Women with DUIs in this high risk sample evidenced greater physiological and behavioral dependence on alcohol than men with DUIs, and may thus represent a particularly severe alcohol dependence phenotype. Women with multiple DUIs were more likely to report wanting to quit and unsuccessful attempts to quit drinking than men. Thus, despite a desire to quit, women with multiple DUIs had less success with quitting drinking than men. Combined with evidence that fewer women than men are referred to treatment after a first DUI offense (Lapham et al., 2000
), that treatment may not be as effective for female as male offenders, and that women are less likely to complete treatment (Maxwell & Freeman, 2007
), this highlights an opportunity to improve treatment for women with DUIs. Examination of gender differences in alcohol criteria endorsement can potentially improve treatment for women in mandated DUI programs and, more broadly, for women at the severe end of the alcohol dependence spectrum.
Previous work found greater prevalence of alcohol abuse and dependence among male then female first-time DUI offenders (Lapham et al., 2001
) but higher rates of alcohol dependence among female than male recidivists (Lapham et al., 2006
; Laplante et al., 2008
). In the current study, inspection of individual alcohol criteria revealed evidence of greater severity among female offenders, including those with just one DUI. Female DUI offenders were more likely than males to drink despite negative consequences such as marital, health and emotional problems. Continued use in the context of negative life consequences may be associated with behavioral under control, a trait which might be over-represented among women with DUIs. In a survey of drivers from the general population, women expressed more social and moral inhibitions against drinking and driving than did men (Marelich, Berger, & McKenna, 2000
), which suggests that women who do drive after drinking are not similarly constrained. Behavioral under control might also be associated with the high endorsement of alcohol withdrawal by women with DUIs, since it shares some genetic risk with alcohol dependence (Slutske et al., 2002
), and a related concept, novelty-seeking, is associated with greater familial risk for alcohol dependence (Grucza et al., 2006
). While there is evidence that individuals with histories of DUI or drinking and driving display more impulsivity, risk-taking, sensation-seeking, and aggressive behavior than individuals without such histories (Donovan, 1993
; Wilson, 1992
), those studies included only small samples of women and did not examine gender differences. Our examination of individual alcohol criteria highlights gender differences among DUI offenders which have not previously emerged.
Our findings stand in contrast to an earlier study of gender differences in alcohol criteria endorsement among alcohol-dependent COGA participants (Schuckit et al., 1998
). While the current study found that women with DUIs endorsed items indicating greater severity more frequently than similar men, the previous study found just the opposite among all alcohol dependent individuals. For example, Schuckit et al. (1998)
found that more men than women reported periods of time dominated by drinking, but the current study found that women with multiple DUIs were more likely to report this than men. The same pattern was true for unsuccessful attempts to quit and for items about negative consequences of drinking, such as problems with family or friends. Regarding measures of physiological dependence, the current study found that women with multiple DUIs were more likely to report tolerance to alcohol, withdrawal, and alcohol-related seizures, whereas men endorsed these items more frequently in the earlier study (Schuckit et al., 1998
). These differences indicate that even in a high-risk sample such as COGA, women with histories of DUI represent a severe alcohol dependence phenotype.
Rates of lifetime drug use and dependence among repeat offenders in this sample were higher than the rates in other samples of repeat DUI offenders (Lapham et al., 2006
; Laplante et al., 2008
). The differences in prevalence rates between our study and those of previous DUI studies may result from our sample selection of families at high risk for alcohol dependence, which may also have correlated high risks for drug dependence (Kendler, Prescott, Myers, & Neale, 2003
). The individuals recruited from treatment programs for DUI offenders in earlier studies may have had fewer or more varied familial vulnerabilities to substance use disorders. Differences in rates of drug use may also be due to reporting differences. Because individuals in earlier studies were recruited from treatment settings where they’d been referred due to DUI legal problems, they may have been reluctant to report drug use for fear it might add to their existing legal troubles. By contrast, the DUI variable in this study was based on self report rather than arrest records, and individuals may have been more willing to accurately report drug use.
This study’s finding that more women than men developed dependence on marijuana and opiates is at odds with findings from other DUI samples (Lapham et al., 2001
), including one study reporting no gender differences in rates of drug dependence (Laplante et al., 2008
), and a study reporting higher rates of hallucinogen dependence among males than females (Lapham et al., 2006
). Because those studies did not condition drug dependence on drug use as we did in our analysis, we tested gender differences in our data a second time without conditioning on use and the gender differences remained significant. It is possible that women were more susceptible than men to the numbing effects of marijuana and opiates, given their higher endorsement of continuing to drink despite health and psychological problems.
Overall, rates of psychiatric comorbidity in this sample were higher than in other studies of DUI offenders (Lapham et al., 2006
; Lapham et al., 2001
; Laplante et al., 2008
), and were consistently high across all DUI categories. Women and men differed in the lifetime prevalence of MDD, CD, and ASPD. In the case of externalizing disorders, it is noteworthy that while men with three or more DUIs had significantly higher rates than men with one or two DUIs, this was not true of women. This suggests that women who commit even a single DUI offense deviate more from women who do not than do men, consistent with a hypothesis of greater behavioral undercontrol among women with DUIs.
This study cannot be interpreted as being representative of all individuals with DUIs or with alcohol dependence, as the study protocol was designed to inform genetic studies and required a certain family size and loading for alcohol diagnoses. The measure for DUI, unlike many studies of DUI populations, is based on self report rather than official documentation. The study also relies on retrospective recall of symptoms for alcohol dependence, psychiatric disorders, and drug use disorders. Nonetheless, it contributes to our understanding of the relationship between DUIs and severe alcohol dependence, and in particular highlights several interesting areas for study among women with DUIs. For example, gender differences in endorsement of behavioral alcohol criteria may indicate areas amenable to psychosocial intervention which may improve treatment outcomes and lower risks for initial or multiple DUIs among women with severe alcohol problems. The gender difference in withdrawal endorsement may indicate areas ripe for gender-specific pharmacological research. In future studies with this population we will examine in detail the time course of alcohol use, psychiatric and drug use disorders in relation to the timing of arrests for DUI.