Abstract
Aims
This retrospective study compared drinking histories of 283 men and 413 women convicted of driving while impaired (DWI) in New Mexico and interviewed 15 years following a first conviction and screening referral.
Design
We characterized drinking course and plotted drinking status (stable abstainers, abstainers, moderate, or risky drinkers) from age 15 to 60.
Setting
Pacific Institute for Research and Evaluation, Albuquerque, New Mexico.
Participants
Community sample of previously convicted DWI offenders.
Measurements
Psychiatric disorders from the Comprehensive International Diagnostic Interview; drinking histories from the Cognitive Lifetime Drinking History.
Findings
Risky drinking was prevalent at all ages for both genders. Almost half the population reported either a lifetime drinking course of risky drinking (19%) or resumed risky drinking after at least one interval of abstinence or moderate drinking (25%), while about one fifth followed a never-risky or risky-to-moderate drinking course. Offenders with a lifetime diagnosis of substance dependence more often transitioned to risky drinking, and those with lifetime alcohol dependence were more prone to transition to abstinence. Across time, those who began risky drinking at age 15 or later quit at double the rate of those who began before age 15. Women’s and men’s drinking courses were similar, but women began risky drinking at a later age and more often moved to abstinence.
Conclusions
Among people convicted of driving while impaired in the US, younger age of initiation of drinking and co-occurrence of psychiatric and substance use appear to be associated with a poorer trajectory of subsequent risky drinking behaviour. Women who are convicted of driving while impaired appear to start drinking later in life and be more likely subsequently to become abstainers.
Introduction
Persons convicted of driving while impaired or “intoxicated” (DWI) are at high risk for having, or developing, alcohol use disorders [1–5] and other psychiatric disorders [5–7]. While a number of studies have focused on the course of alcohol use disorders in general population samples [8–10] long-term changes in drinking patterns among DWI offenders have not previously been examined. Such examination is needed because, as our previous study suggests, alcohol use disorders may persist for many years in this population.[11]
The presence of comorbid psychiatric conditions is an important predictor of poor adherence to treatment and of relapse following treatment for alcohol use disorders.[12,13] However, to our knowledge no study has examined whether lifetime substance-use and other psychiatric disorders of convicted DWI offenders affect long-term drinking patterns. While long-term follow-up studies are rare, a number of studies have followed DWI offenders with substance use and other comorbid disorders over a shorter period. One study of 290 first-time DWI offenders in treatment for substance use disorders ascertained the prevalence of psychiatric comorbidity at intake and followed offenders for 1 year after discharge. The presence of a lifetime anxiety or mood disorder (30% of the sample) was associated with lower coping confidence, greater readiness to change, and greater and more enduring negative consequences of drinking during the DWI intervention and follow-up period.[7] A prospective study examined the demographic and clinical characteristics of 729 repeat DWI offenders with PTSD symptoms who were admitted to a 2-week inpatient facility. At baseline and at 1-year follow-up participants with PTSD had more severe psychiatric comorbidity and reported a higher recidivism rate than those without PTSD.[14] Wells-Parker and colleagues found that following a brief intervention, DWI offenders who scored high on a scale measuring depression were less likely to be re-arrested compared with offenders who showed no evidence of depression [15]. Another study suggests that motivational interventions are more effective among DWI offenders with antisocial personality disorder than those without it [16,17].
This retrospective study interviewed 696 convicted DWI offenders who participated in a follow-up study approximately 15 years after their first conviction and referral for substance abuse screening. The aim of this study was to determine variations in lifetime drinking course and to evaluate gender differences in age of onset and persistence of risky alcohol use.
Methods
Setting and Participants
Participants for this analysis were referred to the Lovelace Comprehensive Screening Program (Screening Program) between April 1989 and March 1992 following a first conviction for DWI and completed interviews 5 and 15 years after this referral. Participants were selected for analysis regardless of Screening Program completion or whether they were referred to treatment. The Screening Program was a court-mandated service for DWI offenders convicted of a first offense in Bernalillo County Metropolitan Court, NM, providing assessment for substance use disorders and treatment referrals for those with identified disorders. Traffic records and participant self-reports indicated that about 80% were truly first offenders.[18] This DWI offender population is similar to those described in other U.S. studies of convicted DWI offender populations with respect to age, marital status, and arrest blood alcohol concentration, but has a higher proportion of women, Hispanics, and Native Americans.[2,18–22] For the initial study we selected 1,208 consecutive female and 1,407 male referrals. At the 15-year follow-up, of the 2,615 selected participants, 156 were deceased, and 716 were interviewed (Figure 1).[11] Comparisons of those interviewed at 15-year follow-up to those originally selected who were not known to be deceased (N=2,459) revealed that men, Mexican nationals, those with an arrest warrant, those without telephones, and those who did not complete screening were under-represented in the 15-year follow-up sample (Table 1). At screening a higher proportion of men reported alcohol dependence (70%) compared to women (61%) but a higher proportion of women had non-substance psychiatric disorders.[5]
Figure 1. Study Flow Diagram for 15-year Follow-up Study.
