Associations between Past Sexual Victimization and Perinatal Depression: A Systematic Review and Meta-Analysis

Background: The birth of a new baby is a transformative event in a woman’s life, often regarded as a joyful and fulfilling experience. However, for some women, the perinatal period can be fraught with significant mental health challenges such as perinatal depression (PND). Experiences of past sexual victimization has important implications for optimizing perinatal care for mothers. Indeed, the American College of Obstetricians and Gynecologist recommends that all women be screened for a history of sexual victimization; however, the low reported rates of sexual victimization among women receiving perinatal care indicates health care providers might not be asking routinely about a history of sexual victimization. A robust synthesis of research assessing a link between past sexual victimization and women’s health outcomes, particularly the presence or absence of PND, could bolster the case for consistent screening practices among practitioners who provide perinatal care.

 

Methods: A protocol was adapted from Cochrane Collaboration guidelines and other expert resources (i.e., Campbell Collection, PRISMA and AMSTAR standards) to develop a priori inclusion/exclusion criteria and conduct a systematic review of English language literature. Inclusion criteria for relevant studies were (a) an explicit focus on past sexual victimization, (b) a measure of depression between conception and one-year postpartum, and (c) sufficient quantitative data to enable calculation of effect sizes assessing the association between sexual victimization and PND. Four bibliographic databases were searched using predefined search strings developed with a social science reference librarian. Data from included studies were tabled and analyzed, and overall and subgroup pooled effect sizes were estimated using random effects meta-analyses using the DerSimonian and Laird (1986) estimation method.  

Findings: Over 500 abstracts were identified for title/abstract review and 59 articles were selected for full text review. Thirty-five articles met all inclusion criteria, resulting in 43 individual effect size estimates. Results from the meta-analysis of the overall effect size indicated a significant average effect size (OR=1.53, 95% CI [1.36, 1.70]), such that past sexual victimization was associated with a 53% increase in the odds that individuals would report experiencing PND. Comparable average effect sizes were yielded from subgroup analyses conducted with respect to (a) the type of analysis conducted (i.e., bivariate versus multivariate), (b) the inclusion of various covariates (i.e., other forms of abuse, past depression), (c) timing of sexual victimization (i.e.,, childhood, adulthood), and (d) the PND measure used. Results from Egger’s test and a contour-enhanced funnel plot suggested that the results could potentially be influenced by various forms of bias, including publication bias.

 

Conclusions/Implications: Results call for the attention of health care providers and social workers to determine how screening practices can be improved during the perinatal period for women who have experienced a history of sexual victimization. Improved screening practices will allow medical providers and social workers to provide more comprehensive perinatal health care for women with past trauma histories. Further, improved screening will ultimately lead to the development of effective interventions to serve this vulnerable population.