Effects of Implementation of a Supervised Walking Program in Veterans Affairs Hospitals : A Stepped-Wedge, Cluster Randomized Trial
Abstract: Background: In trials, hospital walking programs have beenshown to improve functional ability after discharge, but littleevidence exists about their effectiveness under routine practiceconditions. Objective: To evaluate the effect of implementation of asupervised walking program known as STRIDE (AssiSTed EaRlyMobIlity for HospitalizeD VEterans) on discharge to a skilled-nursing facility (SNF), length of stay (LOS), and inpatient falls. Design: Stepped-wedge, cluster randomized trial. (ClinicalTrials.gov: NCT03300336)Setting:8 Veterans Affairs hospitals from 20 August 2017 to19 August 2019. Patients: Analyses included hospitalizations involving patientsaged 60 years or older who were community dwelling andadmitted for 2 or more days to a participating medicine ward. Intervention: Hospitals were randomly assigned in 2 strati-fied blocks to a launch date for STRIDE. All hospitals receivedimplementation support according to the Replicating EffectivePrograms framework. Measurements: The prespecified primary outcomes weredischarge to a SNF and hospital LOS, and having 1 or moreinpatient falls was exploratory. Generalized linear mixed mod-els werefit to account for clustering of patients within hospitalsand included patient-level covariates. Results: Patients in pre-STRIDE time periods (n=6722) weresimilar to post-STRIDE time periods (n=6141). The propor-tion of patients with any documented walk during a poten-tially eligible hospitalization ranged from 0.6% to 22.7% perhospital. The estimated rates of discharge to a SNF were13% pre-STRIDE and 8% post-STRIDE. In adjusted models,odds of discharge to a SNF were lower among eligiblepatients hospitalized in post-STRIDE time periods (odds ratio[OR], 0.6 [95% CI, 0.5 to 0.8]) compared with pre-STRIDE.Findings were robust to sensitivity analyses. There were nodifferences in LOS (rate ratio,1.0 [CI, 0.9 to 1.1]) or havingan inpatient fall (OR,0.8 [CI, 0.5 to 1.1]). Limitation: Direct program reach was low. Conclusion: Although the reach was limited and variable,hospitalizations occurring during the STRIDE hospital walk-ing program implementation period had lower odds of dis-charge to a SNF, with no change in hospital LOS or inpatientfalls. Primary Funding Source: U.S. Department of VeteransAffairs Quality Enhancement Research Initiative (OptimizingFunction and Independence QUERI).
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.