Medical risk factors amongst drivers in single-car accidents

Transcription

1 Journal of Internal Medicine 1997; 241: Medical risk factors amongst drivers in single-car accidents T. GISLASON, K. TO MASSON, H. REYNISDO TTIR, J. K. BJO RNSSON & H. KRISTBJARNARSON From the Department of Psychiatry, National University Hospital, 101 Reykjavı k, Iceland Abstract. T. Gislason, K. To masson, H. Reynisdo ttir, J. K. Bjo rnsson, H. Kristbjarnarson (Department of Psychiatry, National University Hospital, Reykjavı k, Iceland). Medical risk factors amongst drivers in single-car accidents. J Intern Med 1997; 241: Objectives. To estimate medical risk factors amongst drivers in single-car accidents with special reference to sleepiness and alcohol abuse. Design. An epidemiological survey by means of multiple-choice questionnaires that were mailed to drivers and a random control group. Setting. Iceland. Subjects. All drivers in single-car accidents (n 471) during and a control group from the general population (n 1000). Main outcome measure. Medical health profile and answers to questions concerning sleep disorders and alcohol abuse. Results. Compared to the controls the drivers were younger and there were three times more males. The drivers abused alcohol more often. Chronic disorders such as epilepsy, diabetes mellitus, and cardiac disorders were not over-represented. Altogether, 15.4% claimed that sleepiness had caused their traffic accident, logistic regression analyses revealed that these sleepy drivers more often had alcohol abuse and a history of almost falling a sleep whilst driving. Conclusions. Compared to controls, drivers in singlecar accidents are more likely to be young, male, have a history of sleepiness whilst driving, and also have a history suggestive of alcoholism. Chronic disorders like epilepsy and diabetes mellitus were not overrepresented amongst the single-car accidents drivers. These results raise the question of how drivers with a high probability of causing an accident (sleepiness and alcoholism) can be identified in time and proper measures taken to prevent accidents waiting to happen. Keywords: accidents, alcoholism, drivers, epidemiology, epilepsy, sleepiness. Introduction Traffic accidents obviously have different causes, some of which are medical. Consciousness can be disturbed for different reasons such as because of epilepsy, diabetes mellitus, cardiovascular diseases, sleepiness, drug use or alcohol [1]. The possible connection between epilepsy and traffic accidents has received the most attention. In a recent review, however, it was pointed out that although there is a slight increase in prevalence of traffic accidents amongst epileptic patients only one-tenth of their accidents can be connected to epileptic seizures [2]. Some studies have indicated that traffic accidents are more common amongst diabetic patients [3] whilst others claim that there is no need to have special regulations concerning diabetic drivers [4]. Based on their survey in Wisconsin, the authors concluded that there was no reason to limit the right of epileptic or diabetic patients to drive [5]; however, this study was later heavily criticized [6]. In the US, it is estimated that almost 40% of fatal traffic accidents are due to alcohol consumption [7]. Drowsiness fatigue was cited on about 1% of police accident reports on all crashes reported to the national highway traffic safety administration [8]. In a recent research note this estimate of 1% has been considered conservative and other studies on large crash data files have generally yielded estimates in the 1 4% range [8]. When either sleepiness or sleep is the reason for the accident, the consequences are usually very serious [1]. Sleepiness can be caused by sleep apnoea syndrome (SAS), narcolepsy, the use of sedative drugs, alcohol use or disturbed night 1997 Blackwell Science Ltd 213

2 214 T. GISLASON et al. sleep. SAS patients are often involved in traffic accidents [9, 10] and accidents are more common the more severe the sleep apnoea syndrome [11, 12]. Previous assumptions that traffic accidents occur more often amongst certain groups of patients with, for example, epilepsy and SAS, are usually based on some comparison between an arbitrarily selected group of patients and different control groups. Our knowledge about the importance of each disorder in causing traffic accidents is therefore very limited. The aim of this study was to compare a well-defined group of drivers involved in single-car accidents, and a group selected from the general population with special regard to medical health profile, alcohol consumption, and sleepiness. Subjects and methods Drivers In cooperation with the Icelandic Traffic Safety Council, we were given permission by the Icelandic Data Inspection to gather information from all police records in Iceland about drivers involved in singlecar accidents. The study group was confined to those drivers in single-car