Table 1.
Participant characteristics as a percentage of those selected for the original sample and who were not known to be deceased at the time of the 15-year follow-up.
|
Variable at screening |
Selected from Screening Program Referrals |
Interviewed at 15-year follow-up |
p-value |
|---|---|---|---|
| N | N (%) | ||
| Gender | |||
| Men | 1,322 | 285 (22%) | <0.001 |
| Women | 1,137 | 431 (38%) | |
| Age at screening | |||
| <31 | 1,517 | 457 (30%) | 0.16 |
| 31+ | 942 | 259 (27%) | |
| Ethnicity | |||
| Non-Hispanic White | 924 | 267 (29%) | <0.001 |
| Hispanic | 982 | 305 (31%) | |
| Native American | 292 | 96 (33%) | |
| Mexican National* | 123 | 10 (8%) | |
| Other | 138 | 38 (28%) | |
| Arrest warrant | |||
| No | 2,224 | 668 (30%) | 0.002 |
| Yes | 235 | 48 (20%) | |
| Telephone | |||
| No | 281 | 59 (21%) | 0.002 |
| Yes | 2,178 | 657 (30%) | |
| Arrest BAC | |||
| <0.15 | 865 | 250 (29%) | 0.92 |
| 015 or more | 1,239 | 365 (29%) | |
| Unknown | 355 | 101 (28%) | |
| Alcohol diagnosis at screening | |||
| No | 1,189 | 365 (31%) | 0.08 |
| Yes | 731 | 214 (29%) | |
| Unknown | 539 | 137 (25%) | |
| Screening category | |||
| Not Complete | 463 | 112 (24%) | 0.03 |
| Complete and Not Referred to treatment | 965 | 299 (31%) | |
| Complete and Referred to treatment | 1,031 | 305 (30%) | |
Significantly different, p< 0.05 from other ethnic categories
For the 15-year follow-up study bilingual (English and Spanish) staff used comprehensive location and data collection protocols, and informed consent procedures approved by the Pacific Institute for Research and Evaluation Institutional Review Board. Location protocols included a letter sequence, telephone calls, and home visits. Once located, willing participants provided written informed consent and were given $100 to complete the interview. Interviewers were trained in data collection protocols and administering the diagnostic instrument. All interviews were reviewed to monitor consistency.[18] Discrepancies were discussed during team meetings to standardize coding. Participants who resided out of town or who could not travel to the research site were interviewed by telephone.
The Interview
The interview consisted of demographic and health-related information, arrest histories, and opinions regarding law enforcement efforts to reduce impaired driving. It also included a computerized version of the Composite International Diagnostic Interview (CIDI).[23–25] This interview was initially developed for the World Health Organization and the U.S. Mental Health Administration, to estimate prevalence rates of specific psychiatric disorders. Questions are fully scripted and close-ended, highly structured, and appropriate for use by non-clinician interviewers. The version used, the 10th revision, provides DSM-IV diagnoses. Disorders assessed for the present study included lifetime alcohol and drug abuse and dependence, major depressive disorder (MDD), posttraumatic stress disorder (PTSD), generalized anxiety disorder, bipolar, social phobia, specific phobia, agoraphobia, and panic disorder.
We also administered the Cognitive Lifetime Drinking History (CLDH) [26], a measure of lifetime alcohol consumption that assessed onset of regular drinking and ages when drinking patterns changed to define lifetime intervals when those patterns were fairly stable. The CLDH defined intervals of life during which drinking patterns were relatively homogeneous by asking patients when they began to drink regularly, when their drinking changed, whether they continued to drink regularly after it changed, and, if not, whether they ever started drinking regularly again. Drinking patterns were assessed for each of the defined intervals. For intervals during which respondents drank weekly or more often, patterns were assessed by asking how often respondents drank on Fridays during a typical month during that interval and how many drinks they usually had when they drank on a Friday during that interval. These quantity-frequency (QF) questions were repeated for Saturdays, Sundays, week-days, and days when patients drank more than usual. For intervals during which respondents drank less often than weekly, they were simply asked about usual drinking quantity and frequency.
To be included in the present analysis, participants must have reported they had consumed at least 12 drinks in their lifetime, and must have completed both the CIDI and the CLDH. There were 696, 283 men and 413 women, individuals who met these inclusion criteria.
Measures
“Total drinks” was translated into total standardized drinks containing 14 grams of ethanol by obtaining information on the types of alcoholic beverages drunk, drink size for each beverage type, and the percent of all drinks represented by each beverage type; this information was used to estimate total ethanol consumed in grams and divided by 14 to calculate total standardized drinks. Risky drinking intervals were defined as those in which women reported usually drinking 4+ drinks on any drinking day or >7 drinks per week, and in which men reported usually drinking 5+ drinks on any drinking day or >14 drinks per week. Moderate drinking intervals were defined as intervals during which participants reported regular drinking, but did not meet the criteria for risky drinking. Abstinent intervals are those during which participants say they were no longer drinking regularly (e.g., at least one drink a month for at least six months).
