Publications

Background: In the United States, suicide accounts for 1 out of every 20 deaths that occur during pregnancy and the first 12 months postpartum. Although nonfatal suicide attempts are the strongest known predictor of death by suicide, there are no prior population based estimates of the incidence of and clinical risk factors for pregnancy associated suicide attempts. Methods: This retrospective cohort study used statewide, all-payer, longitudinally linked hospital and emergency department (ED) patient records from California. Participants included all California residents with an index hospital delivery of a live infant between 2010 and 2020. Outcomes included ED presentation for nonfatal suicide attempt during pregnancy or up to 12 months postpartum. Clinical risk factors of interest included healthcare utilization patterns during pregnancy and behavioral health diagnoses recorded at index delivery. Results: Among delivering patients with an index delivery (N = 3,737,792), 0.13% (n = 4,968) had a suicide attempt during pregnancy or the postpartum period. After adjusting for background demographic characteristics, risks of a postpartum suicide attempt were increased 4- to 30-fold by several clinical factors, including prenatal suicide attempt ED visits, psychiatric ED visits, and assault ED visits, and by psychotic disorders, bipolar disorder, alcohol use disorder, recurrent and single-episode major depressive disorder, and anxiety disorders recorded at delivery. Conclusions: Risks of postpartum suicide attempt were substantially elevated for patients who had behavioral health related ED visits during pregnancy and by several psychiatric disorders at delivery. Clinical consideration should be given to monitoring these patients for suicide attempt risk.

Objective: Distinctive stressors facing pregnant and postpartum individuals during the COVID-19 pandemic may have affected their emergency department (ED) care-seeking for behavioral health concerns and violence victimization. We tested whether the incidence of postpartum behavioral health and assault injury ED visits differed for individuals according to their months of postpartum pandemic exposure. Methods: We used statewide, longitudinally linked hospital and ED administrative claims data from California to classify all individuals with hospital deliveries between January 1, 2016, and December 31, 2020, according to their months of postpartum pandemic exposure. Outcomes comprised 12-month incidence of any ED visit for a psychiatric disorder, drug use disorder/overdose, alcohol use disorder/intoxication, or assault injury, defined using International Classification of Diseases-Clinical Modification, version 10 codes. Risk ratios compared the incidence of each outcome among people with 1-12 months of postpartum pandemic exposure to those with 0 months of exposure. Results: Compared to people with 0 months of postpartum pandemic exposure (n = 1,163,215), delivering people with 1-12 month' exposure (range: n = 26,836 to n = 273,561) were approximately equally likely to have a postpartum ED visit for a psychiatric disorder, drug use disorder, or alcohol use disorder, after adjusting for demographic differences (most p > 0.10). The incidence of assault injury was significantly lower among delivering individuals with 11 or 12 months of pandemic exposure (RRadj = 0.70 and 0.91, respectively; both p < 0.01) compared to those with 0 months. Conclusions: Contrary to expectations, the pandemic did not appear to have affected ED utilization for most behavioral health conditions among postpartum individuals, but assault injury ED visits declined.