Fingerstick glucose monitoring by cognitive impairment status in Veterans Affairs nursing home residents with diabetes
Abstract: Guidelines recommend nursing home (NH) residents with cognitive impairment receive less intensive glycemic treatment and less frequent fingerstick monitoring. Our objective was to determine whether current practice aligns with guideline recommendations by examining fingerstick frequency in Veterans Affairs (VA) NH residents with diabetes across cognitive impairment levels. Methods: We identified VA NH residents with diabetes aged ≥65 residing in VA NHs for >30 days between 2016 and 2019. Residents were grouped by cognitive impairment status based on the Cognitive Function Scale: cognitively intact, mild impairment, moderate impairment, and severe impairment. We also categorized residents into mutually exclusive glucose‐lowering medication (GLM) categories: (1) no GLMs, (2) metformin only, (3) sulfonylureas/other GLMs (+/− metformin but no insulin), (4) long‐acting insulin (+/− oral/other GLMs but no short‐acting insulin), and (5) any short‐acting insulin. Our outcome was mean daily fingersticks on day 31 of NH admission. Among 13,637 NH residents, mean age was 75 years and mean hemoglobin A1c was 7.0%. The percentage of NH residents on short‐acting insulin varied by cognitive status from 22.7% in residents with severe cognitive impairment to 33.9% in residents who were cognitively intact. Mean daily fingersticks overall on day 31 was 1.50 (standard deviation = 1.73). There was a greater range in mean fingersticks across GLM categories compared to cognitive status. Fingersticks ranged widely across GLM categories from 0.39 per day (no GLMs) to 3.08 (short‐acting insulin), while fingersticks ranged slightly across levels of cognitive impairment from 1.11 (severe cognitive impairment) to 1.59 (cognitively intact). NH residents receive frequent fingersticks regardless of level of cognitive impairment, suggesting that cognitive status is a minor consideration in monitoring decisions. Future studies should determine whether decreasing fingersticks in NH residents with moderate/severe cognitive impairment can reduce burdens without compromising safety.
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.