Publications

BACKGROUND: OBJECTIVE: To characterize hospital phenotypes with respect to trends in observation care use and examine whether patterns differ for people with AD/ADRD.

DESIGN: Retrospective cohort study.

SUBJECTS: Traditional Medicare beneficiaries aged 68 + who visited an emergency department from 2012 to 2019.

MAIN MEASURES: For each beneficiary, we determined visit disposition (e.g., discharge, inpatient admission, or observation stay). We calculated hospital-level slopes for each disposition using linear regression and categorized each hospital as having an increase, decrease, or no change in each disposition category. Cross-tabulations of these trends determined hospital phenotypes (e.g., increasing observation stays along with decreasing admissions). We compared phenotypes by hospital characteristics and repeated these analyses stratified by AD/ADRD diagnosis.

KEY RESULTS: Sample included 22,780,334 ED visits among 5,162,037 beneficiaries at 4835 hospitals. Nationally, the percentage of visits ending in observation increased over time, but there was a substantial decrease in observation stays among 17.5% of hospitals, and 41.4% saw no change. There were 68.8% of hospitals with identifiable phenotypes of observation use trends, the most common of which (N = 788 hospitals, 16.3%) was an increase in the rate of observation stays and a decrease in the rate of admissions. Large, teaching, and urban hospitals saw a disproportionate increase in observation use, while small, non-teaching rural and critical access hospitals saw no meaningful change. A greater share of hospitals saw an increase in observation for beneficiaries with AD/ADRD compared to those without AD/ADRD (49.0% vs. 40.0%).

CONCLUSIONS: While there has been a large national increase in observation use for Medicare beneficiaries, trends among individual hospitals varied substantially, with differential patterns by hospital characteristics and beneficiary AD/ADRD status.

In their previously published article in the Israel Journal of Health Policy Research, Rose and colleagues describe and advocate for greater use of implementation science in Israel. As a discipline, implementation science seeks to traverse the last steps in bringing new science from research to clinical practice, which are often the most difficult of the entire process. Implementation science in general faces substantial challenges, including the extraordinary heterogeneity of the dissemination process, and the obstacles represented by established practices, singular preferences, and questions about generalizability. In our view, implementation science complements classic epidemiology as part of a continuum of population health research that warrants greater attention and funding. For now, however, implementation science will need to show that it can consistently achieve sizable, durable, and widespread results if it is to traverse its own last mile and establish itself as a successful and permanent component of biomedicine in Israel.

OBJECTIVE: Accidental falls are the leading cause of injury for older adults in the United States. Identifying risk factors for falls is a public health priority. Poor sleep is prevalent among aging adults and has been linked to falls risk. We examined late-life sleep medication use and falls risk in a cohort of older adults.

METHODS: The Atherosclerosis Risk in Communities (ARIC) study is an ongoing community-based cohort study. ARIC participants taking any barbiturates, benzodiazepines, antidepressants, non-benzodiazepine receptor agonists, or other hypnotics in the past 4 weeks (2011-2013) were categorized as taking a medication that affects sleep, regardless of indication. Participant hospital discharge records were reviewed through 2019 for ICD codes indicating incident falls. Propensity score matching was used to match participants who used sleep medications with those who did not (1:2). Cox proportional hazards regression models were used to assess the association of sleep medication use with falls with adjustment for demographics, lifestyle, and health characteristics.

RESULTS: In the matched sample (N = 4794; 70% female; mean age 75.5 ± 5 years), 1200 documented falls occurred over 6.5 years of follow-up. In fully adjusted models, sleep medication use was associated with a 33% greater risk of falls compared to nonuse (HR: 1.33; 95% CI: 1.18-1.51). Results did not differ by age, sex, depressive symptoms, baseline cognitive status, or physical functioning status (interaction p-values >.05).

CONCLUSIONS: Late-life sleep medication use is associated with a higher risk of falls. Further research is needed to clarify the mechanisms linking sleep medications to falls risk.

Albuminuria reflects systemic endothelial dysfunction, but its relationships with subclinical brain abnormalities have not been comprehensively catalogued. The Cardiovascular Health Study recruited older adults from four US communities, beginning in 1989-1990. Systematic measurements of albuminuria were performed in 1996-1997; two brain MRIs, in 1992-1994 and 1997-1999; and serum neurofilament light chain (NfL) measurements from 1996-1997 stored samples. We examined the associations of albuminuria with longitudinal progression of white matter hyperintensities (WMH) and ventricular size, incident infarcts, and cross-sectional quantitative brain volumes and circulating biomarkers of neuronal injury (n = 834-1950). Albuminuria was positively associated with ventricular grade worsening (odds ratio per doubling 1.10, 95% confidence interval (CI) 1.01-1.19) and with circulating NfL levels (2% higher per doubling, 95% CI 1-4%), even after adjustment for vascular risk factors. Albuminuria was also associated with worsening of WMH, incident infarcts, and quantitative WMH and hippocampal volumes, but these latter associations appeared to reflect burden of cardiovascular risk factors. Albuminuria is independently associated with worsening ventricular size and circulating NfL, suggesting a specific role of microvascular dysfunction in brain atrophy. It also reflects cardiovascular risk factor burden on markers of vascular brain injury. These results highlight the diverse associations of albuminuria with common brain abnormalities of aging.

