Session on Negative Affect, Drinking Drivers, and Remedial Programs. A Population Survey of the Link Between Depressed Mood and Drinking Driving.

Transcription

1 T2007 Seattle, Washington Session on Negative Affect, Drinking Drivers, and Remedial Programs. A Population Survey of the Link Between Depressed Mood and Drinking Driving. Gina Stoduto 1, Patricia Dill 2, Robert E. Mann 1,3, Elisabeth Wells-Parker 2, Tony Toneatto 1,3, and Rania Shuggi 1. 1 Centre for Addiction and Mental Health, Toronto, ON, Canada; 2 Mississippi State University, Mississippi State, MS, U.S.A.; 3 University of Toronto, Toronto, ON, Canada. Abstract: Since both alcohol and depressed mood exert deleterious effects on psychomotor performance, the possibility that people with depressed mood may be more likely to drive after drinking may have important implications for traffic safety. Available studies in general population samples do not provide strong evidence of a link between depressed mood and drinking-driving. However, studies involving samples of convicted drinking drivers do suggest the presence of a large subgroup characterized by depressed mood, and that this group may be important for understanding recidivism risk and treatment responsiveness. In this work we examine the association between depressed mood and self-reported driving after drinking based on a large representative sample of adults in Ontario. Data are based on the Centre for Addiction and Mental Health Monitor, an ongoing cross-sectional telephone survey of Ontario adults aged 18 and older (n=3,979). Logistic regression analysis was performed to identify the risk of driving after drinking two or more drinks in the previous hour within the past 12 months associated with scores on a screening measure of depressed mood (depression-anxiety subscale of the 12-item General Health Questionnaire (GHQ-12)), while controlling for alcohol use measures (weekly volume and frequency of heavy drinking), driving exposure, and demographic factors. Logistic regression analysis revealed that the odds of reporting driving after drinking within the past year increase significantly as depressed mood increases. These results suggest that higher levels of depressed mood found in samples of convicted drinking-drivers are not simply due to a reaction to their recent experiences with the criminal justice system. Research on the nature of the link between depressed mood and impaired driving, including assessing whether any synergistic effects of depressed mood and alcohol on collision risk exist, and developing and evaluating remedial programs for offenders that address this link, should be considered a research priority. Introduction: Previous general population studies have provided no strong evidence for an association between depressed mood and driving after drinking (Macdonald and Mann 1996). Since both alcohol and depressed mood exert deleterious effects on psychomotor performance, the possibility that people with depressed mood may be more likely to drive after drinking may have important implications for traffic safety. The ability to assess the potential contribution of depressed mood to drinking driving in general population studies has been hampered by small numbers of studies, small sample sizes and a failure to control for potential confounders (Macdonald and Mann 1996). However, studies involving samples of convicted drinking drivers do suggest the presence of a large subgroup characterized by depressed mood (Lapham et al. 2001; Wells-Parker and Williams 2002; Wells-Parker et al. 2006). In this work we examine the association between depressed mood and self-reported driving after drinking based on a large representative sample of adults in Ontario.

2 Method: Our data are based on telephone interviews with 3979 respondents derived from the 2001 to 2004 cycles of the Centre for Addiction and Mental Health (CAMH) Monitor, an ongoing cross-sectional telephone survey of Ontario adults (aged 18 or older), conducted by CAMH and administered by the Institute for Social Research at York University (see Ialomiteanu and Adlaf 2005 for details). Results are based on valid response and responses such as don t know and refusals were considered missing data and excluded from analyses. The percentages reported are based on the weighted sample size and are considered representative for the population surveyed. The drinking driving item was: During the past 12 months, have you driven a motor vehicle after having two or more drinks in the previous hour? (coded yes=1, no=0). The 12-item version of the General Health Questionnaire (GHQ-12) is a widely used screening instrument measuring non-psychotic mental illness. Most factor analytic studies of GHQ-12 identify a 6-item factor of mixed depression and anxiety, commonly labeled dysphoria (Campbell et al. 2003; Wells-Parker et al. 2006); and 6-item factor of social functioning. A Likert scoring ( ) method was used to determine the GHQ-12 subscale scores (scores ranging from 0 to 18). A score of greater than four on depression-anxiety was employed to indicate depressed mood (see Wells-Parker et al. 2006) in order to examine demographic (gender, age, education, marital status, region) and drinking characteristics (weekly alcohol consumption and frequency of drinking 5+ drinks per occasion) of depressed (n=99) and non-depressed drinking drivers (n=285). Logistic regression analysis examined the risk of reporting drinking driving within the past 12 months with increases in depression-anxiety score while controlling for confounding effects of demographic factors, driving exposure (km driven in typical week) and alcohol factors (weekly alcohol consumption and frequency of drinking 5+ drinks). Contrasts for region were made to Toronto residents, household income level was compared to those earning less than $30,000, education level was compared to those with less than high school, those never married were the comparison group for marital status, and those who never drank heavily were the comparison group for drinking 5+. Results: Drinking driving by GHQ-12 subscales is presented in Table 1. The mean depressionanxiety score was significantly higher for drinking drivers compared to non-drinking drivers whereas the social functioning scores did not differ between groups. Examination of demographic and drinking characteristics by depressed mood among drinking drivers revealed that only region, and marital status differed significantly by depressed mood status. Table 2 presents the logistic regression model of drinking driving with demographic factors, driving exposure, alcohol factors and depression-anxiety and social functioning as independent measures. The Hosmer-Lemeshow test shows that the model provides a good fit of the data. Gender, region, income, weekly volume of drinks, frequency of drinking 5+ drinks on an occasion and depression-anxiety score were found to be significant predictors of drinking driving, controlling for other factors. The odds of drinking driving was over 3 times higher for males compared to females, 67% higher in the West region of Ontario compared to Toronto and 102% higher for those with incomes of $50,000-79,999 and $80,000+ compared to those with income less than $30,000. The odds of drinking driving increased significantly as weekly volume of drinks consumed increased (OR=1.03), and the odds were over 4 times greater for those who drank 5+ drinks during one occasion 2-3 times a month or less and 11 times greater from those who drank 5+ drinks once a week or more compared to those who never drank 5+ drinks on one

