R.M. Murray

Prenatal influenza and schizophrenia: Is the effect confined to females?

Four longitudinal studies from Scandinavia, England and U.S.A. have suggested that individuals exposed to epidemics of influenza during fetal development are at an increased risk of adult schizophrenia. In an attempt to replicate their findings, we obtained the English national data on first admission ICD schizophrenics between 1980 and 1986. In the analysis, those born in 1958 were excluded as they were responsible for the significant effect we reported previously for the “Asian Flu” pandemic. For each month in the period 193865, we investigated the relationship between the number of schizophrenic births in the sample and the number of influenza deaths, allowing for the fluctuations in the general population birth rate, and the long term and seasonal variations in schizophrenic births. We applied a generalized linear model to our data. Inclusion of the number of influenza deaths 5 months before birth significantly improved the model; the change in the scaled deviance was 4.92 (df=l, p=O.O3). When the procedure was repeated by sex (male, N=2,252; female, N=l,575), we found this effect to be significant for females (p=O.O2) but not for males. The parameter estimates of our model suggest a 7% increase in the number of schizophrenic births for every 1,000 deaths from influenza in the 5th month before birth. The present study provides further evidence of an increase in the risk of schizophrenia among those exposed to influenza epidemics in fetal life.