Publications

OBJECTIVE: Many post-acute and long-term care settings (PALTCs) struggle to measure antibiotic use via the standard metric, days of therapy (DOT) per 1000 days of care (DOC). Our objective was to develop antibiotic use metrics more tailored to PALTCs.

DESIGN: Retrospective cohort study with a validation cohort.

SETTING: PALTC settings within the same network.

METHODS: We obtained census data and pharmacy dispensing data for 13 community PALTCs (January 2020-December 2023). We calculated antibiotic DOT/1000 DOC, DOT per unique residents, and antibiotic starts per unique residents, at monthly intervals for community PALTCs. The validation cohort was 135 Veterans Affairs Community Living Centers (VA CLCs). For community PALTCs only, we determined the DOT and antibiotics starts per unique residents cared for by individual prescribers.

RESULTS: For community PALTCs, the correlation between facility-level antibiotic DOT/1000 DOC and antibiotic DOT/unique residents and antibiotic courses/unique residents was 0.97 (P < 0.0001) and 0.84 (P < 0.0001), respectively. For VA CLCs, those values were 0.96 (P < 0.0001) and 0.85 (P < 0.0001), respectively. At community PALTCs, both novel metrics permitted assessment and comparison of antibiotic prescribing among practitioners.

CONCLUSION: At the facility level, the novel metric antibiotic DOT/unique residents demonstrated strong correlation with the standard metric. In addition to supporting tracking and reporting of antibiotic use among PALTCs, antibiotic DOT/unique residents permits visualization of the antibiotic prescribing rates among individual practitioners, and thus peer comparison, which in turn can lead to actionable feedback that helps improve antibiotic use in the care of PALTC residents.

BACKGROUND: Urinary tract infections (UTI) are common complications in people with neurogenic bladder (NB). Limited data exist on UTI-related knowledge, experiences, and quality of life (QoL) impacts in this population.

METHODS: We mailed surveys to 289 Veterans with NB due to spinal cord injury/disorder (SCI/D), multiple sclerosis, or Parkinson's Disease who had a UTI diagnosis at four Veterans Affairs Medical Centers between May 2022-May 2023. The survey was adapted from existing instruments and previously collected qualitative data and assessed patient knowledge and experiences with UTI and QoL impacts. Descriptive statistics summarized responses and scaled QoL scores were calculated, with higher scores indicating greater negative impact.

RESULTS: Most respondents (n = 71) were older (mean age = 69), had SCI/D (77%), and used urinary catheters (77%). Over a third had misperceptions about antibiotic risks and the significance of a positive urine culture or bacteriuria for diagnosing UTI. 18% wanted more information about UTIs, with most preferring written materials (77%) or information at healthcare provider visits (62%). The strongest QoL impacts were on daily activities, with many respondents indicating UTIs affect diet (50%), travel (53%), and sex life (60%). Mean [standard deviation (SD)] scaled QoL score was 40.8 (15.3) out of a maximum of 75, with ≥ 3 UTIs in the prior year associated with higher scores (p = 0.02).

CONCLUSIONS: People with NB may have misperceptions about UTI diagnosis and antibiotic risks, and experience substantial QoL impacts from UTIs. Provider encounters for suspected UTI may be good opportunities for delivering written education and assessing QoL impacts.

BACKGROUND: One of the largest-ever retail drug shortages began in 2018 when several angiotensin II receptor blockers (ARBs) for treating hypertension, heart failure, and chronic kidney disease-valsartan, losartan, and irbesartan-were recalled for carcinogenic impurities. The long-term consequences of the ARB shortages and whether certain groups experienced more adverse outcomes is unknown.

OBJECTIVE: To evaluate changes in adherence and health outcomes after ARB recalls and to identify patients who experienced greater changes in access and adverse clinical outcomes.

METHODS: Using an integrated claims and electronic health record dataset and a difference-in-differences design, we evaluated changes in the proportion of days covered (PDC) for ARBs and similar drugs (angiotensin-converting enzyme inhibitors [ACE-Is]), uncontrolled blood pressure, major cardiovascular event (MACE)-related acute care visits, and all-cause ambulatory care visits in the 12 months before vs 18 months after recalls for valsartan, losartan, and irbesartan users vs patients taking similar, nonrecalled drugs (ACE-Is, nonrecalled ARBs). Triple-difference models characterized heterogeneous associations by pre-recall patient demographic (race, ethnicity, age), clinical (baseline indication, mental health conditions), and adherence variables.

