Military veterans and civilians' mental health diagnoses: an analysis of secondary mental health services
Abstract: Healthcare provision in the United Kingdom (UK) falls primarily to the National Health Service (NHS) which is free at the point of access. In the UK, there is currently no national marker to identify military veterans in electronic health records, nor a requirement to record it. This study aimed to compare the sociodemographic characteristics and recorded mental health diagnoses of a sample of veterans and civilians accessing secondary mental health services. The Military Service Identification Tool, a machine learning computer tool, was employed to identify veterans and civilians from electronic health records. This study compared the sociodemographic characteristics and recorded mental health diagnoses of veterans and civilians accessing secondary mental health care from South London and Maudsley NHS Foundation Trust, UK. Data from 2,576 patients were analysed; 1288 civilians and 1288 veterans matched on age and gender. Depressive disorder was the most prevalent across both groups in the sample (26.2% veterans, 15.5% civilians). The present sample of veterans accessing support for mental health conditions were significantly more likely to have diagnoses of anxiety, depressive, psychosis, personality, and stress disorders (AORs ranging 1.41–2.84) but less likely to have a drug disorder (AOR = 0.51) than age- and gender-matched civilians. Veterans accessing secondary mental health services in South London had higher risks for many mental health problems than civilians accessing the same services. Findings suggest that military career history is a key consideration for probable prognosis and treatment, but this needs corroborating in other geographical areas including national population-based studies in the UK.
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.