Intimate partner violence and the risk of incident perinatal depression: A prospective study during pregnancy and the postpartum period

Perinatal depression refers to a major depressive episode that occurs during pregnancy (antenatal depression) or after childbirth (postpartum depression), affecting 11.9% of pregnant women worldwide.1 Preventing perinatal depression is essential to improve maternal and infant wellbeing. One of its risk factors is suggested to be intimate partner violence (IPV)—psychological, sexual, or physical violence by an intimate partner. While accumulating evidence by screening instruments shows the association of IPV with perinatal depressive symptoms,2 only two studies employed a diagnostic instrument at a single time point to suggest the corresponding association with antenatal depression3, 4; no study had utilized diagnostic instruments to determine the new onset of perinatal depression within a specified period. Since screening instruments are not designated for the basis of treatment, evidence based on diagnostic instruments is significant in clinical settings. Furthermore, a survival analysis of incident perinatal depression is expected to detect vulnerable time spans to IPV experiences, contributing to the effective distribution of medical and healthcare resources during the perinatal period. By employing a diagnostic instrument, we aimed to examine whether IPV is associated with the risk of incident perinatal depression across pregnancy and after delivery among Japanese pregnant women. This longitudinal study used data from 2508 Japanese pregnant women without lifetime bipolar disorder or a major depressive episode in the past month who were assigned to a control group of a previous randomized controlled trial that assesses major depressive episodes during pregnancy and postpartum via a smartphone-based app (trial registration number: UMIN000038190).5, 6 We excluded participants with ID duplication (n = 59) and those without consent for this study (n = 14) or a spouse/partner (n = 19), enrolling 2416 women in the analysis. Recent experience of IPV was measured by a self-reporting questionnaire: the Violence Against Women Screen short form at baseline (during 16–20 gestational weeks).7 Incident major depressive episode during the perinatal period from baseline to 3 months postpartum based on the DSM-IV-TR criteria was ascertained by the self-administered version of the World Health Organization Composite International Diagnostic Interview 3.0 depression section,8 which has a good concordance with the interviewer-delivered version.9 Depressive symptoms during the perinatal period were ascertained by the Edinburgh Postnatal Depression Scale (EPDS), and the cutoff points of ≥9, ≥11, and ≥13 were applied to evaluate incident perinatal depressive symptoms.10 A survival analysis of incident perinatal depression and depressive symptoms related to IPV was conducted using a Cox proportional hazards model. The confounding variables were adjusted for age, educational attainment, marital history, occupational status, number of children, and planning of pregnancy. All statistical tests were performed using STATA 18.0 software (Stata Corporation, College Station, Texas, USA) and values of P < 0.05 (two-tailed) were considered statistically significant. Among the participants, the prevalence of IPV was 375 (15.5%). Due to high confidentiality of the smartphone-based IPV questionnaire, participants appeared at ease to disclose their IPV experiences, reducing the fear of retaliation by the partner. During a mean follow-up of 6.5 months (standard deviation, 3.2), we documented 68 new onsets of perinatal depression. According to the Kaplan–Meier curves of incident perinatal depression (Fig. 1a), the incidence of perinatal depression in IPV victims (n = 18) occurred evenly throughout the observation period. Clinicians should monitor the depressive symptoms of IPV victims constantly during both antenatal and postpartum periods. In a Cox proportional hazards regression, IPV was associated with an increased risk of perinatal depression: the multivariable-adjusted hazard ratio was 2.09 (95% CI, 1.21–3.62) for major depressive episode (Fig. 1b; Table S2). Similar association was observed with each risk of EPDS ≥9, ≥11, and ≥13: the corresponding hazard ratio ranged from 1.5 to 1.9 (Fig. 1b; Table S3 to S5). Each exposure to psychological, sexual, and physical violence was similarly associated with an increased risk of incident perinatal depression (Table S6). Monitoring and reducing every type of IPV exposure is suggested crucial to prevent incident perinatal depression. Several limitations of our study need to be considered. Voluntary participation via the app may partially limit the generalizability of our results. All variables were assessed by self-reports. Neither IPV after recruitment nor depression onset beyond 3 months postpartum was assessed. Unmeasured confounding factors may exist, such as cohabitation with their partner and family, household income, hometown delivery, and other psychiatric comorbidities. In conclusion, this is the first study to show that IPV was associated with an increased risk of incident perinatal depression, suggesting the significance of online IPV screening during pregnancy to prevent the development of perinatal depression. Details regarding the introduction, methods, results, and discussion are provided in the Supplementary Materials. DN oversaw this study, supervising the process and providing his expert opinion. KT and DN contributed to the conception and design of the work; DN contributed to the acquisition of data; YI and NS created the dataset for analysis; KT contributed to the analysis and interpretation of data and drafted the manuscript; YI, NS, and DN revised the manuscript. All authors have approved the submitted version of the manuscript to be published. We are grateful to all participants who took part in the study and made this research possible. Informed consent was obtained from all the participants before participation in the study. The study protocol, which complies with the Declaration of Helsinki, was approved by the Graduate School of Medicine and Faculty of Medicine research ethics committee of the University of Tokyo (Number: 2019150NI). This work was supported by a Grant-in-Aid for Scientific Research A (19H01073 to DN) from the Japan Society for the Promotion of Science. The study's sponsor had no role in the study design, data collection, data analysis, data interpretation, or report writing. MTI Ltd. has been related to this study, as stated in the Supplementary Materials. Specifically, the authors paid the necessary expenses to MTI Ltd., and MTI Ltd. cooperated with the authors in recruiting study participants and collecting information from them electronically, but no financial support was provided. YI received personal fees outside the submitted work from IkiIki Work Inc. and Fast Retailing Co., Ltd. NS received personal fees from Medilio Co., Ltd. outside the submitted work. DN received personal fees outside the submitted work from Startia, Inc., MD.net, and an honorarium from Takeda Pharmaceutical Co. and Otsuka Medical Devices Co., Ltd. The other author declares no conflict of interest. Daisuke Nishi is an Editorial Board member of Psychiatry and Clinical Neurosciences and a co-author of this article. To minimize bias, they were excluded from all editorial decision-making related to the acceptance of this article for publication. The data that support the findings of this study are available from the corresponding author upon reasonable request. Fig. S1. Kaplan–Meier failure curves of incident EPDS ≥9 according to intimate partner violence. Fig. S2. Kaplan–Meier failure curves of incident EPDS ≥11 according to intimate partner violence. Fig. S3. Kaplan–Meier failure curves of incident EPDS ≥13 according to intimate partner violence. Table S1. Demographic characteristics of participants according to intimate partner violence. Table S2. Hazard ratios and 95% confidence intervals of incident perinatal depression according to intimate partner violence. Table S3. Hazard ratios and 95% confidence intervals of incident EPDS≥9 according to intimate partner violence. Table S4. Hazard ratios and 95% confidence intervals of incident EPDS≥11 according to intimate partner violence. Table S5. Hazard ratios and 95% confidence intervals of incident EPDS≥13 according to intimate partner violence. Table S6. Hazard ratios and 95% confidence intervals of incident perinatal depression according to intimate partner violence subtypes. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.