Andrew J. Epstein
VerifiedUniversity of Pennsylvania · Rehabilitation Medicine
Active 1959–2026
Research topics
- Medicine
- Internal medicine
- Materials science
- Emergency medicine
- Family medicine
Selected publications
Abstract LB130: High adherence to mammography screening before and after MCED testing
Cancer Research · 2026-04-17
articleAbstract Background: Multi-cancer early detection (MCED) tests have the potential to detect cancers that do not have screening options. However, there is concern that implementation of MCED tests could lead to a lack of adherence to recommended single cancer screening. We examined mammography screening rates before and after MCED testing in a cohort of individuals eligible for breast cancer screening. Methods: Deidentified data for a US cohort of women with a negative (“no cancer signal detected”) MCED (Galleri®, GRAIL, Inc.) test administered during a 1-year ascertainment period (7/01/2022-6/30/2023) were linked via tokenization to the Komodo Research Dataset. This analysis focused on women aged 42-72 years with no prior Galleri test, no positive Galleri test result during ascertainment, and with closed health insurance claims coverage on the index test date. A final study sample requiring ≥24 months of continuous coverage before and after the index test was created to capture women eligible for two complete cycles of mammography screening across the observation period. Diagnostic and screening mammograms were identified by billing codes on insurance claim records. Descriptive statistics of test recipients’ demographic characteristics were calculated by mammogram receipt before and/or after the index MCED test. Results: Of the 6,122 women in the eligible population, 2,820 (46.1%) were in the final study sample. Women in the study sample had similar demographics as the eligible population, with mean age of 57.3 years, 61.9% white non-Hispanic, and 80.4% commercial health insurance. 2,317 (82.2%) and 2,357 (83.6%) had mammograms in the 24-month pre- and post-index periods, respectively. 2,108 (74.8%) women had mammograms in both periods, while 254 (9.0%) did not have a mammogram in either period (Table). Conclusions: Women maintained high adherence to mammography at 24 months following receipt of a negative MCED test result, with >80% of women receiving mammograms in that period. Citation Format: Charles H. McDonnell, Gretchen M. Stipec, Arushi Chadha, Anuraag R. Kansal, Claire O'Connell, Alice Kate Cummings Joyner, Andrew J. Epstein, Eric T. Fung, A. Mark Fendrick. High adherence to mammography screening before and after MCED testing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr LB130.
Determinants of Unmet Demand for Surgery: The Case of Transcatheter Aortic Valve Replacement
Value in Health · 2026-02-01
articleOpen accessOBJECTIVES: To examine the determinants of unmet transcatheter aortic valve replacement (TAVR) needs and their impact on patient survival among Medicare beneficiaries with aortic stenosis. METHODS: We developed a county-level mismatch score measuring the gap between actual TAVR procedures performed and expected need based on population differences. Counties were classified as metropolitan, semiurban, or rural. Factors associated with larger mismatches were identified, and mortality rates among aortic stenosis (AS) patients were examined in relation to mismatch scores. We analyzed Medicare data from 2016 to 2022 across 3129 US counties. The mismatch score was developed to account for population differences and county urbanicity classification. Statistical analyses identified factors associated with TAVR mismatch and its relationship to mortality outcomes. RESULTS: We found substantial geographic variation in TAVR delivery. Counties with higher TAVR mismatch scores showed associations with fewer TAVR-providing hospitals, less market concentration, higher AS prevalence, and lower household incomes. Counties with greater gaps between needed and actual TAVR procedures were also associated with higher mortality rates. This relationship between mismatch and mortality was particularly strong in semiurban counties. CONCLUSIONS: Our findings identify associations between TAVR access gaps and patient outcomes, as well as factors linked to these access patterns. Counties with higher TAVR mismatch scores showed correlations with healthcare capacity constraints, geographic location, and socioeconomic factors. These associations suggest that mismatches may be addressed through targeted approaches based on local needs to improve care delivery for patients with AS in regions currently experiencing access challenges.
Diabetes Obesity and Metabolism · 2026-03-08
articleCorrespondingKaren Chung and Brian Lian are employees of Viking Therapeutics, Inc. Andrew Epstein is an employee of, at the time of the study Yilmaz Akkas was an employee of, and Saif Rathore is an adviser to Medicus Economics, which received funding from Viking Therapeutics. This study analysed data from the National Health and Nutrition Examination Survey, which is publicly available from the US Centers for Disease Control and Prevention. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.70628.
