Relationship between in utero exposure to influenza epidemics and risk of schizophrenia in Denmark.

Several recent epidemiological studies suggest that exposure to influenza during gestation increases the risk of later developing schizophrenia. Inconsistency exists, however, particularly in studies that have examined the relationship between the prevalence of influenza and the monthly number of schizophrenic births, over many years. Our sample (N = 9462) was obtained from a Danish computerized case register, and consisted of schizophrenia patients born between 1915 and 1970, and first admitted to Danish psychiatric hospitals between 1971 and 1991. The study sample was chosen to represent "incidence cases" to allow us to calculate the population attributable risk fraction (PAF). The temporal correlation of fluctuations in the prevalence of influenza and fluctuations in the monthly number of preschizophrenic births was examined using a Poisson regression analysis. Exposure to influenza 4 months prior to birth (i.e., about the 6th month of gestation) was significantly associated with an increased risk of later schizophrenia, especially for narrowly defined schizophrenia. The number of schizophrenic births was found to have risen by 12% (95% confidence interval: 1-24%) for every 100,000 cases of influenza in the 4th month before birth. Our model indicates the PAF to be 1.4%, that is, only 1.4% of the whole schizophrenic sample is attributed to prenatal exposure to influenza. Although maternal exposure to influenza during midgestation is not a major risk factor for schizophrenia, the elucidation of its causal mechanism may open the avenue to understanding the neurodevelopmental origins of the disease.