Suicide in perinatal and non-perinatal women in contact with psychiatric services: 15 year findings from a UK national inquiry

Abstract

Background: Suicide in pregnant and postnatal women is an important cause of maternal death, but evidence to guide suicide prevention in this group is scarce. We aimed to compare the trend, nature, and correlates of suicide in perinatal and non-perinatal women in contact with psychiatric services.

Methods: We used 1997-2012 data from the UK National Confidential Inquiry into Suicides and Homicides by People with Mental Illness, which includes all suicides by people (age ≥10 years) who had been in contact with psychiatric services in the previous year. The study sample comprised all women who died by suicide in pregnancy or the first postnatal year (perinatal suicides), and all women in the same age range who died by suicide outside this period (non-perinatal suicides). We compared suicides among perinatal and non-perinatal women with logistic regression of multiply imputed data.

Findings: The study sample included 4785 women aged 16-50 years who died by suicide, of whom 98 (2%) died in the perinatal period. Of the 1485 women aged 20-35 years, 74 (4%) women died in the perinatal period. Over the course of the study, we recorded a modest downward trend in the mean number of women dying by suicide in the non-perinatal period (-2·07 per year [SD 0·96]; p=0·026), but not the perinatal period (-0·07 per year [0·37]; p=0·58). Compared with non-perinatal women, women who died by suicide in the perinatal period were more likely to have a diagnosis of depression (adjusted odds ratio [OR] 2·19 [95% CI 1·43-3·34]; p<0·001) and less likely to be receiving any active treatment (0·46 [0·24-0·89]; p=0·022) at the time of death. Women who died by suicide within versus outside the perinatal period were also more likely to be younger (crude OR -6·39 [95% CI -8·15 to -4·62]; p<0·0001) and married (4·46 [2·93-6·80]; p<0·0001), with shorter illness duration (2·93 [1·88-4·56]; p<0·001) and no history of alcohol misuse (0·47 [0·24-0·92]; p=0·027). There were no differences in service contact or treatment adherence.

Interpretation: In women in contact with UK psychiatric services, suicides in the perinatal period were more likely to occur in those with a depression diagnosis and no active treatment at the time of death. Assertive follow-up and treatment of perinatal women in contact with psychiatric services are needed to address suicide risk in this group.

Funding: Healthcare Quality Improvement Partnership and National Institute for Health Research.