link data elements relating to the same driver from several different databases in order to combine the elements needed for the study (i.e., crash, violation and driver license databases). Many subgroups contained very small numbers of drivers; analysis was reported only when the number of subjects was sufficiently large to allow for meaningful analysis.

The rates of citation, crashes and at-fault crashes, expressed as events per 10,000 licensed days, varied between the populations and events of interest. No functional ability category appeared to be at greater risk for citations; some levels in the two largest categories, diabetes and cardiovascular, actually had lower rates of citations (for levels 3, and 3-4 respectively) than controls. For several medical conditions, there were significantly higher rates of crashes and at-fault crashes than for controls, mostly for functional ability levels 3-5. Crash rates were significantly higher than controls for diabetes (levels 3 & 4), cardiovascular (level 5), neurological (levels 3-5), epilepsy (level 3-6), psychiatric (levels 3-6), and visual acuity (levels 3-6 and level 8). For at-fault crashes, rates were significantly higher than control for diabetes (levels 3-5), pulmonary (level 3), neurological (levels 3-5), epilepsy (levels 3-6), psychiatric (levels 3-6), and visual acuity (levels 3-6 and level 8). The magnitude of increased frequency for these groups was generally moderate, with odds ratios typically in the 1.5-2.5 range. For a few functional ability categories, crash and at-fault crash rates were higher than comparison drivers at the highest functional ability levels (that is, the most restricted); however the actual number of crash events in these subgroups was very small. Of note, several functional ability categories were so small (memory and communications disorders; alcohol and other drugs) that analysis of each functional ability level for them was not meaningful and was not done.

Drivers who were licensed in more than one functional ability category (reported more than one medical condition) during the study period were considered separately for the most common combinations of conditions. Only two-way combinations were analyzed, since there were few drivers reporting three or more separate medical conditions. Also, many of the possible combinations contained so few subjects that analysis was not meaningful and is not reported. Again because of small subgroup numbers, the functional ability levels were collapsed into two groups, restricted and unrestricted drivers. First, unrestricted drivers (functional ability levels 3-5) were analyzed. No combination had higher rates of citations than controls, while two combinations (diabetes+cardiovascular, and cardiovascular+neurological) actually had lower rates than controls. Crash rates were lower than controls for one group (diabetes+cardiovascular) and higher than controls for several others: diabetes+neurological, diabetes+psychiatric, diabetes+vison, neurological+musculoskeletal, neurological+functional motor, and musculoskeletal+functional motor. The magnitude of difference was moderate, with odds ratios being in the range 1.5-2.3 for the most part. Rates of at-fault crashes for unrestricted drivers with two medical conditions were higher than controls for diabetes+cardiovascular, diabetes+neurological, diabetes+psychiatric, diabetes+vison, neurological+musculoskeletal, and musculoskeletal+functional motor. Odds ratios ranged from 1.4 to 3.0. For this multiple medical conditions group, numbers of crash and at-fault crash events were low, 30-100 except for the largest combination group diabetes+cardiovascular.

Then, the multiple medical conditions group was analyzed for restricted drivers (functional ability level 6-11). No combination had higher citation rates than controls, while one group (cardiovascular+neurological) had lower rates. Crash rates were higher than control for 4 groups: diabetes+cardiovascular, neurological+musculoskeletal, nerulogical + functional motor, and musculoskeletal+functional motor. Odds ratios were 2.0-2.4 and numbers of crash events was low, 11-17. At-fault crash rates were higher than control for 6 groups: diabetes+cardiovascular, diabetes+visual acuity, cardiovascular+visual acuity, neurological+musculoskeletal, musculoskeletal+functional motor, and musculoskeletal+functional motor. Odds ratios were 2.6-3.3, and the numbers of events were again low, 8-12 for each group.

The results of this study provide contextual information on the effects of the medical conditions licensing program on public safety. When analyzed by specific medical condition and functional ability level, medical conditions drivers overall had sporadically elevated rates of crashes and at-fault crashes compared with control drivers. For specific conditions, elevations in adverse driving events, if they occurred, were in the numerically lowest functional ability levels (that is, levels where driving privileges were least restricted). For the subset of drivers with multiple medical conditions, several specific combinations of medical conditions were associated with higher rates of crashes and at-fault crashes compared with controls. For most of these multiple conditions groups, absolute numbers of both drivers and adverse events were very low.

As with any injury control intervention, evaluation is an essential component of the program in order to identify areas of increased risk and to provide feedback to the administering agency. The aim of this research was to evaluate the rates of adverse driving events by individual functional ability levels in order to determine if there were distinct levels for which risk increases or decreases, and to describe the effects of co-existing medical conditions. Some subgroups of medical conditions drivers did appear to have increased rates of crashes and at-fault crashes compared with controls. Almost all of these subgroups were in numerically low functional ability levels, such that their driving privileges were not restricted. The differences in rates were generally of modest magnitudes, and conceivably could be deemed acceptable. The increase in crash and at-fault crash rates might be addressed by changing the functional ability levels to restrict the driving privileges of currently unrestricted drivers. Such a change would have to be broad, covering all or nearly all of the medical conditions categories, since the increase in rates was seen in most of the categories. Further, it is not known that such a change in the program would affect the rates of crashes and at-fault crashes for these drivers. For the drivers with multiple medical conditions, there were groups with increased rates of crashes and at-fault crashes in both restricted and unrestricted functional ability levels. The absolute numbers of crashes, however, were so small that program alterations specifically to address this may not be worth considering.

<Back    TITLE PAGE    Forward>