3 occasion. We also found that as the depression-anxiety score increased the odds of drinking driving increased. Each unit increase in the depression-anxiety score increased the odds that a respondent was a drinking driver by about 8%. The range (0-18) of the depression-anxiety scores indicate that a person at the higher end of this scale would be about twice as likely to be a drinking driver as a person at the lower end of the scale. Discussion: This study, based on population survey data from representative of the Ontario adult driving population, provides an opportunity to assess the impact of depressed mood on driving after drinking while controlling statistically for potential confounders. After controlling for these variables, increases in depressed mood significantly predicted driving after drinking. This observation is in agreement with studies of convicted drinking-driver samples (Lapham et al. 2001; Wells-Parker and Williams 2002; Wells-Parker et al. 2006), and underscores the overlap of samples of individuals with alcohol-related problems with samples of individuals with depressed mood. These results also suggest that higher levels of depressed mood found in samples of convicted drinking drivers may not simply be due to a reaction to their recent experiences with the criminal justice system. Several factors could help explain the elevations in depressed mood among convicted offenders and drinking drivers in the general population. These may include genetic predispositions and reactions to adverse life events other than those related to the impaired driving experience (Beck et al. 1979). Additionally, the elevation in depressed mood may be related to physiological and cognitive processes associated with heavy alcohol consumption (Schuckit et al. 1997; Ramsey et al. 2004). With regard to the latter possibility, Blume et al. (2001) and Wells-Parker et al. (2006) have postulated that the increase in depressed mood seen in some convicted impaired drivers and others with alcohol problems may reflect a motivational state that can be conducive to behavior change in a rehabilitative or remedial program. In support of this hypothesis, Wells-Parker and Williams (2002) found that when brief individual interventions based on Motivational Interviewing (Miller and Rollnick 2002) were added to a group-based remedial program for convicted drunk drivers, significant reductions in recidivism rate (about 30%) were observed, but only for offenders who had elevations in depressed mood The finding of an association between depressed mood and drinking driving is potentially important from several traffic safety perspectives. First of all, depressed mood may serve as a marker for increased likelihood of driving after drinking. As well, it is also possible that depressed individuals who drive after drinking may experience an increase in collision risk greater than that experienced by those not depressed. More serious, or clinical, forms of depression are known to affect psychomotor and cognitive function and may be associated with increased injury risk (e.g., Azorin et al. 1995). If depressed individuals drink to relieve their depression and then drive, their risk of collision involvement may exceed the risks well known to be associated with alcohol use by itself (Borkenstein et al. 1964; Mann et al. 2001). Thus, depressed mood individuals who drive after drinking may be at particularly high risk for collision involvement. If this is the case, education and prevention efforts may be needed to address the increased risks these individuals face. In summary, these results suggest that higher levels of depressed mood found in samples of convicted drinking-drivers are not simply due to a reaction to their recent experiences with the criminal justice system. Research on the nature of the link between depressed mood and impaired driving, including assessing whether any synergistic effects of depressed mood and

