About

Ramesh Johari is a Professor of Management Science and Engineering and, by courtesy, of Electrical Engineering at Stanford University. His research broadly focuses on the design, economic analysis, and operation of online platforms, utilizing statistical and machine learning techniques such as search, recommendation, matching, and pricing algorithms. He is actively involved in initiatives aimed at transforming mobility and ensuring AI systems align with human intent, contributing to cross-campus efforts like the Sustainable Mobility Center and the AI Alignment Project. Johari holds a PhD from MIT, obtained in 2004. He is an affiliated faculty member at the Precourt Institute for Energy and participates in various research symposia and conferences, where he presents on topics related to AI, data science, and platform economics. His work has been recognized through awards and his involvement in pioneering experimentation methods to address real-world challenges in digital and energy systems.

Research topics

• Computer Science
• Economics
• Medicine
• Political Science
• Psychology
• Microeconomics
• Computer Security
• Mathematics
• Medical education
• Engineering
• Operations research
• Telecommunications
• Econometrics
• Internal medicine
• Mathematical optimization
• Endocrinology
• Industrial organization
• Marketing
• Pedagogy
• Simulation
• Business
• Mathematics education

Selected publications

• Cost-Effectiveness of Continuous Glucose Monitoring With Remote Patient Monitoring in Pediatric Patients With Newly Diagnosed Type 1 Diabetes in the U.S.

  Diabetes Care · 2026-03-25

  articleOpen access

  OBJECTIVE: The use of continuous glucose monitoring (CGM) with remote patient monitoring (RPM) continues to grow. We evaluated the cost-effectiveness of CGM with RPM compared with self-monitoring of blood glucose (SMBG) and CGM alone. RESEARCH DESIGN AND METHODS: We simulated type 1 diabetes progression with a Markov model in 5-year-old patients over a 20-year, 50-year, and lifetime horizon. We tracked diabetic ketoacidosis (DKA), severe hypoglycemia (SH), and seven chronic complications: retinopathy, neuropathy, nephropathy, cardiovascular disease, end-stage renal disease, lower-extremity amputation, and blindness. We compared three interventions: SMBG, CGM, and CGM with RPM. Efficacy estimates were derived from meta-analyses of pediatric CGM studies and the results of the Teamwork, Targets, Technology, and Tight Glycemia Study (4T Study 1). We evaluated quality-adjusted life years (QALYs) and health care costs (2022 U.S. dollars) discounted at 3% annually. We performed extensive sensitivity analyses. RESULTS: Compared with SMBG, CGM increased QALYs by 0.09 and costs by $8,900 over 20 years; CGM with RPM increased QALYs by 0.37, and costs by $10,300. CGM with RPM yielded more QALYs at a lower incremental cost-effectiveness ratio compared with CGM ($27,400/QALY vs. $103,700/QALY, respectively). Results were robust across sensitivity analyses and time horizons. CGM with RPM remained cost-effective when achieving at least 30% of 4T's clinical efficacy. CONCLUSIONS: CGM with RPM delivers superior health outcomes compared with SMBG and CGM and is likely cost-effective for patients with newly diagnosed type 1 diabetes. Despite higher intervention costs, CGM with RPM can reduce complications costs and generate net health care savings.

  PublisherOA PDFDOI

• Cost-effectiveness of continuous glucose monitoring with remote patient monitoring in pediatric patients with newly diagnosed type 1 diabetes in the United States

  2026-03-25

  article

  <p dir="ltr"><b>Objective: </b>The use of continuous glucose monitoring (CGM) with remote patient monitoring (RPM) continues to grow. We evaluated the cost-effectiveness of CGM with RPM compared to self-monitoring of blood glucose (SMBG) and CGM alone.</p><p dir="ltr"><b>Research design and methods: </b>We simulated type 1 diabetes progression with a Markov model in 5-year-old patients over a 20-year, 50-year, and age 100 horizon. We tracked diabetic ketoacidosis (DKA), severe hypoglycemia (SH), and seven chronic complications: retinopathy, neuropathy, nephropathy, cardiovascular disease, end-stage renal disease, lower-extremity amputation, and blindness. We compared three interventions: SMBG, CGM, and CGM with RPM. Efficacy estimates were derived from meta-analyses of pediatric CGM studies and the results of the Teamwork, Targets, Technology, and Tight Glycemia Study (4T Study 1). We evaluated quality-adjusted life years (QALYs) and healthcare costs (2022 U.S. dollars) discounted at 3% annually. We performed extensive sensitivity analyses.</p><p dir="ltr"><b>Results: </b>Compared to SMBG, CGM increased QALYs by 0.09 and costs by $8,900 over 20 years; CGM with RPM increased QALYs by 0.37 and costs by $10,300 CGM with RPM yielded more QALYs at a lower incremental cost-effectiveness ratio compared to CGM ($27,400/QALY vs $103,700/QALY, respectively). Results were robust across sensitivity analyses and time horizons. CGM with RPM remained cost-effective when achieving at least 30% of 4T’s clinical efficacy.</p><p dir="ltr"><b>Conclusions: </b>CGM with RPM delivers superior health outcomes compared to SMBG and CGM and is likely cost-effective for patients with newly diagnosed type 1 diabetes. Despite higher intervention costs, CGM with RPM can reduce complications costs and generate net healthcare savings.</p>