Quantifying the Impact of Real-World Evidence: The Sacubitril/Valsartan Experience
Archives of Clinical and Biomedical Research · 2025-01-01
articleOpen access1st authorCorrespondingProjected Impact of Glucagon-Like Peptide-1 Adoption on 5-Year Coronary Stent Use
SSRN Electronic Journal · 2025-01-01
preprintOpen accessJournal of Clinical Gastroenterology · 2025-04-29 · 3 citations
articleOpen accessGOALS: To estimate health care resource utilization (HCRU) and costs among patients with eosinophilic esophagitis (EoE) in the US. BACKGROUND: The EoE prevalence in the US has risen in recent years. Assessing HCRU and costs may assist in understanding the economic burden of EoE in the US. STUDY: In IQVIA's PharMetrics Plus claims database, prevalent patients with EoE were identified and matched with non-EoE controls. The index date was a randomly selected EoE diagnosis date for the EoE cohort (January 2018 to June 2019) and a random date for non-EoE controls. Patients had 1-year of continuous enrollment before and after the index date, with ≥1 EoE diagnostic claim before the index. Descriptive and regression analyses adjusting for comorbidities unrelated to EoE were performed to compare HCRU and costs (EoE vs. non-EoE), 1-year after the index date. RESULTS: The analysis included 15,432 patients with EoE and matched non-EoE controls (mean age: 36.2 y). The annual HCRU, including the mean outpatient visits, was higher in patients with EoE versus non-EoE controls [mean difference (MD): 9.2 d; 95% CI: 8.8-9.6]; with consistent results across age groups. The mean total health care costs (annual) were ∼2.5 times higher in patients with EoE than in non-EoE controls. Patients with EoE who underwent esophageal dilation had higher HCRU (emergency room visit: MD: 0.9 d; 95% CI: 0.8-1.0) and total health care costs (MD: $10,174; $8493-$11,855) than non-EoE controls. CONCLUSIONS: Patients with EoE had higher annual costs and HCRU than non-EoE controls, indicating a substantial economic burden, particularly among patients with EoE with prior esophageal dilation.
The Journal of Urology · 2025-04-08
articleSenior authorJournal of the American Medical Directors Association · 2025-10-10
articleOpen accessOBJECTIVES: Over the past decade, US nursing home workforce experienced considerable growth among advanced practitioners (APs), particularly nurse practitioners. We measured the association between workforce composition and rehospitalization from post-acute care in the nursing home. DESIGN: Cohort study using Medicare claims for 2012 and 2019. SETTING AND PARTICIPANTS: A total of 9171 nursing homes which provided post-acute care to Medicare fee-for-service beneficiaries in both 2012 and 2019. Hospital-based facilities were excluded. METHODS: The change in workforce composition over the study period was measured by calculating the percentages of visits performed by APs and physicians in generalist specialties and categorized into 5 groups: (1) any decrease in the percent of visits by Aps, (2) stable or small [<5 percentage point (ppt)] increase, (3) 5 to 40 ppt increase, (4) 41 to 70 ppt increase, and (5) >70 ppt increase in the proportion of visits performed by APs. Unplanned 30-day rehospitalization rates from post-acute care for each facility were measured using Medicare's quality measure specification. We estimated first-difference linear regression models to compare the rehospitalization rates between the 5 workforce composition categories, adjusting for nursing home characteristics. RESULTS: Most facilities increased the proportion of care provided by APs: 23% experienced a decrease (n = 2127), 19% remained stable (n = 1710), 33% increased by 5 to 40 ppt (n = 3030), 17% increased by 41 to 70 ppt (n = 1529), and 8% increased by >70 ppt (n = 775). Rehospitalization decreased in nursing homes with a relative increase in AP visits, but the association was statistically significant only for group 3 (5-40 ppt increase; 1.1% decrease in rehospitalizations; 95% CI, -1.8% to -0.4%; P = .003). CONCLUSIONS AND IMPLICATIONS: In this cohort study, a moderate shift in visits to APs was associated with a small but statistically significant decrease in rehospitalizations between 2012 and 2019. These findings support incorporation of APs into the nursing home workforce.
