A 4-Fold Risk of Metabolic Syndrome in Patients With Schizophrenia: The Northern Finland 1966 Birth Cohort Study
Saari KM, Lindeman SM, Viilo KM, Isohanni MK, Jarvelin MR, Lauren LH, et al.;
Commented by , 30 May 2005
Aim of the study
People with schizophrenia die on average 10 years earlier than normal controls. This shortened life expectancy is not only due to suicide but also to an association with a number of medical diseases such as cardiovascular disorders, diabetes or sudden death.
Metabolic syndrome is a major risk factor for the latter problems and has been shown to occur at an increased prevalence in schizophrenia. It is composed of:
- abdominal obesity
- hypertriglyceridemia
- decreased high-density lipoproteninemia (HDL)
- high blood pressure
- hyperglycemia
The current article assessed the prevalence of metabolic syndrome in a population-based birth cohort of patients with schizophrenia.
Methods
5613 participants of the Northern Finland 1966 Birth Cohort who took part in the current field study from 1997 to 1998 were analysed. The Northern Finland 1966 Birth Cohort is an unselected, general population based sample of births in the provinces Lapland and Oulu.
The Finnish hospital discharge register was used to obtain information which members of the sample suffered from schizophrenia and the sample was divided into four diagnostic categories according to DSM-III-R:
- schizophrenia (N = 31)
- other functional psychoses (N = 22)
- nonpsychotic disorders (N = 105)
- no psychiatric hospital treatment (N = 5455, i.e. the comparison group).
The presence of metabolic syndrome was assessed in all participants using the criteria of the Finish National Cholesterol Education Program, i.e. the presence of three of the above mentioned criteria.
Results
Participants with schizophrenia had a statistically significant increased prevalence of metabolic syndrome compared to normal controls (19% vs. 6%, p = .010). Subjects with "other psychoses" and "nonpsychotic psychiatric disorders" did not have an increased risk for metabolic syndrome (5% and 9%, respectively).
According to a logistic regression analysis that controlled for sex, the risk of metabolic syndrome in schizophrenia was shown to be 3.7 higher than in normal controls (95% CI = 1.5 to 9.0). Among the single components of metabolic syndrome indicated above only abdominal obesity and hypertriglyceridemia were significantly increased.
Dr Leucht's comments
The report is one of the first epidemiological studies trying to assess risk factors that may explain the well-known increased mortality rates of people with schizophrenia. Methodological strong points are the use of a population based sample and of a cohort with the same year of birth, thus eliminating age as a source of bias.
A confounder may be the use of the Finnish hospital discharge register to identify psychiatric cases. The authors state that almost all patients with schizophrenia are at some stage hospitalised in Finland so that the results should be representative for schizophrenia.
Nonpsychotic disorders have, however, not necessarily been hospitalised at least once so that the sample may not be representative for them. Furthermore, although the authors started out with a high number of participants, there were relatively few cases with schizophrenia due to the low-prevalence of the disorder.
Nevertheless, the high prevalence of metabolic syndrome in schizophrenia identified already at the beginning of the patients’ thirties highlights the importance of the problem. Strategies to control weight and diet and to enhance physical activity are needed.