4 alcohol on collision risk exist, and developing and evaluating remedial programs for offenders that address this link, should be considered a research priority. References: Azorin, J.M., Benhaim, P., Hasbroucq, T., and Possamai, C.A. Stimulus preprocessing and response selection in depression: A reaction time study. Acta Psychol. 89: , Beck, A.T., Rush, A.J., Shaw, B.F., and Emery, G. Cognitive Therapy of Depression, New York: Guilford Press, Blume, A.W., Schmaling, K.B., and Marlatt, G.A. Motivating drinking behavior change: Depressive symptoms may not be noxious. Addict. Behav. 26: , Borkenstein, R.F., Crowther, R.F., Shumate, R.P., Ziel, W.B., and Zylman, R. The Role of the Drinking Driver in Traffic Accidents, Bloomington, Ind.: Dept. of Police Administration, Indiana University, Campbell, A., Walker, J., and Farrell, G. Confirmatory factor analysis of the GHQ-12: Can I see that again? Austral. New Zeal. J. Psychiat. 37: , Hosmer, D.W. and Lemeshow, S. Applied Logistic Regression. New York: John Wiley & Sons, Inc., Ialomiteanu, A. and Adlaf, E.M. CAMH Monitor: Technical Guide 2004, Toronto: Centre for Addiction and Mental Health, 2005, Lapham, S.C., Smith, E., C'de Baca, J., Chang, I., Skipper, B.J., Baum, G., and Hunt, W.C. Prevalence of psychiatric disorders among persons convicted of driving while impaired. Arch. Gen. Psychiat. 58: , 2001 Macdonald, S. and Mann, R.E. Distinguishing causes and correlates of drinking and driving. Contemp. Drug Probl. 23: , Mann, R.E., Stoduto, G., Macdonald, S., Shaikh, A., Bondy, S., and Jonah, B. The effects of introducing or lowering legal per se blood alcohol limits for driving: An international review. Accid. Anal. Prev. 33: 61-75, Miller, W. R. and Rollnick, S. Motivational Interviewing: Preparing People for Change (2nd Edition), New York: Guilford Press, Ramsey, S.E., Kahler, C.W., Read, J.P., Stuart, G.L., and Brown, R.A. Discriminating between substance-induced and independent depressive episodes in alcohol dependent patients. J. Stud. Alcohol 65: , Schuckit, M.A., Tipp, J.E., Bergman, M., Reich, W., Hesselbrock, V.M., and Smith, T.L. Comparison of induced and independent major depressive disorders in 2,945 alcoholics. Amer. J. Psychiat. 154: , 1997 Wells-Parker, E. and Williams, M. Enhancing the effectiveness of traditional interventions with drinking-drivers by adding brief individual intervention components. J. Stud. Alcohol 63: , 2002.

5 Table 1. Drinking driving by GHQ-12, CAMH Monitor GHQ-12 Subscales Non-Drinking Drivers (n=3,643) Drinking Drivers (n=410) Depression-Anxiety *** mean (sd) 2.75 (2.76) 3.36 (3.22) Social Functioning mean (sd) 5.84 (1.70) 5.89 (1.92) Statistical significance *** p<.001

6 Table 2. Logistic regression of drinking driving, CAMH Monitor (n=3,715) Drinking Driving OR (95%CI) Gender*** (male=1) 3.47 (2.58, 4.66) Age.99 (.98, 1.00) Region** (ref.=toronto) Central East 1.01 (.70, 1.45) Central West.84 (.58, 1.20) West** 1.67 (1.15, 2.43) East.87 (.58, 1.32) North.95 (.58, 1.56) Income*** (ref.=<$30,000) $30,000-49, (.78, 2.42) $50,000-79,999** 2.02 (1.20, 3.42) $80,000+*** 2.41 (1.44, 4.03) not stated 1.26 (.67, 2.38) Education (ref.=<high school) Completed high school.77 (.49, 1.22) Some post-secondary.82 (.53, 1.28) University degree 1.06 (.67, 1.67) Marital status (ref.= married/partner) Previously married 1.23 (.79, 1.92) Never married 1.31 (.95, 1.79) Weekly drinks volume*** 1.03 (1.01, 1.04) Freq. five+ drinks*** (ref.=none) 2-3 times per month or less*** 4.15 (2.82, 6.12) Once per week or more*** (7.14, 17.54) Km driven in typical week 1.0 (1.0, 1.0) GHQ-12 depression-anxiety*** 1.08 (1.03, 1.12) GHQ-12 social functioning 1.01 (.95, 1.08) Constant.006 Hosmer-Lemeshow Test χ 2 =5.55, df=8, p<.698

7 Statistical significance Wald Test * p<.05; ** p<.01; *** p<.001