The study protocols were approved by the Research Ethics Committees of Kuopio University/Kuopio University Hospital and Turku University/Turku University Hospital (studies I–III), or the Research Ethics Committee of Kuopio University/Kuopio University Hospital (studies IV–V). All participants gave written informed consent to participate in the studies. The diagnoses of the index subjects were made according to DSM-III-R or DSM-IV (American Psychiatric Association, 1987, or 1994). All index and control subjects were white males of Finnish origin. The exact numbers of the index subjects from the same samples in studies I–III or IV–V show a slight variation between the studies, depending on whether the particular genotype studied was obtained in the genotype analysis.
4.1.1 Type 1 alcoholic subjects
The type 1 alcoholic population (total n=123 in studies I–II, and n=114 in study III) consisted of two independent Finnish samples of male alcoholics from the regions of Turku (n=67–65) and Kuopio (n=56–49). The sample in Turku consisted of alcoholics entering a detoxification program in that area, and patients in Kuopio were recruited with the help of a local rehabilitation center for alcoholics, where they had obtained treatment for their alcoholism. The mean age (±SD) of these subjects was 44.1±8.8 years (studies I–II) or 43.8±8.8 (study III). Inclusion criteria were serious alcohol-related problems (alcohol abuse or dependence resulting in a failure to fulfil obligations at work or in recurrent social problems) starting after the age of 25 years. The onset of alcohol-related problems was determined by interviewing the subjects.
All these type 1 patients fulfilled the Diagnostic and Statistical Manual of Mental Disorders’ (third edition, revised, DSM-III-R; American Psychiatric Association, 1987) criteria for alcohol abuse or dependence, and underwent a clinical examination and a self-administered Michigan Alcoholism Screening
Test (MAST, Selzer, 1971). Exclusion criteria were major mental disorders such as schizophrenia, schizophreniform and schizoaffective disorders, mood disorders with psychotic features, organic mental syndromes and disorders, and paranoid and other psychoses (screened with the Hopkins Symptoms Checklist 90 and a clinical interview by a physician). Subjects with a history of violence or severe antisocial behavior, or severe physical illness, were also excluded. The crucial differential diagnostic factors for the alcoholic subject to be included in the type 1 alcoholic group were the onset of drinking problems after the age of 25 years and the lack of a violent antisocial history, according to Cloninger’s model.
4.1.2 Type 2 alcoholic subjects
The type 2 alcoholics (n=62 in study II, and n=51 in study III) were committed for forensic psychiatric examination in a state mental hospital (Niuvanniemi Hospital, Kuopio) after committing an impulsive violent offense (homicide, attempted homicide, aggravated violent assault, assault, sexual offense, or arson). In Finland, most persons charged with serious violent offenses are committed for forensic psychiatric evaluation: 70% of all homicide offenders, for example, are evaluated (Eronen et al., 1996), and most of the offenders who are considered very violent or dangerous are evaluated in Niuvanniemi Hospital regardless of their residence. Therefore, the offenders included in this study were representative of habitually violent offenders in the Finnish male population. Based on the data gathered, all were recidivist offenders, although just under 10% of them had not been convicted in court before the evaluation.
Forty percent of the subjects had committed at least one homicide or an attempted homicide and aggravated violent assaults, and 25% had committed at least 2 homicides or attempted homicides. The mean age (±SD) of these subjects was 30.4± 8.2 years in study II and 30.1±8.4 years in study III.
All type 2 alcoholics were subjected to an extensive forensic psychiatric examination including a psychiatric evaluation, standardized psychological tests, evaluation of physical condition with laboratory tests,
electroencephalography (EEG), magnetic resonance imaging (MRI), and staff observation in a security ward for 4–8 weeks. Inclusion criteria were serious alcohol-related problems before the age of 25 years and the co-morbid diagnosis of ASP disorder fulfilling DSM-IV (American Psychiatric Association, 1994) criteria. Abuse or dependence had resulted in recurrent social problems and recurrent substance-related legal problems, fulfilling DSM-III-R criteria for alcohol abuse or dependence (American Psychiatric Association, 1987). The onset of alcohol-related problems was determined within a 1-year accuracy on the basis of data gathered from various sources during the forensic psychiatric evaluation. Exclusion criteria were major mental disorders such as schizophrenia, schizophreniform, and schizoaffective disorders, mood disorders with psychotic features, organic mental syndromes and disorders, and paranoid and other psychoses or severe physical illness.