  PublisherDOI

• Cost-effectiveness of continuous glucose monitoring with remote patient monitoring in pediatric patients with newly diagnosed type 1 diabetes in the United States

  2026-03-25

  article

  <p dir="ltr"><b>Objective: </b>The use of continuous glucose monitoring (CGM) with remote patient monitoring (RPM) continues to grow. We evaluated the cost-effectiveness of CGM with RPM compared to self-monitoring of blood glucose (SMBG) and CGM alone.</p><p dir="ltr"><b>Research design and methods: </b>We simulated type 1 diabetes progression with a Markov model in 5-year-old patients over a 20-year, 50-year, and age 100 horizon. We tracked diabetic ketoacidosis (DKA), severe hypoglycemia (SH), and seven chronic complications: retinopathy, neuropathy, nephropathy, cardiovascular disease, end-stage renal disease, lower-extremity amputation, and blindness. We compared three interventions: SMBG, CGM, and CGM with RPM. Efficacy estimates were derived from meta-analyses of pediatric CGM studies and the results of the Teamwork, Targets, Technology, and Tight Glycemia Study (4T Study 1). We evaluated quality-adjusted life years (QALYs) and healthcare costs (2022 U.S. dollars) discounted at 3% annually. We performed extensive sensitivity analyses.</p><p dir="ltr"><b>Results: </b>Compared to SMBG, CGM increased QALYs by 0.09 and costs by $8,900 over 20 years; CGM with RPM increased QALYs by 0.37 and costs by $10,300 CGM with RPM yielded more QALYs at a lower incremental cost-effectiveness ratio compared to CGM ($27,400/QALY vs $103,700/QALY, respectively). Results were robust across sensitivity analyses and time horizons. CGM with RPM remained cost-effective when achieving at least 30% of 4T’s clinical efficacy.</p><p dir="ltr"><b>Conclusions: </b>CGM with RPM delivers superior health outcomes compared to SMBG and CGM and is likely cost-effective for patients with newly diagnosed type 1 diabetes. Despite higher intervention costs, CGM with RPM can reduce complications costs and generate net healthcare savings.</p>

  PublisherDOI

• Analysis of Two-Phase Flow Dynamics in Arteries: Influence of Sodium Chloride on Oxygen and Blood Transport

  Abhi International Journal of Biological Science · 2025-07-07

  articleOpen access

  Blood circulation is a multiphase phenomenon involving plasma, red blood cells (RBCs), and dissolved gases such as oxygen and carbon dioxide. In a healthy individual, the oxygen saturation in the blood flowing through the arteries is typically between 95% and 100%. This is measured as a percentage of hemoglobin that is saturated with oxygen and indicates that the body is receiving sufficient oxygenated blood [1]. While the pulmonary artery carries oxygen-poor blood from the heart to the lungs, where gas exchange occurs and the blood picks up oxygen. In a healthy individual, the oxygen saturation in the pulmonary artery is around 76%, and this value increases to nearly 100% in the pulmonary veins after the blood has been oxygenated in the lungs [2]. The transport of oxygen from arteries to tissues depends on hemodynamic parameters and electrolyte concentration, particularly sodium chloride (NaCl). NaCl plays a crucial role in regulating osmotic pressure, plasma viscosity, and gas diffusion, thereby modifying two-phase flow dynamics. This study develops a two-phase flow framework in which blood is treated as the continuous phase and oxygen as the dispersed phase, incorporating NaCl effects into viscosity and diffusion coefficients. Clinical evidence from recent studies, including the DECIDE-Salt trial, highlights the systemic impact of sodium intake on hypertension and cardiovascular health [3].The combined physiological and mathematical approach presented here provides insights into the dual role of NaCl in micro-scale oxygen transport and macro-scale cardiovascular risk.

