Schizophrenia

Schizophrenia, Alcohol Abuse, and Violent Behavior: A 26-Year Followup Study of an

Unselected Birth Cohort by Pirkko Rasanen, Jari Tiihonen, Matti Isohanni, Paula Rantakallio,

Jari Lehtonen, and Juha Moving

Abstract

It has been suggested that schizophrenia and alco- holism are associated with violent behavior. But so far there are no published studies from unselected cohorts quantifying the actual risk associated with schizophre- nia both with and without comorbid alcoholism. In this study, an unselected birth cohort (n = 11,017) was prospectively followed to the age of 26, and data on psychiatric disorders and crimes were collected from national registers. The odds ratios for violent offenses and recidivism were calculated for each diagnostic group. Men who abused alcohol and were diagnosed with schizophrenia were 25.2 (95% confidence interval (CI) 6.1-97.5) times more likely to commit violent crimes than mentally healthy men. The risk for nonal- coholic patients with schizophrenia was 3.6 (95% CI 0.9-12.3) and for other psychoses, 7.7 (95% CI 2.2-23.9). None of the patients with schizophrenia who did not abuse alcohol were recidivists (>2 offenses), but the risk for committing more crimes among alco- holic subjects with schizophrenia was 9.5-fold (95% CI 2.7-30.0). This study suggests that to prevent the crimes being committed by people with schizophrenia, it is important that clinicians watch for comorbid alco- hol abuse.

Key words: Alcohol abuse, crime, violence. Schizophrenia Bulletin, 24(3):437-441,1998.

Many recent studies suggest an association between schizophrenia and violence (Lindqvist and Allebeck 1990; Cirincione et al. 1992; Hodgins 1992). There is also clear evidence for a relationship between alcohol abuse and violent behavior (Tiihonen et al. 1993; Eronen 1995; Eronen et al. 1996a, 19966). More recently, attention has been paid to the comorbidity rate of schizophrenia and alcoholism, which has been reported to range from 6 to 44 percent among subjects with schizophrenia (Peace and

Mellsop 1987; Regier et al. 1990; Cohen 1995; Rice and Harris 1995). Data obtained from selected populations suggest that subjects with this comorbidity have a more than twofold risk of committing violent crimes than non- alcoholic subjects with schizophrenia (Lindqvist and Allebeck 1989; Swanson et al. 1990; Eronen et al. 19966). According to Rice and Harris (1995), who studied 618 criminal offenders, 26 percent of subjects with schizo- phrenia who abused alcohol were violent offenders, com- pared with 7 percent of subjects with schizophrenia who did not abuse alcohol.

Many studies concerning psychiatric disorders and criminality have been the subject of fair criticism due to their methodological weaknesses, that is, the study mate- rial was limited to criminal offenders or psychiatric patients, the followup periods were short, or the diagnoses were not validated. The first study that can be considered free from this sampling bias was the long-term followup of inpatients with schizophrenia that focused on whether they had committed assault (Lindqvist and Allebeck 1990). The researchers analyzed 644 subjects with schizophrenia and found that 38 (6%) of them had been violent, and of these, 14 (37%) had a drinking problem. Only one of the four violent women in this study abused alcohol, as were five of eight multiple offenders (>2 offenses). The subjects with schizophrenia committed four times as many violent crimes as did the general population.

The largest amount of evidence thus far has been obtained from Hodgins' longitudinal prospective studies of an unselected birth cohort in Stockholm, Sweden (Hodgins 1992), and in Denmark (Hodgins et al. 1996). The Stockholm cohort (n = 15,117) was followed to the age of 30 and the data were collected from national psy- chiatric and police registers. It was found that men with

Reprint requests should be sent to Dr. J. Tiihonen, Dept. of Forensic Psychiatry, University of Kuopio, Niuvanniemi Hospital, FIN-70240 Kuopio, Finland.

437

Schizophrenia Bulletin, Vol. 24, No. 3; 1998 P. Rasanen et al.

major mental disorders (schizophrenia, major affective disorders, paranoid states, and other psychoses) were 2.5 times more likely to commit a crime than men not suffer- ing from such a disorder and 4 times more likely to com- mit a violent offense. Women with major mental disorders were 5 times more likely to commit all crimes and 27 times more likely for violent offenses. Alcoholics were also at greater risk for committing crimes. The study did not determine risks for specific mental disorder diagnoses separately. In the Danish cohort, men with major mental disorder (n = 3,130) were 2.4 to 4.5 times more likely to commit a violent crime, and those who also abused alco- hol (n = 1,731) were 4.2 to 6.7 times more likely to com- mit a violent crime than the mentally healthy group.

