Visual impairment and driving restrictions
Digital Journal of Ophthalmology 2002
Volume 8, Number 1
May 1, 2002
To determine the adverse incident rates of visually impaired drivers at specific levels of driving restriction and to compare these rates to unimpaired, unrestricted drivers.
Visually impaired drivers were identified FROM statewide databases containing over 1.7 million individuals. Probabilistic linkage methodology was used to link databases containing information on drivers' licenses, medical conditions, citations and crashes. Exposed drivers with visual impairment (n=1,527) were compared to an unrestricted, unimpaired cohort (n=2,939) at three levels of driving restriction and visual impairment (low, moderate, high) for rates of citations, crashes, and at-fault crashes. Exposed drivers were case matched to unexposed cohorts based on age, gender, and geographic location, and the degree of driving restrictions included limitations on speed, area, or time of day. The outcomes were expressed as odds ratio for each of three levels of restriction.
When compared to the unimpaired, unrestricted drivers, visually impaired drivers demonstrated a significantly increased number of citations for two of the three levels of driving restriction (OR 1.201-2.514); significantly increased number of crashes for two of the three levels of driving restriction (OR 1.196-2.866); and significantly increased number of at-fault crashes for two of the three levels of driving restriction (OR 1.511-3.836). The highest odds ratios were associated with the most impaired and most restricted drivers.
Visually impaired, restricted drivers are still involved in more motor vehicle crashes and citations than unimpaired, unrestricted drivers, but this increase is modest for drivers with moderate visual impairment and restriction. However, the higher rate of adverse incidents in the most impaired subjects with the most restrictions warrants further investigation.
Driving, Visual Impairment, License
|Driving is a highly valued activity for most individuals, (1) and the loss of this ability greatly reduces a person's sense of independence and well-being.(2-4) However, in the interest of public safety, most states restrict driving privilege for people with certain medical conditions.(5) Adequate vision is arguably one of the most important capacities for safe driving, and visual impairment is a common reason for people to restrict their driving. (6,7)|
Previous studies have shown that 13% of drivers over the age of 55 have a best-corrected visual acuity worse than 20/40.(8) Although visual impairment may lead to a higher rate of adverse driving incidents, the exact effect of visual impairment upon adverse driving incidents is not known because of conflicting data. (9-12) Adverse driving incidents include citations, crashes, and at-fault crashes. Some types of eye disease, such as cataract, may lead to a particularly elevated risk. (13,14) The question becomes even more complicated when one considers that individuals with visual impairment still legally drive, albeit with graduated restrictions, in many states.
We compare the rate of adverse driving incidents in individuals with impaired vision and driving restrictions to unimpaired, unrestricted drivers. We compare the citation rate, crash rate, and at-fault crash rate of non-commercial drivers with and without visual impairment for the entire state for a five-year period. The purpose of this study was to determine the adverse incident rates of impaired drivers at specific levels of restriction and impairment and thereby determine if these restrictions appear to be adequate.
|Materials and Methods|
|Drivers in the State of Utah may be required to enter INTO a medical conditions program based upon the results of a self-administered questionnaire and an eye screening examination administered by the Utah Driver License Division. Drivers in the medical conditions program are assigned a functional ability profile level. Restrictions for each functional ability level range the gamut FROM trivial (listed as a participant in the medical conditions program only but no real driving restrictions) to severe (driving prohibited). These assignments are made by the Utah Driver License Division and take INTO consideration the presence of one or more of the 12 medical conditions shown in TABLE 1. Each medical condition has its own criteria for assignment to a particular functional ability level. The Utah Driver License Division Medical Advisory Board, which is a GROUP of volunteer specialty physicians, developed these guidelines. Division personnel can directly assign a person to a functional level or refer a person for a more comprehensive medical evaluation depending upon the results of the questionnaire or eye-screening test. Because applicants have to pass an objective eye-screening exam consisting of Snellen acuity and gross peripheral fields, visual function is the only one of the 12 medical conditions that does not rely entirely on self-report. Thus, the net product of the medical conditions program is the imposition of certain driving restrictions on drivers who, according to the guidelines, have medical conditions that may impair their ability to drive. The medical conditions program implemented by Utah was implemented almost two decades ago (15) but little information exists to evaluate the efficacy of this and similar programs.|
Although restrictions can be imposed for any of the 12 medical conditions, the focus of this article is visual impairment. Furthermore, we specifically focus our attention on three levels of restriction shown in TABLE 2. TABLE 2 shows both the type of restriction and the degree of visual impairment. Functional ability levels 6, 7, and 8 are the labels used by the Utah Driver License Division and denote actual levels of driving restriction. We labeled the groups with descriptive terms of high, moderate, and low that refer to the highest degree of restrictions and impairment, the intermediate levels of restriction and impairment, and the lowest degree of restrictions and impairment, respectively. Other functional ability levels are used for commercial drivers, drivers prohibited FROM any driving, and individuals who are not restricted but may need more frequent reevaluation; these other levels are not included in this study.
This study was conducted using information FROM several statewide databases. The driver license file obtained FROM the Utah Driver License Division identifies individual drivers and whether or not that person has been assigned a functional ability level because of impairment. This database contained over 1.7 million drivers. Police crash reports were obtained FROM the Utah Department of Transportation, and death certificate DATA was obtained FROM the Utah death certificate database. These DATA sources are population-based, containing information on every relevant individual and/or event in the state of Utah for the years under consideration. Probabilistic linkage was used to link the databases for eventual analysis. Probablistic linkage is an iterative process that joins DATA FROM disparate databases for records relating to the same person or event, and is able to overcome inaccuracies in the databases, including incorrect, missing or duplicate data, typographical errors, and changes in surname. Probabilistic linkage has been amply described elsewhere. (16,17) Automatch Software ® (Matchware Technologies, 1996) was used to carry out this linkage. Ultimately, this technique allowed us to accurately determine the true number of adverse incidents in a given population as opposed to relying on patient's self-report. Previous studies have shown only a modest correlation between state records and patient's self-reported number of adverse incidents. (18,19)
Using a retrospective cohort design, we used the linked database to identify exposed individuals with visual impairment and driving restrictions (n=1,527) and case-match them to unexposed, unimpaired, unrestricted drivers (n=2,939) based on age group, gender, and county of residence for the 5-year period of 1992-1996 for the entire state. Age GROUP was calculated in 5-year increments, and the person was assigned to a GROUP based upon their age at the mid-point in the study period. Both groups were restricted to non-commercial licenses. The exposed GROUP did not have any other non-visual driving restrictions. The unexposed GROUP was obtained by random sampling of the unimpaired, unrestricted driving population with replacement and two-to-one matching. We had to use case matching because we did not know the actual number of miles driven by a given individual. However, we reasoned that individuals of the same location, gender, and age would have similar driving habits.
An individual driver may change his or her level of restriction over a period of time based upon changing impairment (or death, loss of license, etc), so the total number of days licensed in a particular level had to be derived to accurately case match the exposed and unexposed participants. For the unexposed group, the counting period started on 1/1/92 (study start date) or the first day of issue if they were first licensed in the study period and the end date was the study end date, date of death, or license expiration. For the exposed group, the number of days licensed in their particular functional ability GROUP was counted; the start date was the day of initial licensing or renewal during the study period, and the end date was the study end date, date of death, or license expiration. These calculations allowed us to express the rate of adverse events as number of adverse events per 10,000 licensed days. Essentially, this metric allows standardized comparison of risk based upon exposure. TABLE 3 shows the number of exposed and the total number of days of exposure. The latter figure is the more important measurement.
Analysis was conducted for visual impairment for each of the three levels of restriction and impairment. We defined an 'at-fault' crash as one that was caused or contributed to by the driver as per the accident report. An adverse event rate was calculated for all three levels for both exposed and unexposed participants, and an odds ratio was calculated. A Chi-square test was performed to evaluate significance of this ratio based upon one degree of freedom and a 95% confidence interval.
|Individuals with visual impairment and driving restrictions exhibited significantly higher rates of citations, crashes, and at-fault crashes when compared to the unexposed GROUP for the least impaired, least restricted GROUP and the most impaired, most restricted group. The intermediary GROUP did not demonstrate a significantly higher adverse incident rate compared to the unexposed cohort. |
Table 4 shows the results for citation rates. The odds ratio was lowest in the moderate GROUP (OR 1.201) and highest for the high GROUP (OR 2.514). TABLE 5 shows the results for crash rates. The odds ratio was lowest in the moderate GROUP (OR 1.196) and highest in the high GROUP (OR 2.866). Finally, TABLE 6 shows the results for at-fault crashes. In general these odds ratios were higher than those for the citation ratios. The odds ratio was lowest in the moderate GROUP (OR 1.511) and highest in the high GROUP (OR 3.836). In other words, even with high driving restrictions, individuals with visual acuity worse than 20/80 and visual field of 90 degrees were much more likely to be the cause of a motor vehicle crash when compared to their unimpaired, unrestricted peers.