  PublisherOA PDFDOI

• Physical Activity Is Associated With Improved Glycemic Outcomes in Newly Diagnosed Youth With Type 1 Diabetes: 4T Exercise Program

  Diabetes Care · 2025-07-01 · 5 citations

  articleOpen access

  OBJECTIVE: The Teamwork, Targets, Technology, and Tight Range (4T) Exercise Program evaluated physical activity patterns across the first year of type 1 diabetes diagnosis and whether physical activity was associated with changes in glucose outcomes in the 24 h following physical activity. RESEARCH DESIGN AND METHODS: The 4T Exercise Program started newly diagnosed youth with type 1 diabetes on a continuous glucose monitoring (CGM) system and physical activity tracker around 1 month postdiagnosis. A subset of youth opted to participate in up to four quarterly structured exercise education sessions to increase their knowledge around safe physical activity. RESULTS: Ninety-eight youth with type 1 diabetes (median [interquartile range (IQR)] age of 13 [12-15] years, 45% female, 44% non-Hispanic White) completed the study. Compared with sedentary days, days with ≥10 min of vigorous-intensity physical activity were associated with an increase in time in range (TIR) of 2.3% (1.4-3.2%; P < 0.001), a decrease in time above range (TAR) of 3.1% (2.2-4.0%; P < 0.001), and an increase in time below range (TBR) of 0.8% (0.6-0.9%; P < 0.001) in the 24 h following physical activity. From 1-3 months to 10-12 months postdiagnosis, the median (IQR) step count increased by 1,134 (445-1,519) steps per day (P < 0.001), while daily moderate-to-vigorous physical activity increased by 11 (2-23) min per day (P < 0.001). CONCLUSIONS: In the 24 h following physical activity as compared with sedentary days, TIR improved, TAR was lower, and TBR remained within clinical target recommendations. For youth with new-onset type 1 diabetes, each structured exercise education session was associated with a further 0.79% increase in TIR.

  PublisherOA PDFDOI

• Estimation of Treatment Effects Under Nonstationarity via the Truncated Policy Gradient Estimator

  ArXiv.org · 2025-06-05

  preprintOpen access1st authorCorresponding

  Randomized experiments (or A/B tests) are widely used to evaluate interventions in dynamic systems such as recommendation platforms, marketplaces, and digital health. In these settings, interventions affect both current and future system states, so estimating the global average treatment effect (GATE) requires accounting for temporal dynamics, which is especially challenging in the presence of nonstationarity; existing approaches suffer from high bias, high variance, or both. In this paper, we address this challenge via the novel Truncated Policy Gradient (TPG) estimator, which replaces instantaneous outcomes with short-horizon outcome trajectories. The estimator admits a policy-gradient interpretation: it is a truncation of the first-order approximation to the GATE, yielding provable reductions in bias and variance in nonstationary Markovian settings. We further establish a central limit theorem for the TPG estimator and develop a consistent variance estimator that remains valid under nonstationarity with single-trajectory data. We validate our theory with two real-world case studies. The results show that a well-calibrated TPG estimator attains low bias and variance in practical nonstationary settings.

  PublisherOA PDFDOI

• Balancing Producer Fairness and Efficiency via Prior-Weighted Rating System Design

  Proceedings of the International AAAI Conference on Web and Social Media · 2025-06-07 · 1 citations

  articleOpen access

  Online marketplaces use rating systems to promote the discovery of high-quality products. However, these systems also lead to high variance in producers' economic outcomes: a new producer who sells high-quality items, may unluckily receive a low rating early, severely impacting their future popularity. We investigate the design of rating systems that balance the goals of identifying high-quality products (``efficiency'') and minimizing the variance in outcomes of producers of similar quality (individual ``producer fairness''). We show that there is a trade-off between these two goals: rating systems that promote efficiency are necessarily less individually fair to producers. We introduce prior-weighted rating systems as an approach to managing this trade-off. Informally, the system we propose sets a system-wide prior for the quality of an incoming product; subsequently, the system updates that prior to a posterior for each product's quality based on user-generated ratings over time. We show theoretically that in markets where products accrue reviews at an equal rate, the strength of the rating system's prior determines the operating point on the identified trade-off: the stronger the prior, the more the marketplace discounts early ratings data (increasing individual fairness), but the slower the platform is in learning about true item quality (so efficiency suffers). We further analyze this trade-off in a responsive market where customers make decisions based on historical ratings. Through calibrated simulations in 19 different real-world datasets sourced from large online platforms, we show that the choice of prior strength mediates the same efficiency-consistency trade-off in this setting. Overall, we demonstrate that by tuning the prior as a design choice in a prior-weighted rating system, platforms can be intentional about the balance between efficiency and producer fairness.