At present, there are no published data examining extensive bias-free material on the comorbidity of schizo- phrenia and alcoholism and its relationship to violence. We have tackled this issue by evaluating a methodologi- cally firm, unselected birth cohort.

Methods

The Northern Finland 1966 Birth Cohort Study is an un- selected, general population birth cohort based on all pregnant women (n = 12,068) in the provinces of Lapland and Oulu who gave birth to 12,058 live-born children in 1966. Study material consisted of cohort members (n = 11,017) who were living in Finland at the age of 16. Variables relating to the subjects' biopsychosocial condi- tions have been collected prospectively from pregnancy up to the age of 27. The description of the birth cohort has been presented in detail (Rantakallio 1988). Data on psy- chiatric hospitalizations were collected from the Finnish Hospital Discharge Register (FHDR) until the end of 1993, and all subjects with psychiatric diagnoses accord- ing to DSM-III-R (American Psychiatric Association 1987) were identified (n = 561). The FHDR, which was founded in 1967, contains all psychiatric diagnoses of patients treated in all mental and general hospitals, and it has been found to be a valid and reliable tool for epidemi- ological research (Keskimaki and Aro 1991; Isohanni et al. 1997).

The diagnostic validation was made by psychiatrists who scrutinized all hospital case notes for all psychoses, personality disorders, and substance use disorders. The Operational Criteria Checklist for Psychotic Illness (OPCRIT; McGuffin et al. 1991) was used to yield diag- noses according to DSM-III-R criteria. In addition, clini- cal information was also entered into the proforma for DSM-III-R diagnoses used by the Finnish Adoption Study (Tienari et al. 1993). Concordant diagnoses from these two systems were accepted as valid. Finally, dis-

crepant diagnoses as well as all clinical material were evaluated by two senior clinicians in order to reach a con- sensus DSM-III-R diagnosis.

Reliability was ensured in the following way: First, in a cross-national reliability exercise, 40 case summaries were reviewed for DSM-III-R criteria and placed into eight hierarchical categories. The kappa for this approach was good—0.77 (M.I. vs. J.M.). Second, all diagnoses of the cohort members were validated for the DSM-III-R criteria until the end of 1993. Reliability was ensured in many phases, having good kappa values, or range of 0.6 to 0.9. Ultimately, 76 subjects with schizophrenia (51 males, 25 females) were identified (Isohanni et al. 1997). Further, a second revaluation was performed by a psy- chiatric resident, who collected all the available data on alcohol and drug use habits of subjects with schizophrenia from the case records without knowing the results of the criminal record survey. Of the 76 subjects with schizo- phrenia, 11 male and 2 female subjects were classified as abusing alcohol or being alcohol dependent (DSM-III-R), termed here as alcoholism. Three of 13 alcoholics were also using cannabis. The diagnoses (n = 13), settled by a junior researcher, went through the diagnostic validation process described above and all were classified as alco- holism. Data on crimes were collected from files main- tained by the Ministry of Justice until the end of 1992 (Rantakallio et al. 1995). This national register includes all crimes known to police committed anywhere in Finland. Violent crimes were considered to be homicide, assault, robbery, arson, or violation of domestic peace.

Results

The prevalence of violent offenders among nonhospital- ized men in the cohort was 2.2 percent and among non- hospitalized women, 0.2 percent before the age of 27. Of the male subjects with schizophrenia and comorbid alco- holism, 36.4 percent had committed violent crimes, com- pared with 7.5 percent for nonalcoholic subjects with schizophrenia. None of the women with schizophrenia, two of whom were also alcoholics, had committed any crimes.

Table 1 shows the number of subjects, prevalence of violent offenders, odds ratios (OR), 95 percent confidence intervals (CIs), and p-values for each diagnostic group among the male cohort members.

Men with concomitant alcoholism and schizophrenia were 25.2 (95% CI 6.1-97.5) times more likely to commit violent crimes than mentally healthy men. The corre- sponding risk for nonalcoholic subjects with schizophre- nia was 3.6 (95% CI 0.9-12.3) and for other psychoses, 7.7 (95% CI 2.2-23.9).