RESULTS: Adjusting for pre-recall patient characteristics, we observed no significant changes in PDC for ARBs and ACE-Is (combined), uncontrolled blood pressure, or ambulatory care visits among 86,507 recalled ARB users vs 123,583 comparison drug users in the 18 months after the recalls. Following the recalls, medication switches increased on average by an additional 2.08 percentage points (p.p.) per quarter (95% CI = 2.01-2.15) for recalled ARB vs comparison drug users, a 195.9% relative increase. We observed the most switches in the 90-day period immediately after valsartan's recall (difference-in-difference: 9.48 p.p.; 95% CI = 9.36-9.59; relative change = 892%). Cumulatively, 55.2% of valsartan, 7.6% of losartan, and 18.9% of irbesartan users switched medications after 18 months. We observed an increase in the proportion of recalled ARB vs comparison patients who experienced medication gaps exceeding 30 days (1.13 p.p. per quarter on average; 95% CI = 0.97-1.30), which was most apparent after approximately 15 months (5 quarters). Although MACE-related acute care visits did not change in the quarter (90 days) immediately after valsartan's recall, we observed an increase of 1.40 additional visits per 1,000 recalled ARB vs comparison drug patients in each subsequent quarter, a 9.3% relative increase. Results were similar across most subgroups.

CONCLUSIONS: The 2018 ARB recalls were associated with immediate changes in antihypertension medication use. Many patients transitioned to alternative medications. Although overall impacts on clinical outcomes were minimal and not statistically significant, small increases in medication gaps and MACE-related acute care visits among some patients occurred after more than 1 year. The ARB recalls may have been associated with fewer adverse events than other recent shortages owing to the widespread availability of alternative treatments in the same or similar drug class.

PURPOSE: We compare trends in gabapentinoid and opioid utilization overall and by economic development category. We also sought to predict future trends and assess correlations in gabapentinoid and opioid utilization.

METHODS: We conducted a repeated cross-sectional analysis of retail prescriptions for 72 countries from Q1 2012 to Q3 2023. We measured standardized units/1000 population for gabapentinoid and opioid sales, stratified by development category, and used time-series models to predict trends for the following 3 years. Granger causality tests examined predictive relationships between gabapentinoid and opioid sales.

RESULTS: Global gabapentinoid annual sales rose by 114.5% from 2012 to 2022, with a higher increase in developing (180.9%) than developed economies (110.0%). In contrast, annual opioid sales declined globally by 25.4%, with a 27.9% decrease in developed and a 16.8% increase in developing economies. Assuming current trends persist over the following 3 years, gabapentinoid quarterly sales are forecasted to rise by 7.7% in developed and 18.6% in developing economies, while opioid quarterly sales are expected to decrease by 9.5% and increase by 15.1%, respectively. Granger causality tests indicated that gabapentinoids may predict opioid sales globally for the following year, but opioids did not predict gabapentinoid sales.

CONCLUSION: We evaluated the global trends in gabapentinoid and opioid sales, suggesting important differences in pain management practices across developed and developing economies. Our findings highlight the need to ensure the safe use of gabapentinoids and opioids while balancing proper pain management.

Predicting major bleeding in nonvalvular atrial fibrillation (AF) patients on direct oral anticoagulants (DOACs) is crucial for personalized care. Alternatives like left atrial appendage closure devices lower stroke risk with fewer nonprocedural bleeds. This study compares machine learning (ML) models with conventional bleeding risk scores (HAS-BLED, ORBIT, and ATRIA) for predicting bleeding events requiring hospitalization in AF patients on DOACs at their index cardiologist visit. This retrospective cohort study used electronic health records from 2010 to 2022 at the University of Pittsburgh Medical Center. It included 24,468 nonvalvular AF patients (age ≥18) on DOACs, excluding those with prior significant bleeding or warfarin use. The primary outcome was hospitalization for bleeding within one year, with follow-up at one, two, and five years. ML algorithms (logistic regression, classification trees, random forest, XGBoost, k-nearest neighbor, naïve Bayes) were compared for performance. Of 24,468 patients, 553 (2.3%) had bleeding within one year, 829 (3.5%) within two years, and 1,292 (5.8%) within five years. ML models outperformed HAS-BLED, ATRIA, and ORBIT in 1-year predictions. The random forest model achieved an AUC of 0.76 (0.70 to 0.81), G-Mean of 0.67, and net reclassification index of 0.14 compared to HAS-BLED's AUC of 0.57 (p < 0.001). ML models showed superior results across all timepoints and for hemorrhagic stroke. SHAP analysis identified new risk factors, including BMI, cholesterol profile, and insurance type. In conclusion, ML models demonstrated improved performance to conventional bleeding risk scores and uncovered novel risk factors, offering potential for more personalized bleeding risk assessment in AF patients on DOACs.

CONTEXT: In response to the opioid crisis, federal guidelines were implemented, including the Veterans Health Administration's (VA) Opioid Safety Initiative in 2013. The impact of policies on patients near the end of life is unknown.

OBJECTIVES: Examine temporal trends in opioid prescribing, pain, and opioid overdoses among Veterans near the end of life.

METHODS: Retrospective, time series analysis of VA decedents between October 2009 and September 2018 whose next-of-kin participated in VA's Bereaved Family Survey (BFS). Using multivariate regression to adjust for sociodemographic and clinical covariates, we examined temporal trends in outpatient opioid prescribing, uncontrolled pain based on BFS report, and opioid overdose-related hospitalizations, in the last month of life, overall and by clinical diagnosis (cancer versus non-cancer).

