Family predictors of severe mental disorders and criminality in the Northern Finland 1966 Birth Cohort

Abstract Early family characteristics may influence the later development of severe mental disorders and criminality of a child. The association between an adverse family environment during childhood and its later consequences in adulthood, however, are still widely open. The aim of the present study was to analyse in a longitudinal perspective, family risks of severe hospital-treated mental disorders and criminal behaviour in the Northern Finland 1966 Birth Cohort and to develop a descriptive life span model of schizophrenia. A large, general population birth cohort (N =11 017), the Northern Finland 1966 Birth Cohort was used as a study population. This database provides the information of prospectively collected data on both biological and social aspects of pregnancy, the characteristics of family, the mother, the father, and the child. The information of psychiatric outcomes was gathered from the Finnish Hospital Discharge Register (FHDR) and the data on registered criminal behaviour of the cohort members come from computerized files maintained by the Ministry of Justice. Children born to multiparous mothers (GMP) i.e. those that had undergone at least six deliveries were more commonly treated in mental hospitals later in life (4.5% vs. 3.4%; p=0.028) than children born to mothers that have fewer children. Of the diagnostic groups, the risk of psychoses other than schizophrenia (OR 2.3; 95% CI 1.2-4.7), and depressive disorders (OR 2.2; 1.0-4.5) was elevated among adult children of those mothers. Birth order was associated with adult schizophrenia. The risk was elevated among male firstborns (ratio 1.5; 95% CI 1.0-2.2), but it was lower than expected among male lastborns (ratio 0.7; 95% CI 0.5-0.9). The elevated risk was not significantly associated with female schizophrenia patients. On the contrary, the risk was lower than expected among females who were not first, not last or not only children in the family (ratio 0.6; 95% CI 0.3-0.9). Among males the risk for violent crimes later in life was elevated among the only children (OR 1.8; 95% CI 1.1-3.0). If perinatal risk was additional exposure, the risk increased up to 4-fold (OR 4.4; 95% CI 1.9-10.8). Combining with maternal risks increased the risk up to 6-fold (OR 5.9; 95% CI 3.1-11.3) and with paternal risk up to 8-fold (OR 8.4; 95% CI 3.9-18.1), respectively. Among females the absence of the father during childhood until the age of 14 was the strongest risk factor in predicting later criminality (OR 2.5; 95% CI 1.4-4.3). Further, in the families, where the father was present, maternal smoking during pregnancy together with being born unwanted increased the prevalence for criminal offending significantly up to 7.2%. In conclusion, some characteristics of the early childhood family environment were associated with mental disorders and criminality in adulthood and form part of the developmental trajectory of these disorders. Early detection of such children at risk is important in preventing mental disorders and criminality in adulthood