438

Schizophrenia, Alcohol Abuse, and Violent Behavior Schizophrenia Bulletin, Vol. 24, No. 3, 1998

Table 1. Male subjects convicted of violent crimes before the age of 27: The Northern Finland 1966 Birth Cohort

DSM-III-R diagnosis

Schizophrenia Schizophrenia with alcoholism1

Schizophrenia without alcoholism Schizophrenia spectrum2

Other psychoses3

All psychoses No diagnosis

Subjects, n

51 11 40

8 27 86

5,287

Prevalence of violent

offenders, n (%)

7(13.7) 4 (36.4) 3 (7.5) 0(0.0) 4(14.8)

11 (12.8) 117(2.2)

Odds ratios (95% Cl)

7.0(2.8-16.7) 25.2 (6.1-97.5) 3.6(0.9-12.3) 0.0 (0.0-0.0) 7.7 (2.2-23.9) 6.5(3.2-13.0) 1.0 (reference)

p value (Fisher's

exact test)

0.0001 <0.0001

0.06

0.003 <0.0001

Note.—C\ = confidence interval. DSM-III-R » Diagnostic and Statistical Manual of Mental Disorders. 3rd ed., revised (American Psychiatric Association 1987). 1 Alcoholism = alcohol abuse or dependence (DSM-III-R). 2DSM-III-R diagnosis of schizophreniform disorder (295.40), schizoaffective disorder (295.70), delusional disorder (297.10), schizoid personality disorder (301.20), or schizotypal personality disorder (301.22). ^DSM-III-R diagnosis of psychoses due to alcohol or psychoactive substance abuse, organic reasons, major depression, and psychotic disorder not otherwise specified (291-299, except 295 and 297).

We also studied the crime recidivism among the male cohort members and found that none of the nonalcoholic subjects with schizophrenia had committed three or more offenses (violent or any other crimes) during the 13-year followup (1981-93). The risk for recurring crimes among alcoholic subjects with schizophrenia was ninefold (n = 2, OR 9.5, 95% CI 2.7-30.0, p = 0.03, Fisher's exact test) and in the group of all psychoses, fivefold (n = 9, OR 5.0, 95% CI 2.3-10.6, p = 0.0002, Fisher's exact test) more than the value calculated for mentally healthy men.

Discussion

In this study, we were able to avoid the traditional prob- lems of restricted data or poorly validated diagnoses that have often been a problem in previous studies on crimi- nality and psychiatric disorders. To our knowledge, this is the first report in which the comorbidity of alcoholism and schizophrenia and its risk association with criminality have been calculated from a large unselected birth cohort with a prospective followup to age 26. However, we must note some limitations of the present study. First, the ORs may not be reliable in the subgroups with the smallest numbers of cases. Also, the cohort members mainly live in nonurban areas and may use less alcohol than subjects in cities. On the other hand, the causality between alco- holism and violence is not necessarily obvious. Patients may have used alcohol coincidentally during their crimi- nal act, or drinking might have been an attempt to reduce the symptoms of their mental disorder. It is also possible that subjects with violent behavior have been more likely to be seen by a psychiatrist and, therefore, more likely to obtain an alcoholism diagnosis. However, it is probable that very few subjects with schizophrenia (even without

violent behavior) would have remained undetected in our cohort.

Our findings agree with earlier reports that subjects with schizophrenia are four to seven times more likely to commit violent offenses than the general population (Lindqvist and Allebeck 1990; Hodgins 1992; Tiihonen et al. 1993). We consider the study findings important because the cohort members are still so young. Eronen and colleagues (1996ft) found that concomittant alco- holism increased the risk of violence by over twofold compared with nonalcoholic subjects with schizophrenia, and in our study this risk increase was over sevenfold. This was not the case for the female subjects in our study, which may be because of the cohort's small number of females with psychoses or because of the youth of the cohort members. Also, women may start committing vio- lent offenses and develop schizophrenia later in life than men (Hodgins 1992; Hodgins et al. 1996).

The case-record study material might underestimate both alcohol and drug consumption of subjects with schizophrenia. However, in Finland only a small propor- tion (3%) of people with schizophrenia are treated solely as outpatients, and the hospital case notes are generally good from the diagnostic point of view (Isohanni et al. 1997). We think that this study succeeded in avoiding the problem of false-negative cases and misdiagnosis of sub- stance-induced psychoses as schizophrenia.

