About

Amol S. Navathe, MD, PhD, is a Professor of Medical Ethics and Health Policy at the University of Pennsylvania. He is also a Senior Fellow at the Leonard Davis Institute of Health Economics and serves as Co-Director of the Healthcare Transformation Institute at the University of Pennsylvania, as well as Co-Director of the Healthcare Transformation Institute at the Perelman School of Medicine. Additionally, he is the Founding Director of The Parity Center: Equity through Payment Reform at the Perelman School of Medicine. His educational background includes a B.S. in Electrical Engineering & Economic Systems from Stanford University, graduating Magna Cum Laude in 2001; a Ph.D. in Health Care Management and Economics from The Wharton School, University of Pennsylvania, in 2008; and an M.D. from the University of Pennsylvania Perelman School of Medicine in 2010. His research focuses on health policy, healthcare payment reform, health equity, and the application of economics and ethics to medical decision-making and health systems. He has contributed to understanding payment models, hospital participation in bundled payments, virtual specialty palliative care, and addressing racial bias in clinical algorithms, among other topics.

Research topics

• Business
• Actuarial science
• Medicine
• Finance
• Economics
• Public economics
• Environmental health
• Economic growth
• Political Science
• Family medicine
• Computer Science
• Nursing
• Statistics
• Internal medicine
• Psychology
• Marketing

Selected publications

• Nudges to Clinicians and Patients for Influenza Vaccines During Visits

  JAMA Internal Medicine · 2026-01-05 · 3 citations

  articleOpen accessSenior author

  Importance: Annual influenza vaccination reduces burden of disease for older adults, but rates remain suboptimal. Objective: To evaluate if a multicomponent nudge intervention to clinicians and patients increases vaccine completion during primary care visits. Design, Setting, and Participants: The pragmatic BE IMMUNE (Behavioral Economics to Improve and Motivate Vaccination in Primary Care Using Nudges Through the Electronic Health Record) randomized clinical trial took place across 48 primary care clinics through Penn Medicine (Philadelphia, Pennsylvania) and UW Medicine (Seattle, Washington). Patients 50 years and older who were scheduled for a primary care visit and were due for an influenza vaccine within the active intervention period (September 25, 2023, to February 20, 2024) were included. Interventions: Clinics were randomized in a 2:1 ratio to receive (1) previsit text message reminders to patients, (2) an automatic pended order, and (3) monthly comparisons of panel vaccination rates to peer clinicians, or usual care. Additionally, patients in the intervention arm who were identified as high risk for noncompletion were individually randomized 1:1 to receive previsit bidirectional text messaging or a standard text reminder. Main Outcomes and Measures: The primary outcome was influenza vaccination during the visit. Results: Among 80 039 patients across 47 clinics, the mean (SD) age was 65.8 (10.2) years, and 56.0% were female while 43.6% were male. The adjusted odds ratio (AOR) for vaccine completion comparing intervention to usual care was 1.28 (97.5% CI, 1.13-1.45; adjusted P < .001). The probability of completion in the intervention was 31.4% compared to 26.4% under usual care, with a risk difference of 5.1 percentage points (97.5% CI, 2.6-7.5 percentage points; adjusted P < .001). The adjusted odds ratio comparing bidirectional vs standard text messaging among high-risk patients was not statistically significant (1.00; 97.5% CI, 0.98-1.02; adjusted P = .92). Conclusions and Relevance: In this randomized clinical trial, the multicomponent nudge resulted in a statistically significant higher rate of influenza vaccination during the primary care visit, but the bidirectional text messaging did not further increase vaccination in the high-risk group. Trial Registration: ClinicalTrials.gov Identifier: NCT06057727.