accidents where someone was injured and needed medical attention (cases). Accidents involving a death were excluded. The study period covered the years Police files contained the records of 471 drivers involved in a serious accident: 336 men and 135 women, with a mean age of 30 years (Table 1). Controls One thousand individuals 17 years old and over were randomly selected from the National Population Register. One person was later eliminated from the Table 1 Participation and nonresponders Drivers Controls Death 1 1 Sickness 0 21 Living in other countries Refused to participate Not able to locate Replies not usable 1 2 Tel.: answers Mail replies Total sample size control group as he was also in the case group. Thus, 999 individuals constituted the control group (Table 1). Questionnaire We asked about chronic disorders requiring regular medical check-ups in the preceding 5 years, regular drug use, and especially about the use of sedatives or hypnotics. Included were questions about driving licence, mileage driven per year, how the accident occurred, and especially about whether the case considered that sleepiness and or tiredness actually contributed to the accident. There were questions about sleep-related symptoms and the respondents answered on a 5-point scale: 1, never; 2, less than once a week; 3, 1 2 times a week; 4, 3 5 times a week; and 5, daily [13]. We asked about alcohol consumption, possible answers being: yes I do drink or I do not drink. Furthermore, four additional alcohol questions were asked: the CAGE questions (See Appendix). These questions were first introduced in 1968 and their reliability shown in 1974 [14]. Their use in medical work and studies has been described [15]. These questions are considered to be more sensitive than laboratory measures to find individuals suffering from alcoholism [16]. Three or four positive answers are considered to indicate alcoholism. Procedure The questionnaires were mailed in the beginning of January 1992 and 10 days later a reminder or thank-you card was mailed to everyone. A new questionnaire was mailed to all nonrespondents, both 5 and 9 weeks later. One month later we randomly selected 50 controls and 50 cases amongst the nonresponders and tried to locate them, mainly by phone. Participation In all, 342 (72.6%) usable questionnaires came from the cases and 742 (74.2%) from the controls. Male cases responded to a somewhat lesser extent than females (71.4 vs. 75.6%). Amongst male cases, those in the age group years had a low response level (64.1%). Amongst the controls, males and females had similar response levels, but participation

3 MEDICAL RISK FACTORS AMONGST DRIVERS 215 was lowest amongst those 70 years of age (51.8%). Analyses of the nonresponders showed that the majority of the controls were indeed willing to participate, or 34 of 50 (Table 1). Only eight refused, two were sick, and six could not be located. It was more difficult to locate the cases: 25 of 50 answered a telephone inquiry, six refused, 14 could not be located, and five were living abroad. Statistical analyses A logistic regression analysis was used to estimate the probability of having had a traffic accident in both groups together. The dependent variable was the group (control or cases), and the explanatory variables were gender, history of snoring, daytime sleepiness, history of nearly falling asleep whilst driving, positive CAGE answers [0 4], and mean mileage driven per year. These analyses were done separately for each age group and also with age as a dependent variable. A similar model using a forward stepwise procedure was used for the group of cases to estimate the probability of having a sleep-related accident. Results The 342 cases compared to the 742 controls were more often males (70.2 vs. 47.3%). They were much younger as their mean age (mean SD) was years compared to (P 0.001). Altogether, 61.4% of the cases were under 30 years of age, compared to 26.6% of the controls. The cases reported more mileage driven per year: 25.7% of them drove more than miles per year compared to 10.1% of the controls (χ 84.6; P ). Chronic disorders and drug use Six of the controls reported having diabetes mellitus, compared to three of the cases. Five controls reported epilepsy as compared to four cases. Forty-nine of the controls had a history of cardiovascular diseases, compared to 15 of the cases. The use of hypnotics and sedatives was low in both groups and the comparison nonsignificant. Sleep-related symptoms The controls reported more often that they got enough sleep compared to the cases (78.0 vs. 69.0%) (χ 9.5; P 0.002) (Table 2). Enough sleep was more often reported, amongst the older subjects both controls (χ 12.4; P 0.006) and cases (χ 17.9; P ) (Table 3) with no gender difference. Snoring was more common amongst the controls: 13.2% snored habitually and 22.5% occasionally, compared to 6.8 and 26.6% of the cases (χ 9.2; P 0.01) (Table 2). Habitual snoring (score 5) was more common amongst the Table 2 Sleep-related symptoms and alcohol consumption amongst controls, all drivers and those drivers who claimed that sleepiness had caused their accidents ( Sleepy drivers ) Controls All drivers Sleepy drivers (n 742) (n 342) (n 48) Gets enough sleep 563 (78.0%) 231 (69.0%) 27 (60.0%) Snoring habitually 87 (13.2%) 21 (6.8%) 2 (4.9%) occasionally 149 (22.5%) 82 (26.6%) 16 (39.0%) Daytime sleepiness habitually 50 (6.9%) 31 (9.3%) 6 (12.8%) occasionally 319 (44.3%) 183 (54.8%) 32 (68.1%) Yes to 3 4 CAGE questions 47 (6.3%) 49 (14.3%) 12 (27.9%) Almost falling asleep whilst driving never or very seldom 650 (94.1%) 293 (87.5%) 35 (74.5%) less than once per week 34 (4.9%) 37 (11.0%) 11 (23.4%) 1 2 week 6 (0.9%) 4 (1.2%) 1 (2.1%) more often 1 (0.1%) 1 (0.3%) 0

4 216 T. GISLASON et al. Table 3 Total number and proportion (%) by age of those who claim they get enough sleep, complain of habitual snoring, habitual daytime sleepinessor answer yes to 3 4 CAGE questions Controls All drivers Sleepy drivers Age (years) (n 742) (n 342) (n 48) Gets enough sleep (65.6%) 58 (58.6%) 7 (63.6%) (74.2%) 74 (63.8%) 12 (60.0%) (77.7%) 55 (78.6%) 4 (50.0%) (84.8%) 44 (88.0%) 4 (66.7%) Habitual snoring 20 4 (6.4%) 5 (5.1%) (7.8%) 4 (3.3%) (12.7%) 5 (6.9%) 1 (11.1%) (14.7%) 7 (14.0%) 1 (16.7%) Habitual daytime sleepiness 20 7 (11.1%) 16 (16.2%) 1 (9.1%) (6.5%) 9 (7.5%) 2 (9.1%) (3.6%) 4 (5.5%) 2 (22.1%) (10.1%) 2 (4.0%) 1 (16.7%) Yes to 3 4 CAGE questions 20 2 (3.2%) 15 (15.1%) 2 (18.2%) (11.7%) 21 (17.5%) 7 (31.8%) (6.8%) 7 (9.6%) 2 (22.2%) 50 6 (2.8%) 6 (12.0%) 1 (16.7%) older individuals, both controls and cases (χ 37.9; P ) (Table 3), and also three times more common amongst males than amongst females (χ 57.8; P ). Daytime sleepiness was more common amongst cases than controls (χ 15.3; P ) (Table 2). Habitual daytime sleepiness (score 5) was most common amongst those under 20 years of age (χ 92.2; P ) (Table 3). Amongst drivers complaining of habitual daytime sleepiness less than half (48%) reported enough sleep. Complaints of insomnia (difficulties initiating sleep, maintaining sleep and early morning awakening) were not over-represented amongst those with habitual daytime sleepiness. Daytime sleepiness was somewhat more common amongst women, but the difference was not statistically significant (χ 2.1; P 0.3). Altogether, 48 cases (15.4%) reported that sleepiness had contributed to their traffic accidents and 60 (19.1%) reported tiredness as a contributing factor. Amongst these, 37 reported both sleepiness and tiredness. These sleepy cases reported somewhat more often having symptoms of SAS, but the difference was not statistically significant (Table 3). The cases reported more often than the controls that they had almost fallen asleep whilst driving (χ 16.1; P 0.003) (Table 2) and those cases who had reported sleepiness as a contributing factor also reported almost fallen asleep whilst driving proportionately more often than the remaining cases (χ 9.8; P 0.03). Alcohol consumption A larger proportion of the controls did not use alcohol compared to the cases (27.7 vs. 19.4%) (χ 8.4; P 0.005). This difference is not significant if we only compare those under the age of 30 (16.9 vs. 14.3%; χ 0.6; P 0.5] and disappears if we compare only young men (14.1 vs. 13.7%). The cases answered more CAGE questions positively than the controls (mean SD) ( vs ; P ). Altogether, 14.3% of the cases answered three or four CAGE questions positively compared to 6.3% of the controls (χ 18.5; P ) (Table 2). Men answered three or four CAGE questions more often than women (χ 35.3; P ), both amongst the controls (9.7 vs. 3.3%) and the cases (19.2 vs. 2.9%). Those cases that had reported sleepiness as a contributing factor to their traffic accidents answered these questions positively almost twice as often as the other cases (χ 5.9; P 0.015) (Table 2). Table 3 shows that the CAGE answers indicated alcoholism almost five

5 MEDICAL RISK FACTORS AMONGST DRIVERS 217 Table 4 Logistic regression analyses. Estimates the possibility of belonging to the driver group or the control group B SE P Gender Age group CAGE Mileage per year Daytime sleepiness Falling asleep whilst driving Constant times more often amongst cases under the age of 20, as compared to the controls, and there was also a difference amongst the older individuals. The majority of those cases who had answered three or four CAGE questions positively and considered that sleepiness was a contributing factor (n 12) (Table 2) were years of age (Table 3). Based on the univariate analyses above, the following model was designed to estimate the probability amongst all responders of being a case : 1 1 exp A B C constant coefficient gender coefficient age-group coefficient daytime sleepiness coefficient. almost falling asleep whilst driving coefficient mileage per year coefficient CAGE D Table 4 shows the weight of each variable. All variables except