The impact of disease severity measures on survival in U.S. veterans with rheumatoid arthritis-associated interstitial lung disease
Abstract: To determine whether RA and interstitial lung disease (ILD) severity measures are associated with survival in patients with RA-ILD. We studied US veterans with RA-ILD participating in a multicentre, prospective RA cohort study. RA disease activity (28-joint DAS [DAS28-ESR]) and functional status (multidimensional HAQ [MDHAQ]) were collected longitudinally while pulmonary function tests (forced vital capacity [FVC], diffusing capacity for carbon monoxide) were obtained from medical records. Vital status and cause of death were determined from the National Death Index and administrative data. Predictors of death were assessed using multivariable Cox regression models adjusting for age, sex, smoking status, ILD duration, comorbidity burden and medications. We followed 227 RA-ILD participants (93% male and mean age of 69 years) over 1073 person-years. Median survival after RA-ILD diagnosis was 8.5 years. Respiratory diseases (28%) were the leading cause of death, with ILD accounting for 58% of respiratory deaths. Time-varying DAS28-ESR (adjusted hazard ratio [aHR] 1.21; 95% CI: 1.03, 1.41) and MDHAQ (aHR 1.85; 95% CI: 1.29, 2.65) were separately associated with mortality independent of FVC and other confounders. Modelled together, the presence of either uncontrolled disease activity (moderate/high DAS28-ESR) or FVC impairment (<80% predicted) was significantly associated with mortality risk. Those with a combination of moderate/high disease activity and FVC <80% predicted had the highest risk of death (aHR 4.43; 95% CI: 1.70, 11.55). Both RA and ILD disease severity measures are independent predictors of survival in RA-ILD. These findings demonstrate the prognostic value of monitoring the systemic features of RA-ILD.
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.