Our most important findings were that one-fifth of male subjects with schizophrenia were already alcoholics before the age of 27, and they were seven times more likely to commit a violent crime than nonalcoholic sub- jects with schizophrenia (OR 25.2 vs. 3.6). These findings greatly exceed the figures reported in the literature (Lindqvist and Allebeck 1989; Swanson et al. 1990;

439

Schizophrenia Bulletin, Vol. 24, No. 3, 1998 P. Rasanen et al.

Eronen et al. 1996&). We also found that the risk for vio- lence by nonalcoholic subjects with schizophrenia was about half that of subjects with other psychoses. Another important finding was that none of the nonalcoholic sub- jects with schizophrenia were recidivists, whereas there was about a tenfold risk for alcoholic subjects with schizo- phrenia compared with the general population. It is essen- tial to remember that deinstitutionalization and reduction of the number of mental hospital beds can increase the substance abuse problems of severely mentally ill persons (Lamb 1993). We think that in the risk evaluation and the prevention of criminality among subjects with major men- tal disorders, it is important to pay attention to the possibil- ity of coexisting alcoholism. Treatment of this aspect of the patient's condition should not be overlooked.

References

American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed., revised. Washington, DC: The Association, 1987.

Cirincione, C ; Steadman, H.J.; Robbins, P.C.; and Monahan, J. Schizophrenia as a contingent risk factor for criminal violence. International Journal of Law and Psychiatry, 15:347-358,1992.

Cohen, S.I. Overdiagnosis of schizophrenia: Role of alco- hol and drug misuse. Lancet, 346:1541-1542,1995.

Eronen, M. Mental disorders and homicidal behavior in female subjects. American Journal of Psychiatry, 152:1216-1218,1995.

Eronen, M.; Hakola, P.; and Tiihonen, J. Mental disorders and homicidal behavior in Finland. Archives of General Psychiatry, 53:497-501, 1996a.

Eronen, M.; Tiihonen, J.; and Hakola, P. Schizophrenia and homicidal behavior. Schizophrenia Bulletin, 22(l):83-89, 1996fc.

Hodgins, S. Mental disorder, intellectual deficiency, and crime: Evidence from a birth cohort. Archives of General Psychiatry, 49(6):476-483, 1992.

Hodgins, S.; Mednick, S.A.; Brennan, P.A.; Schulsinger, R; and Engberg, M. Mental Disorder and Crime: Evidence from a Danish cohort. Archives of General Psychiatry, 53(6):489-496,1996.

Isohanni, M.; Makikyro, T.; Moring, J.; Rasanen, P.; Hakko, H.; Partanen, U.; Koiranen, M.; and Jones, P. A comparison of clinical and research DSM-III-R diagnoses of schizophrenia in a Finnish national birth cohort. Clinical and research diagnoses of schizophrenia. Social Psychiatry and Psychiatric Epidemiology, 32(5):303-308, 1997.

Keskimaki, I., and Aro, S. The accuracy of data on diag- noses, procedures and accidents in the Finnish Hospital Discharge Register. International Journal of Health Sciences, 2:15-21,1991.

Lamb, H.R. Lessons learned from deinstitutionalisation in the U.S. British Journal of Psychiatry, 162:587-592, 1993.

Lindqvist, P., and Allebeck, P. Schizophrenia and assaul- tive behaviour: The role of alcohol and drug abuse. Ada Psychiatrica Scandinavica, 82:191-195, 1989.

Lindqvist, P., and Allebeck, P. Schizophrenia and crime: A longitudinal follow-up of 644 schizophrenics in Stockholm. British Journal of Psychiatry, 157:345-350, 1990.

McGufFin, P.; Farmer, A.E.; and Harvey, I. A polydiagnos- tic application of operational criteria in psychotic illness: Development and reliability of the OPCRIT system. Archives of General Psychiatry, 48(8):764-770, 1991.

Peace, K., and Mellsop, G. Alcoholism and psychiatric disorder. Australian and New Zealand Journal of Psychiatry, 21:94-101,1987.

Rantakallio, P. The longitudinal study of the Northern Finland birth cohort of 1966. Pediatric and Perinatal Epidemiology, 2:59-88, 1988.

