Acquisition of new medical devices among the persistently critically ill: A retrospective cohort study in the Veterans Affairs
Abstract: Patients who develop persistent critical illness remain in the ICU predominately because they develop new late-onset organ failure(s), which may render them at risk of acquiring a new medical device. The epidemiology and short-term outcomes of patients with persistent critical illness who acquire a new medical device are unknown. We retrospectively studied a cohort admitted to the Veterans Affairs (VA) ICUs from 2014 to 2019. Persistent critical illness was defined as an ICU length of stay of at least 14 days. Receipt of new devices was defined as acquisition of a new tracheostomy, feeding tube (including gastrostomy and jejunostomy tubes), implantable cardiac device, or ostomy. Logistic regression models were fit to identify patient factors associated with the acquisition of each new medical device. Among hospitalized survivors, 90-day posthospitalization discharge location and mortality were identified. From 2014 to 2019, there were 13,184 ICU hospitalizations in the VA which developed persistent critical illness. In total, 30.4% of patients (N = 3998/13,184) acquired at least 1 medical device during their persistent critical illness period. Patients with an initial higher severity of illness and prolonged hospital stay preICU admission had higher odds of acquiring each medical device. Among patients who survived their hospitalization, discharge location and mortality did not significantly differ among those who acquired a new medical device as compared to those who did not. Less than one-third of patients with persistent critical illness acquire a new medical device and no significant difference in short-term outcomes was identified. Future work is needed to understand if the acquisition of new medical devices is contributing to the development of persistent critical illness.
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.