Responses to Concerning Posts on Social Media and Their Implications for Suicide Prevention Training for Military Veterans: Qualitative Study
Abstract: Background: A “concerning post” is a display of a user’s emotional crisis on a social media platform. A better understanding of concerning posts is relevant to suicide prevention, but little is known about social media users’ attitudes and responses to concerning posts. Military veterans in the United States are disproportionately affected by suicide, often use social media, and may have exposure to individuals with elevated suicide risk via concerning posts. Objective: The objective of the study was (1) to obtain insight into whether and how US military veterans respond to members of their social network on social media (ie, “friends”) who are experiencing substantial emotional distress, and (2) to identify potential interventions that could assist in users’ response to concerning posts. Methods: We recruited veterans through Facebook and conducted semistructured interviews with 30 participants between June and December 2017. We used a summary template for rapid analysis of each interview, followed by double-coding using a codebook based on topic domains from the interview guide. Members of the research team met regularly to discuss emerging patterns in the data, generate themes, and select representative quotes for inclusion in the manuscript. Results: Veterans were reluctant to disclose emotional and health issues on Facebook, but they were open to reaching out to others’ concerning posts. There was a complex calculus underlying whether and how veterans responded to a concerning post, which involved considering (1) physical proximity to the person posting, (2) relationship closeness, (3) existing responses to the post, and (4) ability to maintain contact with the person. Veterans desired additional training, backed by community-based veteran organizations, in how to respond to concerning posts from peers. Conclusions: There is a need to incorporate features that will help veterans effectively respond to concerning posts from peers into suicide prevention training and to expand access for veterans to such training.
Abstract: Novel and automated means of opioid use and relapse risk detection are needed. Unstructured electronic medical record data, including written progress notes, can be mined for clinically relevant information, including the presence of substance use and relapse-critical markers of risk and recovery from opioid use disorder (OUD). In this study, we used natural language processing (NLP) to automate the extraction of opioid relapses, and the timing of these occurrences, from veteran patients' electronic medical record. We then demonstrated the utility of our NLP tool via analysis of pre-/post-COVID-19 opioid relapse trends among veterans with OUD. For this demonstration, we analyzed data from 107,606 veterans OUD enrolled in Veterans Health Administration, comparing a pandemic-exposed cohort (n = 53,803; January 2019-March 2021) to a matched prepandemic cohort (n = 53,803; October 2017-December 2019). The recall of our NLP tool was 75% and our precision was 94%, demonstrating moderate sensitivity and excellent specificity. Using the NLP tool, we found that the odds of opioid relapse postpandemic onset were proportionally higher compared to prepandemic trends, despite patients having fewer mental health encounters from which to derive instances of relapse postpandemic onset. In this research application of the tool, and as hypothesized, we found that opioid relapse risk was elevated postpandemic. The application of NLP Methods: to identify and monitor relapse risk holds promise for future surveillance, risk prevention, and clinical outcome research.