a1 Centre for Women's Mental Health Research, The University of Manchester, Manchester, UK
a2 Biostatistics/Health Methodology Research Group, The University of Manchester, Manchester, UK
a3 National Centre for Register-based Research, University of Aarhus, Denmark
Background Few large studies describe links between maternal mental illness and risk of major birth defect in offspring. Evidence is sparser still for how effects vary between maternal diagnoses and no previous study has assessed risk with paternal illnesses.
Method A population-based birth cohort was created by linking Danish national registers. We identified all singleton live births during 1973–1998 (n=1.45 m), all parental psychiatric admissions from 1969 onwards, and all fatal birth defects until 1 January 1999. Linkage and case ascertainment were almost complete. Relative risks were estimated using Poisson regression.
Results Risk of fatal birth defect was elevated in relation to history of any maternal admission and also with affective disorders specifically, although the strongest effect found was with maternal schizophrenia. The rate was more than doubled in this group compared to the general population [relative risk (RR) 2.34, 95% confidence interval (CI) 1.45–3.77], which also represented a significant excess risk compared with all other admitted maternal disorders (p=0.018). Risk of death from causes other than birth defect was no higher with schizophrenia than with other maternal conditions. There was no elevation in risk of fatal birth defect if the father was admitted with schizophrenia or any other psychiatric diagnosis.
Conclusions There are many possible explanations for a higher risk of fatal birth defect with maternal schizophrenia and affective disorder. These include genetic effects directly linked with maternal illness, lifestyle factors (diet, smoking, alcohol and drugs), poor antenatal care, psychotropic medication toxicity, and gene–environment interactions. Further research is needed to elucidate the causal mechanisms.
(Received May 08 2007)
(Revised October 16 2007)
(Accepted October 30 2007)
(Online publication December 13 2007)
c1 Address for correspondence: Dr R. T. Webb, Centre for Women's Mental Health Research/Health Methodology Research Group, Williamson Building, The University of Manchester, Oxford Road, Manchester M13 9PL, UK. (Email: firstname.lastname@example.org)