To determine differences in drinking patterns among participants, we categorized individuals according to six lifetime drinking courses, as follows: (a) participants who never reported a risky drinking pattern (never-risky course); (b) those who reported risky drinking during at least one interval and then became abstinent, with no subsequent drinking intervals (risky-to-abstinent course); (c) those who reported risky drinking during at least one interval and subsequently reported moderate drinking, with no return to risky drinking (risky-to-moderate course); (d) those who resumed risky drinking after at least one interval of abstinence or moderate drinking (resumed-risky course); and (e) those who reported risky drinking throughout their lifetimes (all-risky course). Participants who reported some other pattern were classified as having a variable course. We compared the percentage of individuals following each drinking course by gender, ethnicity (non-Hispanic white, Hispanic, Native American, other), education group (< 12 years, 12 years, >12 years), presence of lifetime alcohol and/or drug dependence, and the presence of a lifetime non-substance psychiatric disorder (MDD, PTSD, generalized anxiety disorder, bipolar, social phobia, specific phobia, agoraphobia, or panic disorder).
We defined drinking status by their reports of current drinking at any given age (risky, moderate, recent abstainer, stable abstainer) from 15 to 60 years during each reported drinking interval. Recent abstainers were abstinent for under 3 years. Stable abstainers were abstinent for 3 or more years. We then graphed rates of risky drinking, moderate drinking, recent abstention, and stable abstention by age, from 15 to up to 60 years, by whether or not they qualified for a lifetime diagnosis of alcohol dependence and by gender. Data were truncated at age 60 due to small sample sizes.
Statistical Methods
Chi-square tests were used to test for differences among groups. Where there were 3 or more groups per category, we used chi-square terms for individual cells to examine differences. Preliminary analysis indicated that those who started risky drinking at a very young were more likely to be in the group that resumed risky drinking. Therefore, we developed a multivariate Cox proportional hazards model of the number of years from initiation of risky drinking until quitting risky drinking. This model included all participants except those who were never risky drinkers. The dependent variable (outcome) for the multivariate model was the time from the onset of risky drinking to stopping, and not returning to, risky drinking. The independent variables were age at initiation of risky drinking, gender, lifetime alcohol dependence (Y/N), lifetime drug dependence (Y/N), and non-substance psychiatric diagnosis (Y/N).
Results
Participants
Women comprised 59% of the sample. There was a higher proportion of Hispanics among female than male participants, and women were less likely than men to report drug dependence (Table 2). Rates of alcohol dependence were similar by gender, but women were far more likely than men to have a lifetime diagnosis of a psychiatric disorder.
Table 2.
Participant characteristics at 15-year follow-up.*
| Characteristic | Men (N=283) |
Women (N=413) |
Chi-square p-value |
|---|---|---|---|
| n (%) | N (%) | ||
| Race/Ethnicity | |||
| Non-Hispanic White | 102 (36%) | 127 (31%) | 0.03 |
| Hispanic | 117 (41%) | 213 (52%) | |
| Native American | 58 (20%) | 70 (17%) | |
| Other | 6 (2%) | 3 (1%) | |
| Education group | |||
| < 12 year | 35 (12%) | 49 (12%) | 0.78 |
| 12 year | 70 (25%) | 112 (27%) | |
| > 12 year | 178 (63%) | 252 (61%) | |
| Alcohol dependence | |||
| No | 129 (46%) | 211 (51%) | 0.14 |
| Yes | 154 (54%) | 201 (49%) | |
| Drug dependence | |||
| No | 178 (63%) | 299 (73%) | 0.006 |
| Yes | 105 (37%) | 112 (27%) | |
| Non-Substance diagnosis** | |||
| No | 180 (64%) | 158 (38%) | <0.001 |
| Yes | 102 (36%) | 254 (62%) | |
Values may not add to total due to missing data.
Non-substance diagnosis includes major depressive disorder, PTSD, generalized anxiety disorder, bipolar, social phobia, specific phobia, agoraphobia, and panic disorder.
Drinking Course
Almost half the population reported either a lifetime of risky drinking (19%) or resumed risky drinking after at least one interval of abstinence or moderate drinking (25%), while about one fifth of all offenders followed a never-risky (8%) or risky-to-moderate (14%) drinking course. (Table 3). Twenty-one percent became abstinent and the remaining (13%) reported a variable pattern of drinking intervals. There were no gender or ethnic differences in the proportion of participants in each drinking course category (Table 3).
Table 3.