daytime sleepiness contributed significantly to the model above. The probability of being a case was higher amongst the younger subjects and amongst males. Symptoms of alcoholism, long mileage driven and reporting almost falling asleep whilst driving increased the probability of being a case. The same analyses were performed in different age groups (Table 5). The Table shows that men aged years and 61 years are at increased risk. The same is true for the very young and middle aged who score high on the CAGE and or have high yearly mileage. Having the tendency to fall asleep whilst driving appears to be a special risk factor amongst those aged The same type of model was also used for the cases only to estimate the probability of having reported sleepiness as a factor contributing to the traffic accident. Only two variables contributed significantly to the model: alcohol consumption (CAGE) (P 0.01) and reporting almost falling asleep whilst driving P 0.05). Thus from the model: 1 1 exp A B CAGE (0 4) 0.63 almost falling asleep whilst driving (0 3) the probability for all cases of having an accident with sleepiness as a contributing factor can be estimated to be 5% for those who claim that they have never almost fallen asleep whilst driving and answer no CAGE questions positively. On the other hand the probability estimate of having sleepiness as a contributing factor to the accident increased to 57% for those who answered all CAGE questions positively and said they often almost fall asleep whilst driving. Discussion One of the advantages of investigating well-defined populations is that we can collect reliable epidemiological data about certain groups. We chose to look for medical risk factors amongst drivers involved in serious single-car accidents involving personal injury in comparison to a random group from the total population. The response rate was acceptable in both groups except amongst the oldest controls, and most of the cases as well as the controls were indeed C D Table 5 Logistic regression analyses. Estimates the possibility of belonging to the driver group or the control group in different age groups (beta values presented) * P 0.05; **P Age group (years) Gender ** * CAGE 0.478* * Mileage per year 0.412** ** Daytime sleepiness Falling asleep whilst driving ** Constant ** 3.130** 0.686

6 218 T. GISLASON et al. willing to participate once we had contacted them (Table 1). The typical case was a young male with a history of alcohol abuse. Chronic disorders like epilepsy, diabetes mellitus, and cardiac diseases were not commonly encountered amongst the drivers and do not seem to be important in considering the total number of traffic accidents. A survey like this, however, does not describe the increased risk of each individual with a specific chronic disease of having an accident as there were so few accidents that could be related to the chronic disorders above. Altogether there were 48 drivers (15.4%) who considered that sleepiness had contributed to their traffic accidents. This is a much higher proportion than has been attributed to sleepiness in previous reports [8]. As a group our sleepy drivers were young (Table 3), and had a high prevalence of sleeprelated symptoms except habitual snoring. For the drivers as a group, there was only a statistical difference in relation to the answers to CAGE questions and to the question about almost falling asleep whilst driving. The increased probability of a sleep-related accident amongst those with symptoms of alcohol abuse and a history of almost falling asleep whilst driving was clearly illustrated in the logistic regression analyses. Almost falling asleep whilst driving is particularly a problem in the age group years old (Table 5). Symptoms of alcohol abuse are especially common in the youngest age group (17 19 years) and amongst the middle aged. A recent study found sleepiness and low-dose ethanol to impair simulated automobile driving in a synergetic way [17]. Thus, when a patient has a history of alcohol abuse or sleepiness whilst driving and in particular the combination of both, his doctor should inform him that his lifestyle involves an increasing risk of being involved in a traffic accident. In a previous Icelandic investigation on the prevalence of alcoholism amongst individuals years of age during , the prevalence of alcoholism amongst men was 12.1% but 2.8% amongst women [18]. Our results concerning the prevalence of alcoholism are very comparable to these. From the point of view of prevention, some of the results of our survey must be brought to public attention. Alcoholism was three times more common amongst the men than the women in the control group, but amongst drivers alcoholism was six times more common amongst the men than the women. Amongst very young drivers alcoholism was five times more common than amongst the controls in the same age group. Measures taken by society to decrease the number of alcohol-related