The Prevalence and Correlates of Self-Harm in the Perinatal Period: A Systematic Review

OBJECTIVE: To perform a systematic review and meta-analysis on research on prevalence and correlates of self-harm in pregnancy and the postpartum year ("perinatal self-harm"). DATA SOURCES: Six databases (EMBASE, MEDLINE, PsycINFO, Maternity and Infant Care Database, CINAHL, Cochrane Controlled Register of Trials) were searched from inception through October 31, 2018. STUDY SELECTION: Inclusion criteria were (1) peer-reviewed articles with (2) data available for estimating prevalence and correlates. Exclusion criteria were (1) studies of women seeking abortion and (2) letters, editorials, or case reports/series. DATA EXTRACTION: Two reviewers independently screened all articles, extracted data, and appraised quality. RESULTS: Of 3,913 articles screened, 39 (reporting 19,191,431 pregnancies) were included. Prevalence ranges were as follows: self-harm during pregnancy (14 studies): 0%-2.39% (median = 0.0004%; interquartile range [IQR], 0.0002%-0.18%); self-harm during postpartum year (10 studies): 0%-2.41% (median = 0.17%; IQR, 0.04%-1.05%); self-harm during pregnancy in women with serious mental illness (SMI) (6 studies): 0%-23.78% (median = 2.16%; IQR, 0.26%-7.9%); self-harm during postpartum year in women with SMI (7 studies): 0%-21.9% (median = 7.97%; IQR, 0%-18%). Key correlates of self-harm during pregnancy and the postpartum year include mental disorder, substance misuse, younger age, being unmarried, and obstetric and neonatal complications. Additionally, a history of self-harm and fetal/infant loss were associated with postpartum self-harm. There were limited data on correlates of perinatal self-harm in women with SMI. CONCLUSIONS: Perinatal self-harm appears to be rare but is associated with adverse obstetric and neonatal outcomes. However, it is common in women with SMI, though there is limited evidence regarding correlates and outcomes in this population. More research into the prevalence, correlates, and outcomes of perinatal self-harm, particularly in women with SMI, is needed.