RESULTS: Among 79,409 decedents, mean daily outpatient opioid dose in morphine milligram equivalents in the last month of life decreased from 4.6 mg in 2010 to 2.1 mg in 2018 (adjusted change -0.20 mg/year; P < .001). Opioid overdose-related hospitalization decreased from 0.8% in 2010 to 0.1% in 2018 (adjusted percentage point [PP] change -0.06 PP/year; P < .001). Among the 63,965 Veterans with pain data, the percentage with frequent uncontrolled pain increased from 48.8% in 2010 to 52.2% in 2018 (adjusted PP change +1.37 PP/y; P < .001). Patterns were similar among patients with cancer versus non-cancer conditions.

CONCLUSIONS: Over a time period during which opioid safety initiatives were implemented, opioid prescribing near the end of life decreased, accompanied by decreases in opioid-related hospitalizations but increases in pain. These findings suggest that important tradeoffs may exist between reducing opioid-related serious adverse events and undertreating patient pain in the last month of life. Opioid prescribing guidelines could consider incorporating prognosis into recommendations.

BACKGROUND: Sarcoidosis is an idiopathic multiorgan disease with variable clinical outcomes. Comprehensive analysis of sarcoidosis mortality in US veterans is lacking.

RESEARCH QUESTION: What are the trends in all-cause mortality among US veterans with sarcoidosis, and how are these trends influenced by demographics, Black vs White racial disparities, and geographic variability in relationship to mortality?

STUDY DESIGN AND METHODS: Using Veterans Health Administration (VHA) electronic health records (EHRs), we conducted a population-based retrospective cohort study of adjusted all-cause mortality from 2004 through 2022 among veterans with a diagnosis of sarcoidosis who received care through the VHA. Demographics, region of residence, service branch, tobacco use, and comorbidities were extracted from the EHR. Annual trends in all-cause mortality and patient-level characteristics associated with mortality were examined with multivariable ungrouped Poisson regression. We visualized trends and analyzed state-by-state mortality using the marginal means procedure. In subgroup analysis (2015-2022), we considered the impact of neighborhood-level socioeconomic disparities using the Area Deprivation Index (ADI).

RESULTS: In all, 23,745 veterans received a diagnosis of sarcoidosis between 2004 and 2019 and were followed up through 2022. After adjustment, including age and sex, all-cause mortality increased annually by 4.7% (P < .0001) and was 6.4% higher in Black than White veterans (mortality rate ratio, 1.064; P = .02). A subgroup analysis comparing models with and without ADI adjustment showed no meaningful change in mortality trends. Risk factors for increased all-cause mortality included older age, male sex, Black race, Northeast residence, and lower risk with other service branches. Despite distinct geographical variations in mortality rates, no clear patterns emerged.

INTERPRETATION: Mortality among veterans with sarcoidosis is rising. Differences identified by service branch and higher risk among male Veterans raise questions about differences in environmental exposures. The narrower racial disparities and smaller impact of ADI than in other studies may highlight the role of universal health care access in achieving equitable outcomes.

AIMS: To develop machine-learning algorithms for predicting the risk of a hospitalization or emergency department (ED) visit for opioid use disorder (OUD) (i.e. OUD acute events) in Pennsylvania Medicaid enrollees in the Opioid Use Disorder Centers of Excellence (COE) program and to evaluate the fairness of model performance across racial groups.

METHODS: We studied 20 983 United States Medicaid enrollees aged 18 years or older who had COE visits between April 2019 and March 2021. We applied multivariate logistic regression, least absolute shrinkage and selection operator models, random forests, and eXtreme Gradient Boosting (XGB), to predict OUD acute events following the initial COE visit. Our models included predictors at the system, patient, and regional levels. We assessed model performance using multiple metrics by racial groups. Individuals were divided into a low, medium and high-risk group based on predicted risk scores.

RESULTS: The training (n = 13 990) and testing (n = 6993) samples displayed similar characteristics (mean age 38.1 ± 9.3 years, 58% male, 80% White enrollees) with 4% experiencing OUD acute events at baseline. XGB demonstrated the best prediction performance (C-statistic = 76.6% [95% confidence interval = 75.6%-77.7%] vs. 72.8%-74.7% for other methods). At the balanced cutoff, XGB achieved a sensitivity of 68.2%, specificity of 70.0%, and positive predictive value of 8.3%. The XGB model classified the testing sample into high-risk (6%), medium-risk (30%), and low-risk (63%) groups. In the high-risk group, 40.7% had OUD acute events vs. 16.5% and 5.0% in the medium- and low-risk groups. The high- and medium-risk groups captured 44% and 26% of individuals with OUD events. The XGB model exhibited lower false negative rates and higher false positive rates in racial/ethnic minority groups than White enrollees.

CONCLUSIONS: New machine-learning algorithms perform well to predict risks of opioid use disorder (OUD) acute care use among United States Medicaid enrollees and improve fairness of prediction across racial and ethnic groups compared with previous OUD-related models.