traffic accidents should therefore be concentrated on young drivers who abuse alcohol, and especially on men. These measures should also be aimed towards those with a history of almost falling asleep whilst driving. Considering the high prevalence of alcohol abuse amongst young drivers, the question arises whether year-olds with a history of alcohol abuse should be allowed to have a driving licence. Recent studies have shown the relationship between sleep apnoea syndrome and traffic accidents [19, 20]. It was pointed out, however, that although there was an increase in traffic accidents related to increased alcohol use and perhaps to diabetes and cardiac diseases [20], the treatment of these chronic disorders was often difficult. On the other hand, sleep apnoea syndrome could easily be diagnosed and treated [19, 20]. In a recent study it has been shown that the positive effect on vigilance and driving performance remains after 4 years [21]. It has been proposed that considering the seriousness of traffic accidents amongst patients with sleep apnoea syndrome, doctors should ask as many patients as possible about the major symptoms of the sleep apnoea syndrome [22]. This survey shows that answering more than three questions about alcohol use positively hugely increased the probability of being involved in a traffic accident, as did a history of almost falling asleep whilst driving. These results raise the question of how drivers with a high probability of causing an accident can be identified in time and proper measures taken to prevent accidents waiting to happen. Only if measures are taken jointly by health authorities, traffic security authorities, and insurance companies can we expect any success in preventing serious accidents. References 1 Parsons M. Fits and other causes of loss of consciousness while driving. Q J Med 1986; 277: Krumholtz A, Fisher RS, Lesser RP, Hauser WA. Driving and epilepsy a review and reappraisal. JAMA 1991; 265: Davis TG, Wehing EH, Carpenter RE. Oklahoma s medically restricted drivers a study of selected medical conditions. Oklahoma State Medical Association Journal 1973; 66: Saunders CJP. Driving and diabetes mellitus. Br Med J 1992; 305: 1265.

7 MEDICAL RISK FACTORS AMONGST DRIVERS Hansotia P, Broste SK. N Engl J Med 1991; 324: Correspondence (four letters). N Engl J Med 1991; 324: Stinson FS, DeBakey SF. Alcohol-related mortality in the United States, Br J Addiction 1992; 87: Knipling RR, Wang J-S. Crashes and fatalities related to driver drowsiness fatigue. Research note, November, National Highway Traffic Safety Administration. Washington DC: Office of Crash Avoidance Research. Research and Development. 9 Guillminault C, van den Hoed J, Mitler M. Clinical overview of sleep apnea syndrome. In: Guillminault C, Dements W, eds. Sleep Apnea Syndromes. New York: A. R. Liss, 1978; George CF, Nickerson PW, Hanly PJ, Millar TW, Kryger MH. Sleep apnea patients have more automobile accidents. Lancet 1987; 22, August: Findley LJ, Unverzagt ME, Suratt PM. Automobile accidents involving patients with obstructive sleep apnea. Am Rev Respir Dis 1988; 138: Findley LJ, Fabrizio M, Thommi G, Suratt PM. Severity of sleep apnea and automobile crashes. N Engl J Med 1989; 320: Partinen M, Gislason T. Basic Nordic Sleep Questionnaire (BNSQ): a quantitated measure of subjective sleep complaints. J Sleep Res 1995; 4 (Suppl.): Mayfield DG, McLeod G, Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry 1974; 131: Ewing JA. Detecting alcoholism. The CAGE questionnaire. JAMA 1984; 252: Beresford TP, Blow FC, Hill E, Singer K, Lucey MR. Comparison of CAGE questionnaire and computer-assisted laboratory profiles in screening for covert alcoholism. Lancet 1990; 336: Roehrs T, Beare D, Zorick F, Roth T. Sleepiness and ethanol effects on simulated driving. Alcohol Clin Exp Res 1994; 18: Helgason T. Prevalence and incidence of alcohol abuse in Iceland. In: Cooper, B, ed. Pyschiatric Epidemiology, 1987: London: Croom Helm. 19 Stradling JR. Obstructive sleep apnoea and driving sufferers need medical advice. Br Med J 1989; 298: Findley LJ, Bonnie RJ. Sleep apnea and auto crashes what is the doctor to do? Chest 1988; 94: Haraldsson PO, Carenfelt C, Lysdahl M, Tornros J. Long-term effect of uvulopalatopharyngoplasty on driving performance. Arch Otolaryngol Head Neck Surg 1995; 121: Findley LJ, Woodrow W, Jabour ER. Drivers with untreated sleep apnea a cause of death and serious injury. Arch Intern Med 1991; 151: Appendix The CAGE questions (C) Have you ever felt you should cut down on your drinking? (A) Have people annoyed you by criticizing your drinking? (G) Have you ever felt bad or guilty about your drinking? (E) Have you ever had a drink first thing in the morning to steady your nerves or to get rid of a hangover (eye-opener)? Received 7 May 1996; accepted 8 October Correspondence: Thorarinn Gislason MD PhD, Vifilsstadir Pulmonary Department, 210 Gardabaer, Iceland ( thorarig rsp.is).