Intended for healthcare professionals


Mental disorders in prison populations aged 15-21: national register study of two cohorts in Finland

BMJ 2005; 330 doi: (Published 09 June 2005) Cite this as: BMJ 2005;330:1364
  1. Eila S Sailas (eila.sailas{at}, medical doctor1,
  2. Benjamin Feodoroff, bachelor of medicine1,
  3. Matti Virkkunen, professor3,
  4. Kristian Wahlbeck, professor2
  1. 1 STAKES, National Research and Development Centre for Welfare and Health, PO Box 220, 00531 Helsinki, Finland,
  2. 2 Vaasa Central Hospital, Psychiatric Unit, 65130 Vaasa, Finland,
  3. 3 Helsinki University Central Hospital, PO Box 320, 00029 HUCH, Finland
  1. Correspondence to: E S Sailas
  • Accepted 9 February 2005


Juvenile delinquency is linked to psychiatric morbidity.1 w1-w9We were interested in temporal changes in psychiatric morbidity in offenders. Criminal policy in many Western countries emphasises the need for alternative methods of punishment for adolescent prisoners.2 In Finland, successful policy to diminish the number of adolescents in prisons has been internationally recognised.3 We studied the changes in psychiatric hospitalisations in Finnish prisoners aged 15 to 21 to see whether a selection process occurs as the number of young prisoners decreases.

Participants, methods, and results

We linked the unique personal identification numbers of all prisoners aged 15 to 21 years from 1984-5 and 1994-5 to the Finnish healthcare register, which includes data of all hospitalisations in Finland. We retrieved occurrences of depression, psychosis, personality disorder, and substance dependence, and we analysed temporal changes. The observation period for the first cohort was 1980-9 and for the second 1990-9. We compared hospitalisations with those of a control group, matched for sex, age, and place of birth, derived from the population register. The full method is on

The earlier cohort from 1984-5 comprised 656 prisoners (719, with 63 (8.8%) missing identification numbers), and the later cohort included 370 prisoners (377, with 7 (1.9%) missing identification numbers). This temporal decrease shows the effect of the policy to reduce the number of young prisoners. The cohorts did not differ in terms of sex (Fisher's exact test, P = 0.19); in the earlier cohort there were 18 (0.03%) women, in the later 5 (0.01%).

The number of inmates with at least one hospital treatment for any mental disorder increased significantly (odds ratio 1.8, 95% confidence interval 1.3 to 2.3, age adjusted) over time compared with the general population, in which we detected no increase in hospitalisations for mental disorders (1.0, 0.7 to 1.4; 1). The increase in treatment for psychosis was significant between the two cohorts (2.7, 1.4 to 5.1, age adjusted) and it was significant for substance dependence (3.0, 2.0 to 4.6, age adjusted). In the control groups the changes in prevalence of these disorders were not significant (1.6, 0.7 to 3.5 and 0.9, 0.3 to 3.0).

Table 1

Young prisoners and controls with psychiatric hospital treatments, Finland, 1984-5 and 1994-5

View this table:
  • What is already known on this topic

  • Prisoners are more likely to have serious mental disorders than the general population, and efforts have been made to reduce the number of young prisoners

  • What this study adds

  • More mentally ill young people end up in prison as the prison population diminishes

The absolute number of prisoners who had had inpatient treatment for psychosis or substance dependence increased from the earlier to the later cohort. The odds of being hospitalised for schizophrenia in the earlier prisoner group were fourfold greater (3.9, 1.6 to 9.4) than in the control group but were eightfold greater (8.0, 2.7 to 23.5) in the later cohort.


Relatively more mentally ill people end up in prisons as the prison population diminishes. Hospitalisations of young prisoners for mental disorders increased during the 10 year period separating the two study cohorts. We did not detect a similar increase in the matched general population sample. Young prisoners had considerably more inpatient treatments for mental disorders than did the controls. The most alarming finding is the increase of psychoses in the more recent group. This cannot be explained by changes in the population morbidity, as the number of hospitalisations for psychotic disorders in the general population control group decreased, as seen before.4

Our findings reflect the failure of healthcare systems and emphasise the necessity for early screening of mental disorders in delinquents.

Embedded ImageThe full method and references w1-w9 are on

This article was posted on on 13 April 2005:


  • Contributors ESS conceived the study. ESS and BF collected the data. ESS, BF, and KW analysed the data and drafted the paper. KW supervised and coordinated the research. MV discussed the data and supervised the research. ESS is guarantor.

  • Funding Supported by a grant from Finska Läkaresällskapet and done as a part of the Merttu-project of the Finnish Academy of Finland (Grant 105218).

  • Competing interests None declared.

  • Ethical approval Criminal Sanctions Agency, Ministry of Social Affairs and Health, and the department of psychiatry, Helsinki University Central Hospital.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
View Abstract