|In this study, we determined the rate of adverse driving incidents for visually impaired, restricted drivers in three categories, and compared this rate to an unimpaired, unrestricted, unexposed cohort. The DATA suggests that some groups of visually impaired, non-commercial drivers exhibit a higher rate of citations, crashes, and at-fault crashes compared to their unimpaired, unrestricted peers in spite of driving restrictions. For the drivers with minor or moderate impairment and restrictions, the increase in risk was either not significant, or if it was significant, the odds ratio was generally between 1 and 2. Drivers with higher levels of impairment and higher levels of restriction were at much higher risk compared to their matched unexposed cohorts.|
The GROUP with the highest restriction and highest impairment showed a rather sharp increase in crash and citation rates compared to the other groups. This shift deserves special comment. The authors note that the high category represents a step-wise decrease in visual acuity to 20/80 or worse, and a large reduction in the total visual field compared to the moderate category. One explanation for this phenomenon may be that the patients in this category have reached a certain level of visual impairment that compromises driving ability regardless of restrictions. Some studies have suggested that binocular visual field impairment in particular has a great affect upon driving performance; (20-23) controlled field test have supported this notion (24). One study showed a 2.2 times increase crash rate over a 3-year period with a 40% or greater loss of visual field (25). Another study reported a 2.5 times rate increase of at-fault crashes in drivers with cataracts (26). These latter statistics appear to be comparable to our results. Viewed in the context of our new findings, those studies support the possibility that driving restrictions may not be sufficient once a person reaches a level of visual function that corresponds to impairment as described in the high category. On the other hand, this category represented a relatively small number of drivers and even smaller number of crashes; this figure probably does not indicate a public health hazard.
There are two aspects of this study that deserve special comment. First of all, the investigators do not know the driving exposure rate of the individual drivers. The rates of adverse driving events could theoretically be expressed in terms of real exposure, that is, miles driven. Previous studies confirm that older drivers with visual impairment are two times more likely to report reductions in days driven and number of destinations per week, driving slower than other drivers, and preferred somebody else to drive (27). Unfortunately, this type of true exposure DATA is virtually impossible to obtain for large groups of people. To control for this issue, the investigators had to use case matching as a proxy for true exposure. Furthermore, using miles-driven in the denominator is not a panacea, for such approach ignores the element of time. For instance, individuals may present little risk to themselves or society if their total miles-driven per year is small, even if their risk as expressed in incidents per miles-driven is high.
The second aspect of this study that deserves comment is the problem of comparing a study GROUP with two types of exposure: a level of driving restriction and a level of visual impairment. At least one other study suggests that graduated, restricted licensing based on visual test reduces the fatal crash rate (28). We cannot, however, make a direct conclusion that driving restrictions prevent adverse driving incidents. After all, the rate of adverse events may be the same regardless of the presence of a restriction. While we would ideally like to have a study controlling only one exposure, we have difficulty envisioning an acceptable study design with sufficient power to answer such a question.
We can say, however, that there does not appear to be a major increase in the rate of adverse incidents in the lesser-impaired drivers, but more attention should be directed toward the most impaired, restricted drivers. FROM a public health perspective, we may direct more attention to drivers classified INTO this higher GROUP for additional analysis or restriction. The study had other strong points. First of all, the linked database is very large and comprehensive; it contains all Utah drivers and covers all urban and rural counties. Review of TABLE 3 demonstrates the necessity of a very large study population. Even with 1.6 million drivers and a 5-year study interval, the number of people in several of the categories was not very large. Secondly, the database matched over 97% of the motor vehicle incidents in the entire state to the appropriate individual. This accuracy allowed us to accurately case-match the visually impaired, restricted individuals to unimpaired, unrestricted drivers to calculate reliable odds ratios.
As a clinical recommendation, we suggest that providers pay particular attention to those drivers who exhibit a higher degree of visual impairment; this level is defined by 20/80 - 20/100 best-corrected acuity and 90° or greater total visual field. The DATA may not be conclusive, but at the very least, physicians need to recognize that this particular level of vision places patients at higher driving risk in spite of restrictions and additional counseling may be considered. Like most other States, Utah has implemented a licensing program that attempts to balance public safety with the needs of the individual to be mobile and independent. (29-30) This study was conducted, in part, to evaluate the ongoing success of this strategy and provide objective DATA for policy-makers.
Table 1: Medical Categories
Learning, Memory, Communication Problems
Alcohol and Drugs
Functional Motor Impairment
Table 2: Study Class and Driving Restrictions
1 Utah Driver License Division Functional Ability Level 2 Descriptive label applied for this study
Table 3: Exposed Group
Table 4: Visual Impairment and Citations
Table 5: Visual Impairment and Crashes
Table 6: Visual Impairment and At-Fault Crashes
|Intermountain Injury Control Research Center, University of Utah, Salt Lake City|
National Highway Traffic Safety Administration, Contract DTNH22-96-H-59017
National Library of Medicine, Grant 5 T15 LM0712
|1. Wilkinson ME. Driving and visual impairment. Insight. 1998 Jun; 23(2): 48-52.|
2. Johnson JE. Urban older adults and the forfeiture of a driver's license. J Gerontol Nurs 1999 Dec; 25 (12): 12-8.
3. Park W. Vision rehabilitation for age-related macular degeneration. Int Ophthalmol Clin 1999 Fall; 39 (4): 143-62.
4.Marattoli RA, Ostfeld AM, Merrill SS: Driving cessation and changes in mileage driven among elderly individuals. J Gerontol 48: S255-S260, 1993.
