Comment on “Association Between Mental Illness and Risk of Severe Maternal Morbidity in Pregnant Veterans”

Maternal mental health problems are a growing concern, affecting ~20% of pregnant or postpartum people. Mental illness is a leading contributor to maternal mortality in the United States and has been associated with severe maternal morbidity (SMM), though it remains unclear whether mental illness preceded or was the result of SMM. Antenatal mental illness has gone understudied and is subject to ascertainment bias due to inconsistent screening practices, reporting stigma, and barriers to psychiatric care. The Veterans Health Administration (VHA) offers a strong setting to study this relationship because of its robust mental health screening, accessible treatment, and a high prevalence of mental illness among veterans. This study evaluated the risk of SMM among veterans with active or historical mental illness before childbirth, hypothesizing that risk would be highest with active illness. This observational cohort study included live births to veterans from 2010 through 2020 covered by the VHA. The VHA does not directly provide prenatal or delivery care and instead outsources to community-based care while capturing related billing data. These data were linked to inpatient and outpatient encounters and delivery discharge records. The exposure was antenatal mental illness, identified through diagnosis codes before delivery and categorized as active, historical, or no prior mental illness. Active mental illness was defined as any diagnosis within 12 months before delivery, while historical mental illness was defined as 12 to 36 months before delivery. Diagnoses included depression, anxiety, posttraumatic stress disorder, bipolar disorder, and schizophrenia. The primary outcome was SMM without transfusion, categorized by organ system. Secondary analyses investigated associations between individual mental illness categories, antepartum pregnancy complications, and SMM subtypes. Modified Poisson regression adjusted for demographic, clinical, obstetric, geographic factors, and multiple births to the same veteran. The cohort included 47,883 births among 37,982 veterans. Of these, 78.9% had no prior mental illness, 3.6% had historical mental illness only, and 17.4% had active mental illness during pregnancy. The most common active diagnoses were depression and anxiety, followed by PTSD and other serious mental illnesses. Substance use disorders were more common among veterans with historical or active mental illness. Active mental illness was associated with increased SMM compared with no prior mental illness (2.0% vs. 1.2%; aRR: 1.38, 95% CI: 1.15-1.67), while historical mental illness was not associated with increased risk (1.3% vs. 1.2%; aRR: 0.87, 95% CI: 0.57-1.31). This increased risk was seen across all psychiatric categories except serious mental illness, the smallest subgroup. SMM risk among veterans with active mental illness was greatest for cardiac, pulmonary, renal, and sepsis-related categories. Mediation analysis suggested that the association largely occurred through a direct pathway rather than through substance use disorder. Hypertensive disorders, gestational diabetes, and preterm birth occurred more frequently among veterans with active or historical mental illness. Overall, active mental illness during pregnancy was independently associated with increased SMM risk, while historical mental illness was not. The increased frequency of cardiopulmonary, renal, and sepsis-related SMM events, along with hypertensive disorders, suggests possible pathways linking active mental illness to SMM. Strengths include the large national cohort, longitudinal VHA data, universal mental health and substance use screening, and the ability to distinguish active from historical illness. Limitations include limited generalizability to nonveterans, reliance on diagnostic coding, residual confounding, and lack of data on parity, mental illness severity, or treatment. The authors conclude that active mental illness should be considered an independent risk factor for SMM and that remission before pregnancy may help reduce adverse pregnancy outcomes among veterans. (Summarized from Panelli DM, Esmaeili A, Joyce VR, et al. Association between mental illness and risk of severe maternal morbidity in pregnant veterans. Obstet Gynecol. 2026;147:528–539. doi:10.1097/AOG.0000000000006107)