This study aimed to describe changes in Swedish alcohol habits during a period of "harmonization" with European Union alcohol policy. Three random samples collected during the years 1997 (n=997), 2001 (n=893) and 2005 (n=914) were compared. Alcohol habits and hazardous consumption was measured with the AUDIT (Alcohol Use Disorders Identification Test) distributed to the respondents by post. Results were somewhat diverse, but significant changes in alcohol habits occurred among two subgroups of the population: for women and the age group between 28 and 60 years, AUDIT scores peaked in 2001. The results are discussed in relation to the changes made in Swedish alcohol policy during the investigation period.
OBJECTIVE: To investigate the prevalence of alcohol problems in a representative sample of Swedish drivers suspected of drunk driving in comparison with control drivers and the general Swedish population in relation to mode and time of detection. Is the time of day or night or the detection mode important for the prevalence of alcohol problems and which are the best predictors for identifying alcohol problems among DUI offenders? METHODS: Two thousand and one hundred drivers (169 females) suspected of DUI offence during 1997-2001 who agreed to respond to the AUDIT questionnaire (Alcohol Use Disorders Identification Test) and 785 control drivers (266 females) not suspected of DUI recruited at general traffic controls were investigated. RESULTS: Both mode and time of detection were found to be important. The greatest impact on the prevalence of alcohol problems emanated from the predictors in the following order: high BAC; unlicensed driving; detection hours between 12.00 and 18.00, and age under 26 years. Age over 55 years and detection in general traffic controls were the two strongest factors negatively correlated to alcohol problems prevalence. The differences between regions with regard to alcohol problems incidence could only partly be explained by police routines and resources. CONCLUSIONS: Mode and time of detection affect the prevalence of DUI offenders with alcohol problems, and to a lesser degree also BAC level. Both the detection mode and the time of detection are significant for the proportion of identified DUIs with alcohol problems. Because the majority of committed DUI offenses are never identified, it is important to optimize the detection strategies of the police with the purpose of minimizing public damage and expenses.
The prevalence and types of crime offences, as well as predictors of relapse, among drivers suspected of driving under influence (DUI) were investigated. A total of 1830 Swedish DUI drivers responded to the Alcohol Use Disorders Identification Test--AUDIT. Information about previous DUI offences, other traffic offences along with other types of criminal offences was taken from a crime register. A total criminality (including all traffic offences) of 64% in the period of five years before investigation was analyzed. 40% of the sample had other criminality besides traffic violations during that period. 14.3% of the drivers relapsed to DUI in the two-year period after the investigation. In terms of DUI relapse, the following factors were the main predictors: previous traffic violations, previous DUI offences, previous other criminality (frauds or other acts of dishonesty) and detection hours between 12.00 and 19.00. Detection in general traffic controls and high BAC (blood alcohol concentration) when detected were the strongest factors with negative correlation to DUI re-offence.
Among clients who have been screened already for drug-related problems, the Drug Use Disorders Identification Test--Extended (DUDIT-E) maps the frequency of illicit drug use (D), the positive (P) and negative (N) aspects of drug use, and treatment readiness (T). D scores correlated with Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnoses among 154 heavy drug users from criminal justice and drug detoxification settings, as well as with urine test results in drug detoxification units. One-week test/retest intraclass correlation coefficients among 92 male prison inmates were .90, .78, .75, and .84 for D, P, N, and T scores, respectively. Cronbach's alpha were .88-.95 for P score, .88-.93 for N score, and .72-.81 for T score. Principal components analysis supported construct validity for P, N, and T scores. T scores were higher in prison treatment units than in motivational and regular units without treatment emphasis. Motivational index scores differentiated between three categories of heavy drug users; they did not differentiate between prisons and unit types, but this corresponded to unclear structural differentiation between units.