5.Fishbaugh J. Look who's driving now- visual standards for driver licensing in the United States. Insight 1995 Dec; 20(4):11-20.
6.Owsley C, McGwin G Jr. Vision impairment and driving. Surv Ophthalmol. 1999 May-June; 43(6): 535-50.
7.Sivak M. The information that drivers use: is it indeed 90% visual? Perception 1996; 25(9): 1081-9.
8.Rubin GS, West SK, Munoz B, et al, and the Salisbury Eye Evaluation Project Team. A comprehensive assessment of visual impairment in a population of older Americans. Invest Ophthalmol Vis Sci. 1997; 38: 557-68.
9.McCloskey LW, Keopsell TD, Wolf ME, Buchner DM. Motor vehicle collision injuries and sensory impairments of older drivers. Age Ageing 1994; 23: 267-73.
10. Gresset JA, Meyer FM. Risk of accidents among elderly car drivers with visual acuity equal to 6/12 or 6/15 and lack of binocular vision. Ophthalmic Physiol Opt. 1994; 14: 33-37.
11.Decina LE, Staplin L: Retrospective evaluation of alternative vision screening criteria for older and younger drivers. Accid Anal Prev 25:267-275, 1993.
12.Crancer JA, McMurray L: Accident and violation rates of Washington's medically restricted drivers. JAMA 205: 272-276, 1968.
13.Osley C, Stalvey B, Wells J, Sloane ME. Older drivers and cataract: driving habits and crash risk. J Gerontol A Biol Sci Med Sci 1999 Apr; 54(4): M203-11.
14.Klein R. Age-related eye disease, visual impairment, and driving in the elderly. Hum Factors 1991 Oct; 33(5): 521-5.
15.Hales RH. Functional ability profiles for driver licensing. Exemplification by visual profile. Arch Ophthalmol 182 Nov; 100(11): 1780-83.
16.Jara MA. Probabilistic linkage of large public health DATA files. Statistics in Medicine 1995; 14: 491-98.
17.Shevchenko IP, Lynch JT, Mattie AS, Reed-Fourquet LL. Verification of information in a large medical database using linkages with external databases. Statistics in Medicine 1995;14: 511-30.
18. McGwin G Jr, Owsley C, Ball K. Identifying crash involvement among older drivers: aggreement between self-report and state records. Accid Anal Prev 1998 Nov; 30(6): 781-91.
19.Marottoli RA, Cooney LM, Tinetti ME. Self-report verses state records for identifying crashes among older drivers. J Gerontol 52A:M184-M187, 1997.
20.Owsley C, McGwin G Jr, Ball K. Vision impairment, eye disease, and injurious motor vehicle crashes in the elderly. Ophthalmic Epidemiol 1998 Jun; 5(2): 101-13.
21.Owsley C. Vision and driving in the elderly. Optom Vis Sci 1994 Dec; 71 (12): 727-35.
22.Johnson CA, Keltner JL. Incidence of visual field loss in 20,000 eyes and its relationship to driving performance. Arch Ophthalmol 1983; 101: 371-375.
23.Gresset JA, Meyer FM. Risk of accidents among elderly car drivers with visual acuity equal to 6/12 or 6/15 and lack of binocular vision. Ophthalmic Physiol Opt 1994 Jan; 14 (1): 33-7.
24.Wood JM, Troutbeck R. Effect of visual impairment on driving. Hum Factors 1994 Sep; 36(3): 476-87.
25.Owsley C, Ball K, McGwin G Jr, Sloane ME, Roenker DL, White MF, Overley ET. Visual processing impairment and risk of motor vehicle crash among older adults. JAMA 1998 Apr 8; 279 (14): 1083-8.
26.Owsley C, Stalvey B, Wells J, Sloane ME. Older drivers and cataract: driving habits and crash risk. J Gerontol A Bio Sci Med Sci. 1999 Apr; 54(4): M203-11.
27.Ball K, Owsley C, Stalvey B, Roenker DL, Sloane ME, Graves M. Driving avoidance and functional impairment in older drivers. Accid Anal Prev 1998 May; 30(3): 313-22.
28.Levy DT, Vernick JS, Howard KA. Relationship between driver's license renewal policies and fatal crashes involving drivers 70 years or older. JAMA 1995 Oct 4; 274(13): 1026-30.
29.Retchin SM, Anapolle J. An overview of the older driverl.Clin Geriatr Med 1993; 9: 279-96.
30.Waller JA. Chronic medical conditions and traffic safety.N England J Med 1965; 273 (26): 1413-20.