  PublisherOA PDFDOI

• Aligning AI Payment Policy With Desired Outcomes Rather Than Inputs May Require Customized Pathways

  Health Affairs · 2026-01-01

  articleSenior author

  Artificial intelligence (AI) has the potential to create health care value independent of traditional inputs, such as clinicians' time, skill, and resources. However, Medicare's current structuring of reimbursement around human inputs has the potential to miscalculate the value of AI in clinical practice. We examine the tension between input-based prices and outcome-based care by comparing and contrasting payment for AI with the approach for prescription drug pricing. We propose a classification system to distinguish between the types of AI that differ in their implications for clinician time and cost. By aligning AI reimbursement policy with desired outcomes rather than inputs, policy makers can ensure that innovators, clinicians, and patients alike benefit from novel AI technologies.

  PublisherDOI

• Bundled Payments—Looking Beyond the Episode

  JAMA Internal Medicine · 2026-04-27

  articleSenior author
  PublisherDOI

• A Machine Learning Model to Improve Risk Adjustment Accuracy in Medicare

  Health Services Research · 2026-03-05

  articleOpen access

  ABSTRACT Objective To develop a machine learning (ML) algorithm that improves accuracy compared to the Hierarchical Condition Category (HCC) score used by the Centers for Medicare and Medicaid Services to risk‐adjust payments for &gt; 65 million Americans. Study Design and Setting Prognostic study using Medicare claims data to train “Franklin”, an ML algorithm predicting one‐year costs, trained using identical data to HCC. Predictive accuracy was evaluated using R 2 log cost, Spearman rho, and sensitivity and specificity. Data Sources and Analytic Sample Random sample of 2018–2019 Part A and B claims from aged, community‐based enrollees in Traditional Medicare who were not dually eligible and did not have end‐stage renal disease. Principal Findings The sample consisted of 4,176,666 Medicare beneficiaries (mean [SD] age 74.9 [7.2] years, 55.9% women; 85.9% Non‐Hispanic white, 5.6% African‐American, 3.4% Hispanic). Franklin was more accurate than HCC ( R 2 log cost 0.44 vs. 0.15; Spearman rho 0.61 vs. 0.41, p &lt; 0.001 for both). Accuracy improved for the 47% of beneficiaries with 0 HCCs and the 27% of beneficiaries with one HCC (Spearman rho 0.59 vs. 0.08 and 0.46 vs. 0.16, respectively; p &lt; 0.001 for both). Franklin outperformed HCC in detecting the 20% lowest‐cost beneficiaries (sensitivity 0.60 vs. 0.34, specificity 0.90 vs. 0.83; p &lt; 0.001 for both). Franklin improved accuracy over HCC for racial/ethnic minorities and rural‐dwelling beneficiaries ( R 2 log cost Black 0.48 vs. 0.14, Hispanic 0.55 vs. 0.09, rural 0.36 v. 0.11; p &lt; 0.001 for all), although Franklin disproportionately classified Black (15.8% vs. 10.1%) and Hispanic (22.9% vs. 12.2%) beneficiaries in the lowest predicted cost decile. Conclusions Franklin is an ML risk adjustment model that significantly improves risk‐adjustment accuracy for Medicare beneficiaries compared to HCC. Franklin could generate improvement in payment accuracy, reduction in selection incentives, and financial savings to Medicare. Clarifying the equity impacts of more accurate risk adjustment is necessary.

  PublisherDOI

• Comparison of Alternative Policy Definitions for Safety-Net Ambulatory Practices

  JAMA Health Forum · 2026-04-17

  articleOpen access

  This cross-sectional study uses Medicare data to compare 2 definitions of safety-net ambulatory practices to investigate the extent to which safety-net practices overlap.

  PublisherDOI

• Transplantation in Mandatory Kidney Payment Models: Understanding the Potential Influence of the ESRD Treatment Choices Model on the Increasing Organ Transplant Access Model