Journal of Managed Care & Specialty Pharmacy · 2025-12-04
articleBACKGROUND: Chronic kidney disease (CKD) is common in older adults and is often associated with type 2 diabetes (T2DM) and heart failure (HF). However, little is known about the burden of newly diagnosed CKD in Medicare Fee-for-Service (FFS) beneficiaries, including those with comorbid T2DM or HF. OBJECTIVE: To quantify the clinical and economic burden of CKD in Medicare FFS beneficiaries, including those with comorbid T2DM or HF. METHODS: In this retrospective cohort study using 100% Medicare FFS claims data (Parts A, B, D) from 2014 to 2022, beneficiaries with incident CKD (based on a diagnosis code on 2 distinct dates) from January 1, 2015, to December 31, 2021, were included; index date was the date of earliest CKD diagnosis. Beneficiaries with a diagnosis of CKD, acute kidney injury, dialysis, kidney transplantation, or a claim for any condition other than T2DM that could cause kidney disease during a 365-day baseline period prior to index date were excluded. Beneficiaries with CKD were categorized into 4 mutually exclusive cohorts: CKD-only; CKD+HF; CKD+T2DM; and CKD+HF+T2DM based on claims during the baseline period. Clinical burden was measured as prevalence at baseline and incidence of key clinical outcomes at 12 months of follow-up. Economic burden was measured as all-cause and CKD-related health care resource utilization (HCRU) and inflation-adjusted costs in the baseline period and at 12 months of follow-up. RESULTS: < 0.001 for comparison across cohorts). Relative spending across cohorts was similar for CKD-related total costs at 12 months. CONCLUSIONS: Older adults with a new diagnosis of CKD experienced considerable clinical and economic burden, and presence of T2DM and HF was associated with larger burden. All-cause mean total costs at 12 months after a new diagnosis of CKD ranged from $24,180 for the CKD-only cohort to $54,477 for the CKD+HF+T2DM cohort.
Advances in Therapy · 2025-02-13 · 4 citations
articleOpen accessINTRODUCTION: This study investigates the impact of geographic and socioeconomic barriers on access to transcatheter aortic valve replacement (TAVR). METHODS: Utilizing Medicare data from the US Centers for Medicare and Medicaid Services, this study analyzed TAVR and surgical aortic valve replacement (SAVR) procedures among beneficiaries from 2017 to 2022. Geographic units were defined by 5-digit zip codes, categorized on the basis of TAVR/SAVR volume into four categories: (1) no TAVR or SAVR, (2) no-TAVR zone (SAVR present, no TAVR), (3) low-TAVR zone (TAVR/SAVR ratio ≤ 0.5), and (4) TAVR accessible (TAVR/SAVR ratio > 0.5). The differential distance index (DDI) was developed to measure travel hurdles, calculated as the difference in miles from a patient's zip code center to the treatment hospital (TAVR versus SAVR, CABG (coronary artery bypass grafting), and PCI (percutaneous coronary intervention) comparators). This study maintained a continuous access variable to model outcomes such as the ratio or volume of TAVR/SAVR and the percentage share of TAVR/AVR within each zip code over biennial periods (2017-2018, 2019-2020, 2021-2022). Covariates in the model included population density, area deprivation index (ADI), and calendar time, with an exploration of the interaction between DDI and ADI. RESULTS: The analysis revealed significant geographic disparities in TAVR access across the USA, with no-TAVR zone and low-TAVR zone areas often featuring lower population densities, higher ADIs, and more rural settings. Increased travel distance (DDI) significantly correlated with lower TAVR utilization, emphasizing distance as a critical barrier. Furthermore, both ADI and DDI emerged as significant predictors of TAVR volume and share, underlining the compound effect of socioeconomic status and geographic distance on healthcare access. CONCLUSIONS: This study highlights the critical role of geographic and socioeconomic barriers in accessing advanced medical treatments like TAVR. Addressing these barriers may ensure equitable healthcare distribution, guiding policymakers and providers towards more accessible healthcare solutions for all populations.
Frequent coauthors
- 202 shared
Harlan M. Krumholz
Yale New Haven Health System
- 182 shared
Yongfei Wang
Center for Outcomes Research and Clinical Epidemiology
- 182 shared
Joseph S. Ross
Yale University
- 115 shared
Jeph Herrin
Yale University
- 104 shared
Brahmajee K. Nallamothu
- 103 shared
Frederick A. Masoudi
Ascension
- 102 shared
Vivian Ho
Rice University
- 100 shared
Sherin Stephen
Education
- 2004
PhD, Health Care Management, Wharton School of Business
University of Pennsylvania
- 1998
MPP, Public Policy
Georgetown University
- 1994
BA, Computer Science
Amherst College
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Andrew J. Epstein
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup