Prevalence of a loaded firearm in the home among birth parents experiencing symptoms of postpartum depression.

Firearm access increases the risk of suicide among all household members. The prevalence of loaded firearms in the home among those experiencing symptoms of postpartum depression (PPD) is unknown. We conducted a cross-sectional study using Pregnancy Risk Assessment Monitoring System (PRAMS) data from 2012 to 2019. We included participants from the nine jurisdictions that asked about loaded firearms in the home and who screened positive for PPD. We excluded participants whose infants were not alive at time of survey completion and who did not respond to the firearm question, resulting in an analytic sample of 4986 participants. Using PRAMS analytic weights, we estimated the prevalence of a loaded firearm in the home and the prevalence of screening for PPD based on having a loaded firearm in the home. Among PRAMS participants experiencing symptoms of PPD, 8.8% (95% CI: 7.6%, 10.1%) reported there was a loaded firearm in their home. Participants with a loaded firearm in their home were more likely to be White (81.3% vs. 60.6%) and live in a rural area (57.9% vs. 27.5%) than those without. Among participants who reported attending a postpartum checkup, 78.6% (95% CI: 67.0%, 90.2%) of those with a loaded firearm in their home reported having been asked by a provider if they were feeling depressed, compared to 88.7% (95% CI: 85.3%, 92.0%) of those without. About 1 in 11 birth parents experiencing symptoms of PPD report a loaded firearm in their home. Further screening for firearm access in this population may need to be considered.

Characterizing Female Firearm Suicide Circumstances: A Natural Language Processing and Machine Learning Approach.

INTRODUCTION: Since 2005, female firearm suicide rates increased by 34%, outpacing the rise in male firearm suicide rates over the same period. The objective of this study was to develop and evaluate a natural language processing pipeline to identify a select set of common and important circumstances preceding female firearm suicide from coroner/medical examiner and law enforcement narratives. METHODS: Unstructured information from coroner/medical examiner and law enforcement narratives were manually coded for 1,462 randomly selected cases from the National Violent Death Reporting System. Decedents were included from 40 states and Puerto Rico from 2014 to 2018. Naive Bayes, Random Forest, Support Vector Machine, and Gradient Boosting classifier models were tuned using 5-fold cross-validation. Model performance was assessed using sensitivity, specificity, positive predictive value, F1, and other metrics. Analyses were conducted from February to November 2022. RESULTS: The natural language processing pipeline performed well in identifying recent interpersonal disputes, problems with intimate partners, acute/chronic pain, and intimate partners and immediate family at the scene. For example, the Support Vector Machine model had a mean of 98.1% specificity and 90.5% positive predictive value in classifying a recent interpersonal dispute before suicide. The Gradient Boosting model had a mean of 98.7% specificity and 93.2% positive predictive value in classifying a recent interpersonal dispute before suicide. CONCLUSIONS: This study developed a natural language processing pipeline to classify 5 female firearm suicide antecedents using narrative reports from the National Violent Death Reporting System, which may improve the examination of these circumstances. Practitioners and researchers should weigh the efficiency of natural language processing pipeline development against conventional text mining and manual review.