  Kidney Medicine · 2025-11-06

  articleOpen accessSenior author
  PublisherOA PDFDOI

• Aligning incentives: the importance of behavioral economic perspectives in AI adoption

  npj Health Systems · 2025-05-26

  articleOpen access
  PublisherOA PDFDOI

• Variation, Overlap, and Stability in Defining Safety Net Hospitals

  JAMA Network Open · 2025-07-30 · 3 citations

  articleOpen accessSenior author

  Importance: The lack of universally accepted definitions for safety net hospitals (SNHs) has made it difficult to effectively design policies to support these hospitals and the populations they serve. Objective: To evaluate the overlap, variation, and consistency across different definitions for SNH status. Design, Setting, and Participants: This retrospective cohort study used a hospital year-level dataset on short-term acute care US hospitals from 2014 to 2022. Hospital-level and area-level measures were used to define SNHs. Hospital characteristics under each definition, overlap across definitions, and stability of SNH samples produced by each definition from were described. Data analyses were performed from August 2024 to June 2025. Exposure: Nine hospital-level and 4 area-level SNH definitions. Main Outcomes and Measures: Hospital characteristics under each definition, overlap across definitions, and stability of SNH samples over time. Hospital-level definitions included Medicare Disproportionate Share Hospital (DSH) index, Medicare inpatient day share, dual-eligible or low-income subsidy (DLIS) inpatient day share, Medicaid inpatient day share, Medicare Safety-Net Index, teaching status, public ownership, uncompensated care share, and operating margins. Area-level measures included Area Deprivation Index, Social Vulnerability index, proportion Hispanic population, and proportion Black population. Safety net status was assigned based on quartiles defined nationally (or within a state for Medicaid-specific definitions). For a subset of measures, this quartile-based approach was compared between the absolute number of inpatient days attributed to each patient group and the relative number (or share) of inpatient days. Results: Among 4531 short-term acute care hospitals, between 992 (21.9%) and 1326 (29.3%) were SNHs in 2022, depending on definition. SNHs defined based on the absolute level of inpatient days or absolute level of DLIS populations were often large (51% [242 of 476] or 67% [537 of 801]) and were not often rural (9% [45 of 476] or 2% [17 of 801]). Meanwhile, SNHs defined based on relative level of Medicaid inpatient days or relative level of DLIS patients were more often small (63% [298 of 476] and 82% [660 of 801]) and rural (48% [228 of 476] and 69% [555 of 801]) hospitals. The largest overlap across definitions was between a hospital's Medicaid inpatient day share and Medicare DSH index (55% overlap [808 of 1466 hospitals]), which tended to represent large, teaching hospitals. Public ownership, teaching status, and Medicare DSH index produced the most stable definitions of SNHs over time from 2014 to 2022, with 83% (862 of 1043), 74% (1000 of 1354), and 60% (809 of 1358) of similar hospitals, respectively, meeting safety net criteria. The least stable definitions were based on low operating margins, high uncompensated care share, and high DLIS day share, with only 15% (263 of 1796), 20% (362 of 1823), and 25% (436 of 1725) of similar hospitals, respectively, meeting safety net criteria in 2014, 2018, and 2022. Conclusions and Relevance: In this cohort study of US hospitals, different SNH definitions produced different samples, and candidate measures had variable overlap and stability over time. These findings highlight the trade-offs when considering different options to define SNHs.

  PublisherOA PDFDOI

• Savings Associated With Bundled Payments for Outpatient Spine Surgery Among Medicare Beneficiaries