4.1.3 Controls (studies I–III)
Previously published data on COMT genotypes of 3140 blood donors (Syvänen et al., 1997) were used as a model of the Finnish general population. There were also 267 unrelated controls (mean age±SD=54.6±6.92 years) from the Kuopio Ischemic Heart Disease Risk Factor Study (KIHD) (Salonen, 1988) reporting low or moderate alcohol use (self-reported avarage alcohol intake = 1–7 drinks/week and MCV<101 FI, GGT<80 u/l). (For more details about the KIHD subjects, see below.) The controls in study III were 54 unrelated healthy males from the Turku and Kuopio regions in Finland (mean age±SD=44.1±7.9 years).
4.1.4. The representative sample of socially drinking males (studies IV–V) The Kuopio Ischemic Heart Disease Risk Factor Study (KIHD) is a population- based epidemiological study launched in the 1980s to investigate previously unestablished risk factors for myocardial infarction, progression of atherosclerosis, and other major health outcomes in middle-aged men (Salonen, 1988; Lakka et al., 1994; Lynch et al., 1994). The total sample of the
KIHD study consists of 2,682 men who were recruited in two cohorts. The present study is based on the second cohort, an age-stratified sample of 42-, 48-, 54-, and 60-year-old men (n=1516, participation rate 82.6%) enrolled in the study between 1984 and 1989. Men who had reported no use of alcohol for at least 12 months were defined as abstainers (n=123, 11% in study IV; and n=100, 10% in study V). Since abstainers are a fairly heterogenous group consisting of lifetime non-drinkers as well as those who have quit because of health problems or other reasons, they were excluded from final analyses. The proportion of abstainers in this sample is similar to the 10% oserved in the general population in Finland (Nylander et al., 2007). From the remaining subjects, both sociodemographic and genotype data (COMT in study IV and DRD2 in study V) were available for 896 subjects and 884 subjects, respectively.
The study population was a random sample of men living in the city of Kuopio, or neighboring rural areas. The mean age (±SD) of the volunteers was 51.9 (±6.73) years in study IV, and 56.1 (±6.7) years in study V. A variety of sociodemographic, behavioral and medical assessments were done according to the KIHD protocol as described earlier (Lakka et al., 1994; Lynch et al., 1994). Age, place of residence (urban/rural), marital status, educational level (represented by a 7-point scale from less than elementary=1 to academic degree=7), current income, history of smoking in cigarette-years, and history of physician-diagnosed chronic diseases (ischemic heart disease, diabetes, stroke, cancer, diseases of the liver or pancreas, mental disorder) and severe trauma were recorded in the questionnaire and double-checked by the research staff in the interview.
A self-report quantity-frequency questionnaire derived from the Scandinavian Drinking Survey (Hauge and Irgens-Jensen, 1981) was used to record the level of alcohol use. The average weekly consumption of alcohol in pure ethanol (g/week) was calculated based on the known alcoholic content of each beverage type and the reported doses and frequencies of drinking sessions (Kauhanen et al., 1997). Serum gamma-glutamyltranspeptidase (GGT) and
mean corpuscular volume (MCV) was determined from baseline blood samples as biomarkers of excessive alcohol use. These biochemical measures were checked to validate the self-report data. The correlation of self-reported alcohol consumption with the GGT and MCV measures separately at various levels of alcohol consumption has been examined earlier. There was no indication of differential misclassification due to erroneous reporting (Kauhanen et al., 1992).