Rantakallio, P.; Myhrman, A.; and Koiranen, M. Juvenile offenders, with special reference to sex differences. Social Psychiatry and Psychiatric Epidemiology, 30:113-120, 1995.

Regier, D.A.; Farmer, M.E.; Rae, D.S.; Locke, B.Z.; Keith, S.J.; Judd, L.L.; and Goodwin, F.K. Comorbidity of mental disorders with alcohol and other drug abuse: Results from the Epidemiological Catchment Area (ECA). Journal of the American Medical Association, 264(19):2511-2518,1990.

Rice, M.E., and Harris, T. Psychopathy, schizophrenia, alcohol abuse and violent recidivism. International Journal of Law and Psychiatry, 18:333-342, 1995.

Swanson, J.; Holzer, C ; Ganju, V.; and Jono, R. Violence and psychiatric disorder in the community: Evidence from the Epidemiologic Catchment Area surveys. Hospital and Community Psychiatry, 41:761-770, 1990.

Tienari, P.; Wynne, L.C.; Moring, J.; Wahlborg, K.-E.; Sorri, A.; Naarala, M.; and Lahti, I. Genetic vulnerability or family environment? Implications from the Finnish adoptive family study of schizophrenia. Psychiatrica Fennica, 24:23-41, 1993.

Tiihonen, J.; Eronen, M.; and Hakola, P. Criminality asso- ciated with mental disorders and intellectual deficiency. Archives of General Psychiatry, 50:917-918,1993.

440

Schizophrenia, Alcohol Abuse, and Violent Behavior Schizophrenia Bulletin, Vol. 24, No. 3, 1998

The Authors

Pirkko Rasanen, M.D., Ph.D., is Associate Professor, and Jari Tiihonen, M.D., Ph.D., is Professor and Chairman, Department of Forensic Psychiatry, University of Kuopio, Niuvanniemi Hospital, Kuopio, Finland, Matti Isohanni,

M.D., Ph.D., is Professor and Chairman, Jari Lehtonen, M.D., is Psychiatrist, and Juha Moring is Medical Director, Department of Psychiatry, University of Oulu, Finland, Paula Rantakallio, M.D., Ph.D., is Emeritus Professor, Department of Public Health Science and General Practice, University of Oulu, Finland.

Back Issues Available

Several back issues of the Schizophrenia Bulletin are still available to requesters:

Schizophrenia Bulletin, Vol. 19, No. 4,1993 (Issue theme: Late-Life Schizophrenia)

Schizophrenia Bulletin, Vol. 20, No. 2,1994 (Featured topics: Etiology, Affect, and Treatment)

Schizophrenia Bulletin, Vol. 20, No. 3,1994 (Featured topics: Neurodevelopmental Factors, Genetics, and Family Issues)

Schizophrenia Bulletin, Vol. 21, No. 1,1995 (Featured topics: Schizotypy, Dopamine, and Services Research)

Schizophrenia Bulletin, Vol. 21, No. 2, 1995 (Issue theme: Israeli High-Risk Study)

Schizophrenia Bulletin, Vol. 21, No. 3, 1995 (Featured topics: Psychosocial Factors and Medication Side Effects)

Schizophrenia Bulletin, Vol. 21, No. 4,1995 (Issue theme: Treatment Outcomes Research)

Schizophrenia Bulletin, Vol. 22, No. 1, 1996 (Featured topics: Treatments, Violence, and Central Nervous System Function)

Schizophrenia Bulletin, Vol. 22, No. 2, 1996 (Issue theme: Early Detection and Intervention in Schizophrenia)

Schizophrenia Bulletin, Vol. 22, No. 3, 1996 (Issue theme: Medical Comorbidity in Schizophrenia)

Schizophrenia Bulletin, Vol. 22, No. 4, 1996 (Featured topics: Treatment, Schizotypy, Communication Patterns, and Neuroimaging)

Schizophrenia Bulletin, Vol. 23, No. 1, 1997 (Featured topics: Neuropsychology, Schizotypy, Treatment, and Clinical Course)

Schizophrenia Bulletin, Vol. 23, No. 2, 1997 (Issue theme: Substance Abuse and Schizophrenia)

Schizophrenia Bulletin, Vol. 23, No. 3,1997 (Issue theme: The Neuroanatomy of Schizophrenia)

441