UNLABELLED: In 2023 the Centers for Medicare and Medicaid Services (CMS) expanded coverage for transfemoral carotid artery stenting (tfCAS) to include standard-risk patients. Given this shift, we sought to identify predictors of postoperative stroke in patients undergoing tfCAS.

METHODS: We analyzed VQI data from 2011-2024, identifying tfCAS patients treated for atherosclerosis or restenosis. We excluded patients with interventions outside the internal carotid artery (ICA) or bifurcation and procedures with concurrent intracranial treatment. We then stratified patients based on in-hospital postoperative stroke status. Baseline characteristics and outcomes were compared using X2 tests, and predictors of stroke risk were compared using logistic regression. Ten-fold cross-validation was used to identify predictors of stroke, and a point system was developed to predict the risk of in-hospital postoperative stroke.

RESULTS: In our cohort of 35,346 tfCAS cases, 693 (2%) patients had an in-hospital post-procedure stroke. Patients who had a post-procedure stroke were older, more likely to be female, non-white, and had a higher overall comorbidity burden. After 10-fold cross-validation using the original logistic regression model, the factors most strongly associated with increased odds of stroke included uncontrolled hypertension (23% vs. 14%, adjusted odds ratio (aOR) 2.66 [95% confidence interval (CI) 1.94-3.71] p<.001), emergent surgery (41% vs. 32%, aOR 2.39,[1.74-3.27] p<.001), severe distal tortuosity (13% vs. 8%, aOR 1.62 [1.25-2.27] p<.001), age greater than 75 (41% vs. 32%, aOR1.53[1.23-1.92] p<.001), preoperative stroke (44% vs. 29%, aOR 1.57[1.30-1.90] p<.001), prior CEA (14% vs. 12%, aOR 1.44 [1.14-1.81] p=.002 ), Type III arch (7.1% vs 5.5%, aOR 1.45 [1.02-2.03] p=.03) and circumferential or protruding calcification (3.3% vs 1.7%, aOR 1.43 [1.04-1.93] p=.02). Dual antiplatelet therapy (DAPT) and having a high-volume physician or center were associated with lower odds of stroke. Based on variable importance, a simplified point system was developed to predict postoperative stroke (ROC AUC = 0.68).

CONCLUSION: The identified mediators of stroke risk after tfCAS offer the opportunity to improve patient selection. These data highlight the importance of managing modifiable factors preoperatively, such as hypertension and antiplatelet therapy. The timing of surgery also emerged as a strong predictor of stroke, suggesting the need for careful consideration of the need for emergent or urgent surgery. Anatomic considerations such as vessel tortuosity, arch type and calcification should also prompt consideration for an alternate revascularization strategy.

BACKGROUND: In Haiphong, Vietnam, most hepatitis C virus (HCV) infections occur among people who inject drugs (PWID). As part of multiple respondent-driven sampling (RDS) surveys among PWID in Haiphong, an intervention (DRIVE-C) provided HCV testing and treatment in 2019. Centres providing opiate agonist treatment (OAT) or antiretroviral therapy (ART) also provided HCV testing and linkage-to-treatment in 2021/22. We modelled the impact and cost-effectiveness of HCV testing and treatment for PWID in Haiphong.

METHODS: An HCV transmission model among PWID and former injectors was calibrated in a Bayesian framework using data from Haiphong. A status quo (SQ) scenario modelled past interventions, with no future HCV treatment. A future intervention scenario modelled the impact of providing HCV testing and linkage-to-treatment in OAT and ART centres, and annual RDS survey interventions over 2025-2030, each testing 1400 PWID. We estimated the incremental cost-effectiveness ratio (ICER) per disability adjusted life-year (DALY) averted for the future scenario compared to SQ over 2025-2054 (3 % annual discount rate).

RESULTS: For the SQ scenario, HCV incidence decreased from 8.1 (95 % credibility interval 5.1-13.6) per 100 person-years (/100pyrs) in 2015 to 5.3/100pyrs (3.0-9.6) in 2023 and increases to 6.2/100pyrs (3.5-10.7) in 2030. In the future intervention scenario, incidence decreases to 2.7/100pyrs (1.0-6.4) by 2030. The mean ICER is €884/DALY averted; cost-effective at a willingness-to-pay threshold of €2334 (57 % of Vietnam's 2023 GDP per capita).

CONCLUSIONS: Using RDS surveys and other care settings to scale-up HCV-testing and treatment are cost-effective strategies to reduce HCV incidence among PWID in Vietnam.

The vast majority of placebos are administered with concealment or deception. Most clinicians and researchers believe that either deception or concealment is necessary to achieve a placebo response; however, recent studies with open-label placebo (OLP), conditioning, and conditioning + OLP (COLP) have shown that placebos can be effective even when patients know they are receiving a placebo. To date, no studies have examined COLP in a pediatric surgical population, which is vulnerable to developing opioid dependence. COLP may assist in lowering opioid doses administered and quicker tapering off opioids. This study investigates the effectiveness of COLP on postoperative treatment of patients ages 10 through 17 undergoing posterior fusion for adolescent idiopathic scoliosis (AIS). It is a randomized controlled trial of 64 AIS patients assigned to one of the two arms: COLP + treatment as usual (TAU) and TAU control. Randomization and baseline assessments occur at the preoperative visit. Opioid consumption is recorded weekly after hospitalization until the first in-person postoperative visit at 6 weeks. The primary outcome measure is postoperative opioid consumption, both amount and duration. Secondary outcomes include pain, functional ability, and mental health scores.