Characteristics associated with lifetime drinking course.
| Drinking course |
|||||||
|---|---|---|---|---|---|---|---|
| Variable | All-Risky 134 (19%) |
Never- Risky 54 (8%) |
Risky-to- Abstinent 143 (21%) |
Risky-to- Moderate 96 (14%) |
Resumed- Risky 177 (25%) |
Variable 92 (13%) |
p-value |
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | ||
| Gender | |||||||
| Men | 60 (21%) | 22 (8%) | 58 (20%) | 41 (14%) | 64 (23%) | 38 (13%) | 0.76 |
| Women | 74 (18%) | 32 (8%) | 85 (21%) | 55 (13%) | 113 (27%) | 54 (13%) | |
| Race/ethnicity1 | |||||||
| Non-Hispanic White | 42 (18%) | 16 (7%) | 44 (19%) | 28 (12%) | 63 (28%) | 36 (16%) | |
| Hispanic | 66 (20%) | 30 (9%) | 70 (21%) | 48 (15%) | 75 (23%) | 41 (12%) | 0.51 |
| Native American | 23 (18%) | 8 (6%) | 29 (23%) | 20 (16%) | 38 (30%) | 10 (8%) | |
| Other | 3 (33%) | 0 (0%) | 0 (0%) | 0 (0%) | 1 (11%) | 5 (56%) | |
| Education2 | |||||||
| <12 years | 25 (30%) | 6 (7%) | 24 (29%) | 8 (10%) | 15 (18%) | 6 (7%) | 0.005 |
| 12 years | 30 (16%) | 19 (10%) | 47 (26%) | 21 (12%) | 42 (23%) | 23 (13%) | |
| >12 years | 79 (18%) | 29 ( 7%) | 72 (17%) | 67 (16%) | 120 (28%) | 63 (15%) | |
| Alcohol Dependence3 | |||||||
| No | 64 (19%) | 51 (15%) | 53 (16%) | 61 (18%) | 66 (19%) | 45 (13%) | <0.001 |
| Yes | 70 (20%) | 3 (1%) | 90 (25%) | 35 (10%) | 110 (31%) | 47 (13%) | |
| Drug Dependence4 | |||||||
| No | 89 (19%) | 50 (10%) | 97 (20%) | 75 (16%) | 103 (22%) | 63 (13%) | <0.001 |
| Yes | 45 (21%) | 4 (2%) | 46 (21%) | 21 (10%) | 72 (33%) | 29 (13%) | |
| Non-substance | |||||||
| Diagnosis5 | |||||||
| No | 62 (18%) | 37 (11%) | 71 (21%) | 54 (16%) | 70 (21%) | 44 (13%) | 0.005 |
| Yes | 71 (20%) | 17 ( 5%) | 71 (20%) | 42 (12%) | 107 (30%) | 48 (13%) | |
Those of other race/ethnicity were not included in calculating the chi-square test because of sparse data for that category.
Education: <12 years was overrepresented in the all-risky group.
Alcohol dependence was underrepresented in the never-risky group and overrepresented in the risky-to-abstinent and the resumed-risky groups.
Drug dependence was underrepresented in the never-risky group and overrepresented in the resumed-risky group.
Non-substance diagnosis includes major depressive disorder, PTSD, generalized anxiety disorder, bipolar, social phobia, specific phobia, agoraphobia, and panic disorder. These diagnoses were underrepresented in the never-risky group.
Those in the highest education group were underrepresented in the group that resumed risky drinking. Participants with alcohol or drug dependence were less likely to evidence a never-risky course and were more likely to follow a resumed-risky course, compared with participants without these lifetime disorders. Those with alcohol dependence were more likely than others to follow a risky-to-abstinent course, but this was not evident in those with drug dependence. Those with a non-substance lifetime psychiatric disorders were underrepresented in the never-risky group.
Among the “all-risky” subgroup the genders differed with respect to the age of initiation of risky drinking. Among men who followed an “all-risky” course, 85% had begun this pattern by their early 20s, while only about 60% of women with an all-risky course had done so. Among all those who ever reported a risky drinking interval, more men than women began this pattern before age 19, and more women than men began it at age 25+ (Figure 2). Among risky drinkers who became abstinent, the average age at change to abstinence was about age 36 for both genders.
Figure 2. Age at risky drinking initiation for all who reported a history of risky drinking, by gender, N males =261, N females =381.
Males > females for ages 15–19
Females > males for ages 25+
Age at risky drinking initiation was defined as the age at the first drinking interval in which women reported usually drinking 4+ drinks on any drinking day or >7 drinks per week, and in which men reported usually drinking 5+ drinks on any drinking day or >14 drinks per week.
Drink Preference
We classified the type of alcoholic drink consumed most frequently during the last drinking interval as beer only, wine only, spirits only, or “mixed”. Beer was most frequently consumed by both men and women. However, a significantly larger percentage of women reported drinking “wine only” compared with “wine only” consumption by men (8% compared to <1%; p<0.001).
Drinking Status
Drinking status by age among study participants varied substantially, according to whether or not the person reported lifetime alcohol dependence (Figures 3, 4). Those without alcohol dependence had a much higher proportion in the moderate drinker group while those with alcohol dependence had a disproportional area representing risky drinking for almost all ages.
Figure 3. Drinking status by age, all participants who reported no lifetime alcohol dependence, (N=340).
Stable abstain = no regular alcohol consumption (at least one drink a month) in the past 3 years.
Recent abstain = drank within the past 3 years but are no longer drinking at least one drink a month for at least 6 months.
Moderate drinker = regular drinking, but did not meet the criteria for risky drinking.
Risky drinking women = usually drink 4+ drinks on any drinking day or >7 drinks per week.