  JAMA Health Forum · 2025-07-11 · 2 citations

  articleOpen accessSenior author

  Importance: Few value-based payment programs have targeted outpatient surgery, although these procedures comprise most surgeries performed in hospitals. In 2018, the Centers for Medicare and Medicaid Services introduced Bundled Payments for Care Improvement Advanced (BPCI Advanced), the first episode-based payment model to include an outpatient surgical condition-spine surgery. It is not known whether bundled payments reduce spending or improve quality for outpatient surgery, despite plans to expand outpatient episodes in future models. Objective: To determine whether hospital participation in the first year of BPCI Advanced for outpatient and inpatient spine surgery (back and neck except spinal fusion procedures [BNESF]) was associated with changes in spending and quality. Design, Setting, and Participants: A retrospective cohort study using Medicare claims and differences-in-differences analysis adjusting for patient and market characteristics was conducted comparing outcomes for patients receiving outpatient and inpatient BNESF from hospitals that participated in BPCI Advanced vs those receiving these procedures from a matched comparison group of nonparticipating hospitals. Medicare beneficiaries receiving outpatient and inpatient BNESF between 2013 and 2019 were included. Analyses were conducted between March 2023 and February 2024. Exposures: Hospital participation in BPCI Advanced. Main Outcomes and Measures: The primary outcome was total episode spending, including spending incurred for the index procedure and 90-day follow-up period. Secondary outcomes included 90-day return inpatient admissions, emergency department visits, and mortality. Results: Among 14 280 patients who received outpatient BNESF, hospital participation in BPCI Advanced was associated with a differential reduction in total episode spending (-$1201; 95% CI, -2184 to -219) and return inpatient admissions (-2.2 percentage points; 95% CI, -4.2 to -0.1). For outpatient procedures, the mean (SD) age was 71.8 (8.6) years; 43.9% were women, 3.9% were Black; and 3.2% were Hispanic. Among 23 440 patients who received inpatient BNESF, hospital participation in BPCI Advanced was not associated with differential changes in total episode spending or return inpatient admissions. There were no significant changes for emergency department visits or mortality for either group. Conclusions and Relavance: In this cohort study, participation in the first year of a bundled payment program for outpatient spine surgery was associated with nearly 10% lower spending. No changes in spending were observed for similar inpatient spine surgery procedures. Further evaluation of bundled payments for outpatient surgical conditions and associated changes in care delivery is needed to inform plans to include these episodes in future models.

  PublisherDOI

• Comparative Evaluation of Difference in Differences Methods for Staggered Adoption Interventions

  ArXiv.org · 2025-08-20

  preprintOpen access

  Staggered adoption is a common approach for implementing healthcare interventions, where different units adopt the program at different times. Difference-in-differences (DiD) methods are frequently used to evaluate the effects of such interventions. Nonetheless, recent research has shown that classical DiD approaches designed for a single treatment start date can produce biased estimates in staggered adoption settings, particularly due to treatment effect heterogeneity across adoption and calendar time. Several alternative methods have been developed to address these limitations. However, these methods have not been fully systematically compared, and their practical utility remains unclear. Motivated by a payment program implemented by a healthcare provider in Hawaii, we provide a comprehensive review of the staggered adoption setting and a selection of DiD methods suitable for this context. We begin with a theoretical overview of these methods, followed by a simulation study designed to resemble the characteristics of our application, where the intervention is implemented at the cluster level. Our results show that the current methods tend to under-perform when the number of clusters is small, but improve as the number of clusters increases. We then apply the methods to evaluate the real-world payment program intervention and offer practical recommendations for researchers implementing DiD methods for staggered adoption settings. Finally, we translate our findings into practical guidance for applied researchers choosing among DiD methods for staggered adoption settings.

  PublisherOA PDFDOI

Frequent coauthors

• Joshua M. Liao

  The University of Texas Southwestern Medical Center

  201 shared

• Ezekiel Emanuel

  94 shared

• Ravi B. Parikh

  Penn Center for AIDS Research

  72 shared

• Jingsan Zhu

  University of Pennsylvania

  72 shared

• Kevin G. Volpp

  University of Pennsylvania

  64 shared

• Kevin G. Volpp

  Penn Center for AIDS Research

  54 shared

• Deborah S. Cousins

  University of Pennsylvania

  36 shared

• Rachel M. Werner

  30 shared

Labs

• Medical Ethics and Health PolicyPI

Education

• B.S., Electrical Engineering & Economic Systems

  Stanford University

  2001

• Ph.D., Health Care Management and Economics

  The Wharton School, University of Pennsylvania

  2008

• M.D.

  University of Pennsylvania Perelman School of Medicine

  2010

Awards & honors

• Senior Fellow, Leonard Davis Institute of Health Economics
• Co-Director, Healthcare Transformation Institute, University…
• Founding Director, The Parity Center: Equity through Payment…