Risky drinking men = usually drink 5+ drinks on any drinking day or >14 drinks per week.
Figure 4. Drinking status by age for all participants who reported lifetime alcohol dependence, N=355.
Stable abstain = no regular alcohol consumption (at least one drink a month) in the past 3 years.
Recent abstain = drank within the past 3 years but are no longer drinking at least one drink a month for at least 6 months.
Moderate drinker = regular drinking, but did not meet the criteria for risky drinking.
Risky drinking women = usually drink 4+ drinks on any drinking day or >7 drinks per week.
Risky drinking men = usually drink 5+ drinks on any drinking day or >14 drinks per week.
Women reported higher rates of abstinence at all ages compared with men, while men reported higher rates of risky drinking than women, particularly when they were under age 30 (Figures 5, 6). Men over age 35 were more likely than older women to be moderate drinkers, but rates of risky drinking were fairly similar between the genders. Risky drinking was less prevalent among female DWI offenders in their 20s than it was among males, with rates of about 55% compared to 75% in men, and overall abstinence rates were substantially higher among women. Among men, shifts from risky drinking to abstinence or moderate drinking began to occur by their mid-20s. Risky drinking rates fell from about 75% in their 20s to about 45% by age 35, and stabilized until male participants were in their 50’s, after which rates of moderate drinking increased and risky drinking decreased to about 20%.
Figure 5. Drinking status by age, females, N=413.
Stable abstain = no regular alcohol consumption (at least one drink a month) in the past 3 years.
Recent abstain = drank within the past 3 years but are no longer drinking at least one drink a month for at least 6 months.
Risky drinking women = usually drink 4+ drinks on any drinking day or >7 drinks per week.
Figure 6. Drinking status by age, males, N=283.
Stable abstain = no regular alcohol consumption (at least one drink a month) in the past 3 years.
Recent abstain = drank within the past 3 years but are no longer drinking at least one drink a month for at least 6 months.
Moderate drinker = regular drinking, but did not meet the criteria for risky drinking.
Risky drinking men = usually drink 5+ drinks on any drinking day or >14 drinks per week.
Abstinence rates among men leveled off after age 35 and appeared to decrease when they reached their late 40s. Reported rates of risky drinking among women tended to be relatively stable in their 20s and 30s, and started to decrease in their 40s and 50s. Unlike men, abstinence rates increased among women as the population aged.
Quitting Risky Drinking
Seventy-three percent of those who initiated risky drinking before age 15 became alcohol dependent, compared with 57% for ages 15–19; 50% for ages 20–24; and 46% for ages 25+. Tests for interactions (effect modification) were not significant. Hazard ratios for stopping drinking showed that among those who began risky drinking, the rate of stopping risky drinking for those who initiated it at 15–19 years old was twice that for those who initiated it at ages <15 (Table 4). There were no differences by gender or psychiatric diagnoses.
Table 4.
Multivariate cox regression predicting time to stopping risky drinking
| Variable | HR (95% CI) | p-value |
|---|---|---|
| Age initiating risky drinking <15 15–19 20–24 25+ |
1.0 2.0 (1.3, 3.2) 2.3 (1.5, 3.7) 2.7 (1.6, 4.4) |
0.001 |
| Gender Female Male |
1.0 0.9 (0.7, 1.1) |
0.17 |
| Alcohol Dependence No Yes |
1.0 1.0 (0.8, 1.2) |
0.88 |
| Drug Dependence Yes No |
1.0 1.0 (0.8, 1.3) |
0.81 |
| Non-Substance diagnosis1 Yes No |
1.0 0.9 (0.7, 1.1) |
0.30 |
Non-substance diagnosis includes major depressive disorder, PTSD, generalized anxiety disorder, bipolar, social phobia, specific phobia, agoraphobia, and panic disorder. These diagnoses were underrepresented in the never-risky group.
Discussion
This unique study is the first to our knowledge to examine lifetime drinking course among a large sample of DWI offenders. Convicted DWI offenders, particularly those with lifetime alcohol or drug dependence, remained at high risk for alcohol-related problems throughout their lives. Almost half (44%) either were risky drinkers during all their drinking intervals or became moderate drinkers, but at some point “relapsed” to risky drinking (Table 3). Low levels of education, but not gender or ethnicity were associated with continuing a pattern of risky drinking throughout the lifespan up to age 60. Lifetime trajectories of drinking course by age differed dramatically according to whether or not the participant reported lifetime alcohol dependence (Table 3). Indeed as McClellan points out, alcohol dependence is a chronic, relapsing disorder [12], but with an increased emphasis on diagnosis and treatment of this population, these statistics could be improved. A previous study of this DWI offender population revealed that a history of hospitalization for alcohol or drug use problems, but not having received outpatient therapy for alcohol or drug use disorders, was associated with abstinence from alcohol in the past 3 months.[27] Of 23 participants who were hospitalized for substance use disorders, 13 (57%) had achieved abstinence.
Those with non-substance-related lifetime psychiatric disorders also differed with respect to drinking course compared with other offenders, having fewer individuals who never were risky drinkers and more individuals who resumed risky drinking. Another study of this population found that having a lifetime non-substance-related psychiatric disorder did not impart increased long-term risk of driving after drinking in this first offender population. In fact, offenders with a lifetime diagnosis of PTSD were more likely than those without these disorders to report alcohol abstinence at the 15-year follow-up.[27]
When comparing lifetime drinking trajectories by gender we found several important differences. First, although the preferred drink of both genders was beer, more women drank wine as their preferred drink than men. This preference difference is found among Americans in general.[28] According to the Gallup Poll Social Series, beer is the preferred alcoholic beverage by men (54%) but only 27% of women preferred it to wine or liquor.[29]
A second difference is that among men the proportion of risky drinkers decreased rapidly until about age 35 then declined more slowly, while among women there was a gradual reduction in risky drinking with age. The CLDH was used to characterize lifetime drinking trajectories in a case-control study of risk factors for cardiovascular disease in Western New York State (WNYS).[28] Two distinct trajectories were observed, an early peak trajectory characterized by heavy alcohol use in late adolescence and early adulthood followed by a dramatic decrease in drinking during middle age, and a stable trajectory characterized by moderate alcohol use that tended to persist throughout middle age. Drinkers having early peak lifetime trajectories generally began drinking earlier than stable drinkers, and they drank fewer years, less frequently, and consumed lower volumes of alcohol over their lifetimes. This description of early peak drinking is consistent with that of risky drinking reported by DWI offenders in this study, but there was no dramatic decrease in risky drinking in our sample.
A third gender difference is that while female DWI offenders in the present study reported engaging in risky drinking at the same rates as men, female offenders generally were older than men when they initiated risky drinking. Female early peak drinkers in the WNYS study also started drinking at older ages than males.[28] Hazard ratios show that the rate of quitting risky drinking among those who initiated risky drinking at age 15 or later is double the rate for those who initiated risky drinking before age 15. After controlling for age at which offenders began risky drinking, we found that neither gender nor psychiatric diagnosis was significant in predicting time to stopping risky drinking.
While the findings of other studies on this topic are not comparable for a number of reasons, including the study methodologies and differences among the populations sampled, it may be useful to examine results in this population in relation to other studies that investigated alcohol use and gender differences over the lifespan. A longitudinal study of drinking patterns among a representative sample of U.S. African American, white, and Hispanic adults, found that from 1984 to 1992, there were ethnic and gender differences in heavy drinking and rates of abstention,[9] whereas drinking course in this study among DWI offenders did not vary significantly by these factors. Delucchi & Kaskutas report changes in alcohol consumption over 11 years in problem drinkers identified through random digit dialing in California. Similar to the present study, drinking quantity and frequency generally declined over an 11-year period, but unlike our study abstention rates never exceeded 10%.[10] Vaillant led a landmark study that followed over 600 individuals for 40 years.[30,31] That sample, however, was limited to men and consisted of college students and working class men.
Study limitations include low participation rates in the follow-up study which limit the generalizability of findings to all DWI offenders. Another potential limitation of findings based on the CLDH concerns the validity of retrospective measures of lifetime drinking patterns. Prospective ascertainment of alcohol intake poses fewer problems concerning memory than retrospective ascertainment. However, despite these positive attributes of prospective studies, there are also drawbacks, particularly in the case of DWI offenders who may be reluctant to acknowledge patterns of heavy drinking. The CLDH uses a number of cognitive techniques to maximize memory and has demonstrated high reliability in test-retest studies.[26] Retrospective assessments of drinking habits may provide several advantages. For example, diet recall validity studies have reported high correlations between alcohol intake reported in the distant past and recalled after many years. [32,33] Moreover, several studies have found that heavy drinkers report higher alcohol intakes retrospectively than prospectively [34–36], which suggests that people may be more comfortable reporting past heavy drinking than current heavy drinking.
In conclusion, risky drinking was prevalent at all ages for both genders. Drinking course was adversely affected by young age at risky drinking initiation and the presence of substance-use and other psychiatric disorders. Across time, those who began risky drinking at age 15 or later quit at double the rate of those who began before age 15. While male and female DWI offenders’ drinking courses were similar, their drinking status differed by age. Women began risky drinking later in life and were more likely to achieve abstinence. The high rate of lifetime risky drinking in this population suggests that DWI offenders should be a high priority group for receiving intensive treatment services.
Acknowledgments
We thank our research participants for their generous assistance and cooperation. We thank Jan Alroy, Catherine Cummins, Vivian Fernandez, and Michael Lackey for conducting the interviews, and Patricia Hokanson for manuscript preparation.
The authors disclose receipt of the following financial support for the research and/or authorship of this article. This study was funded by a grant from the National Institute on Alcohol Abuse and Alcoholism (NIAAA R01 AA014750). The methods, observations, and interpretations put forth in this article do not necessarily represent those of the funding agency.
Footnotes
Declaration of Interest
The authors declare no conflicts of interest with respect to the authorship and/or publication of this article.
Reference List
- 1.Hedlund JH, McCartt AT. Drunk driving: Seeking additional solutions. Trumbull, CT: Preusser Research Group; 2002. pp. 18–34. [Google Scholar]
- 2.Lapham SC, Skipper BJ, Simpson GL. A prospective study of the utility of standardized instruments in predicting recidivism among first DWI offenders. J Stud Alcohol. 1997;58(5):524–530. doi: 10.15288/jsa.1997.58.524. [DOI] [PubMed] [Google Scholar]
- 3.Current Trends: Drivers with repeat convictions or arrests for driving while impaired - United States. Morb Mortal Wkly Rep. 1994;43(41):759–761. [PubMed] [Google Scholar]
- 4.Bakker LW, Hudson SM, Ward T. Reducing recidivism in driving while disqualified. Crim Justice Behav. 2000;27(5):531–560. [Google Scholar]
- 5.Lapham SC, Smith E, C'de Baca J, Chang I, Skipper BJ, Baum G, et al. Prevalence of psychiatric disorders among persons convicted of driving while impaired. Arch Gen Psychiatry. 2001;58(10):943–949. doi: 10.1001/archpsyc.58.10.943. [DOI] [PubMed] [Google Scholar]
- 6.Shaffer HJ, Nelson SE, LaPlante DA, LaBrie RA, Albanese M, Caro G. The epidemiology of psychiatric disorders among repeat DUI offenders accepting a treatment-sentencing option. J Consult Clin Psychol. 2007;75(5):795–804. doi: 10.1037/0022-006X.75.5.795. [DOI] [PubMed] [Google Scholar]
- 7.Palmer RS, Ball SA, Rounsaville BJ, O'Malley SS. Concurrent and predictive validity of drug use and psychiatric diagnosis among first-time DWI offenders. Alcohol Clin Exp Res. 2007;31(4):619–624. doi: 10.1111/j.1530-0277.2007.00346.x. [DOI] [PubMed] [Google Scholar]
- 8.Vaillant GE. A long-term follow-up of male alcohol abuse. Arch Gen Psychiatry. 1996;53(3):243–249. doi: 10.1001/archpsyc.1996.01830030065010. [DOI] [PubMed] [Google Scholar]
- 9.Caetano R, Kaskutas LA. Changes in drinking patterns among Whites, Blacks and Hispanics; 1984–1992. J Stud Alcohol. 1995;56(5):558–565. doi: 10.15288/jsa.1995.56.558. [DOI] [PubMed] [Google Scholar]
- 10.Delucchi KL, Kaskutas LA. Following problem drinkers over 11 years: Understanding changes in alcohol consumption. J Stud Alcohol Drugs. 2010;71(6):831–836. doi: 10.15288/jsad.2010.71.831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lapham SC, Stout R, Laxton G, Skipper BJ. Persistence of Addictive Disorders in a First-Offender Driving While Impaired Population. Arch Gen Psychiatry. 2011;68(11):1151–1157. doi: 10.1001/archgenpsychiatry.2011.78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.McLellan TA, Lewis D, O'Brien CP, Kleber HD. Is drug dependence a chronic medical illness: Implications for treatment, insurance and outcome evaluation. JAMA. 2000;284(13):1689–1695. doi: 10.1001/jama.284.13.1689. [DOI] [PubMed] [Google Scholar]
- 13.Weisner C, Matzger H, Kaskutas LA. How important is treatment? One-year outcomes of treated and untreated alcohol-dependent individuals. Addiction. 2003;98(7):901–911. doi: 10.1046/j.1360-0443.2003.00438.x. [DOI] [PubMed] [Google Scholar]
- 14.Peller AJ, Najavits LM, Nelson SE, LaBrie RA, Shaffer HJ. PTSD among a treatment sample of repeat DUI offenders. J Trauma Stress. 2010;23(4):468–473. doi: 10.1002/jts.20550. [DOI] [PubMed] [Google Scholar]
- 15.Wells-Parker E, Williams M. Enhancing the effectiveness of traditional interventions with drinking drivers by adding brief individual intervention components. J Stud Alcohol. 2002;63(6):655–664. doi: 10.15288/jsa.2002.63.655. [DOI] [PubMed] [Google Scholar]
- 16.Woodall WG, Kunitz SJ, Zhao H, Wheeler DR, Westerberg V, Davis J. The prevention paradox, traffic safety, and driving-while-intoxicated treatment. Am J Prev Med. 2004;27(2):106–111. doi: 10.1016/j.amepre.2004.04.012. [DOI] [PubMed] [Google Scholar]
- 17.Woodall WG, Delaney HD, Kunitz SJ, Westerberg VS, Zhao H. A randomized trial of a DWI intervention program for first offenders: intervention outcomes and interactions with antisocial personality disorder among a primarily American-Indian sample. Alcohol Clin Exp Res. 2007;31(6):974–987. doi: 10.1111/j.1530-0277.2007.00380.x. [DOI] [PubMed] [Google Scholar]
- 18.Lapham SC, Skipper BJ, Owen JP, Kleyboecker K, Teaf D, Thompson B, et al. Alcohol abuse screening instruments: Normative test data collected from a first DWI offender screening program. J Stud Alcohol. 1995;56(1):51–59. doi: 10.15288/jsa.1995.56.51. [DOI] [PubMed] [Google Scholar]
- 19.The Century Council. Looking back, moving forward. Los Angeles, CA: The Century Council; 1997. [Google Scholar]
- 20.Windle M, Miller BA. Alcoholism and depressive symptomatology among convicted DWI men and women. J Stud Alcohol. 1989;50(5):406–413. doi: 10.15288/jsa.1989.50.406. [DOI] [PubMed] [Google Scholar]
- 21.Perrine MW, Peck RC, Fell JC. Epidemiologic perspectives on drunk driving. Rockville, MD: Department of Health and Human Services, Public Health Service, Office of the Surgeon General; 1989. [Google Scholar]
- 22.Donovan DM, Marlatt GA. Personality subtypes among driving-while-intoxicated offenders: Relationship to drinking behavior and driving risk. J Consult Clin Psychol. 1982;50(2):241–249. doi: 10.1037//0022-006x.50.2.241. [DOI] [PubMed] [Google Scholar]
- 23.Robins LN, Helzer JE, Ratcliff KS, Seyfried W. Validity of the Diagnostic Interview Schedule, Version II: DSM-III diagnoses. Psychol Med. 1982;12(4):855–870. doi: 10.1017/s0033291700049151. [DOI] [PubMed] [Google Scholar]
- 24.Robins LN, Wing J, Wittchen HU, Helzer JE, Babor TF, Burke J, et al. The Composite International Diagnostic Interview. An epidemiologic Instrument suitable for use in conjunction with different diagnostic systems and in different cultures. Arch Gen Psychiatry. 1988;45(12):1069–1077. doi: 10.1001/archpsyc.1988.01800360017003. [DOI] [PubMed] [Google Scholar]
- 25.Wittchen HU. Reliability and validity studies of the WHO-Composite International Diagnostic Interview (CIDI): A critical review. J Psychiatr Res. 1994;28(1):57–84. doi: 10.1016/0022-3956(94)90036-1. [DOI] [PubMed] [Google Scholar]
- 26.Russell M, Marshall JR, Trevisan M, Freudenheim JL, Chan AW, Markovic N, et al. Test-retest reliability of the cognitive lifetime drinking history. Am J Epidemiol. 1997;146(11):975–981. doi: 10.1093/oxfordjournals.aje.a009225. [DOI] [PubMed] [Google Scholar]
- 27.Lapham SC, Skipper B. Do psychiatric disorders affect long-term drinking and driving outcomes among convicted impaired drivers. J Consult Clin Psychol. 2012 In review. [Google Scholar]
- 28.Fan AZ, Russell M, Stranges S, Dorn J, Trevisan M. Association of lifetime alcohol drinking trajectories with cardiometabolic risk. J Clin Endocrinol Metab. 2008;93(1):154–161. doi: 10.1210/jc.2007-1395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Newport F. [cited 2011 Jan 19];US Drinking Rate Edges Up Slightly to 25-Year High. 2011 Available from: URL: http://www.gallup.com/poll/141656/Drinking-Rate-Edges-Slightly-Year-High.aspx. (Archived by WebCite® at http://webcitation.org/61qN5mrYZ)
- 30.Vaillant GE. The natural history of alcoholism, revisited. Cambridge, MA: Harvard University Press; 1995. [Google Scholar]
- 31.Vaillant GE. The natural history of alcoholism. Cambridge, MA: Harvard University Press; 1983. [Google Scholar]
- 32.Dwyer JT, Gardner J, Halvorsen K, Krall EA, Cohen A, Valadian I. Memory of food intake in the distant past. Am J Epidemiol. 1989;130(5):1033–1046. doi: 10.1093/oxfordjournals.aje.a115404. [DOI] [PubMed] [Google Scholar]
- 33.Thompson FE, Lamphiear DE, Metzner HL, Hawthorne VM, Oh MS. Reproducibility of reports of frequency of food use in the Tecumseh Diet Methodology Study. Am J Epidemiol. 1987;125(4):658–671. doi: 10.1093/oxfordjournals.aje.a114579. [DOI] [PubMed] [Google Scholar]
- 34.Czarnecki DM, Russell M, Cooper ML, Salter D. Five-year reliability of self-reported alcohol consumption. J Stud Alcohol. 1990;51(1):68–76. doi: 10.15288/jsa.1990.51.68. [DOI] [PubMed] [Google Scholar]
- 35.Ernhart CB, Morrow-Tlucak M, Sokol RJ, Martier S. Underreporting of alcohol use in pregnancy. Alcohol Clin Exp Res. 1988;12(4):506–511. doi: 10.1111/j.1530-0277.1988.tb00233.x. [DOI] [PubMed] [Google Scholar]
- 36.Simpura J, Poikolainen K. Accuracy of retrospective measurement of individual alcohol consumption in men; a reinterview after 18 years. J Stud Alcohol. 1983;44(5):911–917. doi: 10.15288/jsa.1983.44.911. [DOI] [PubMed] [Google Scholar]