  PublisherOA PDFDOI

• Regression discontinuity in Time: Evaluating the impact of evolving digital health interventions

  International Journal of Medical Informatics · 2025-07-16

  articleOpen accessSenior author
  PublisherOA PDFDOI

• 1139-P: Assessing Clinic Readiness for Implementation of Equitable and Sustainable New-Onset Care in the T1DX-QI

  Diabetes · 2025-06-13

  article

  Introduction and Objective: The 4T Program emphasizes early and equitable initiation of diabetes technology with an interdisciplinary team approach to achieve optimal glycemic outcomes. The T1D Exchange Quality Improvement Collaborative (T1DX-QI) focuses on healthcare equity and improving care and outcomes. A collaboration between the 4T Program and the T1DX-QI is proposed to equitably improve new-onset care and address existing gaps in care for youth with T1D in the US. Methods: A survey of pediatric new-onset programs, resources, and potential challenges was sent to 16 T1DX-QI centers that attended a 4T program information webinar to assess implementation readiness. These findings will inform redesigning implementation and dissemination of new-onset care at T1DX-QI centers. Results: Fifteen centers completed the survey via REDCap describing their center’s characteristics (Table). Potential key challenges identified in implementing the 4T program included provider time for program delivery (93%) and informatics support concerns (47%). Other challenges included provider buy-in (20%) and the ability to provide device and technology support to families (20%). Conclusion: These data are being utilized in a 4T-T1DX-QI workshop to collaboratively discuss strategies with the same 15 centers to design equitable and sustainable implementation of improved new-onset care at T1DX-QI centers. Disclosure F.K. Bishop: None. A. Addala: None. G.T. Alonso: Advisory Panel; MannKind Corporation. L.C. Chao: None. A. Choudhary: None. M.A. Clements: Consultant; Glooko, Inc. Research Support; Dexcom, Inc., Abbott. D. DeSalvo: Consultant; Dexcom, Inc. Advisory Panel; Insulet Corporation. M. Desai: None. R. Johari: None. D.M. Maahs: Advisory Panel; Abbott, Medtronic. Research Support; Dexcom, Inc. Consultant; Sanofi. A. Mucci: None. P. Prahalad: Consultant; Sanofi. N. Rioles: None. S. Thapa: None. D.P. Zaharieva: Research Support; Hemsley Charitable Trust. Speaker's Bureau; Dexcom, Inc. Research Support; Insulet Corporation, International Society for Pediatric and Adolescent Diabetes. O. Ebekozien: Research Support; Abbott. Advisory Panel; Sanofi. Research Support; Sanofi, Lilly Diabetes, Medtronic. Funding The Leona M. and Harry B. Helmsley Charitable Trust

  PublisherDOI

• Optimal Empirical Risk Minimization under Temporal Distribution Shifts

  ArXiv.org · 2025-07-17

  preprintOpen access

  Temporal distribution shifts pose a key challenge for machine learning models trained and deployed in dynamically evolving environments. This paper introduces RIDER (RIsk minimization under Dynamically Evolving Regimes) which derives optimally-weighted empirical risk minimization procedures under temporal distribution shifts. Our approach is theoretically grounded in the random distribution shift model, where random shifts arise as a superposition of numerous unpredictable changes in the data-generating process. We show that common weighting schemes, such as pooling all data, exponentially weighting data, and using only the most recent data, emerge naturally as special cases in our framework. We demonstrate that RIDER consistently improves out-of-sample predictive performance when applied as a fine-tuning step on the Yearbook dataset, across a range of benchmark methods in Wild-Time. Moreover, we show that RIDER outperforms standard weighting strategies in two other real-world tasks: predicting stock market volatility and forecasting ride durations in NYC taxi data.

  PublisherOA PDFDOI

Recent grants

• CAREER: An Economically Sustainable Internet: From Technology and Architecture to Contracting Models

  NSF · $400k · 2007–2012

• EAGER: Efficient Pricing for Dynamic Resource Allocation in Transportation Systems

  NSF · $200k · 2010–2012

• SCH: Improving Patient Health and Equity through the Digital Transformation of Diabetes Care Delivery

  NSF · $1.2M · 2022–2026

• CPS: Breakthrough: Collaborative Research: The Interweaving of Humans and Physical Systems: A Perspective From Power Systems

  NSF · $250k · 2015–2019

• TRIPODS+X:RES: Collaborative Research: The Future of the Road - A Data-Driven Redesign of the Urban Transit Ecosystem

  NSF · $175k · 2018–2021

Frequent coauthors

• David Scheinker

  65 shared

• David M. Maahs

  61 shared

• Priya Prahalad

  Stanford University

  60 shared

• Dessi P. Zaharieva

  Stanford University

  56 shared

• Ananta Addala

  University of Fort Lauderdale

  53 shared

• Annie Chang

  Icahn School of Medicine at Mount Sinai

  53 shared

• Paul Dupenloup

  University of Bari Aldo Moro

  52 shared

• Ryan Pei

  University of Adelaide

  51 shared

Labs

• Management Science and EngineeringPI

Awards & honors

• Faculty MS&E faculty among AI Alignment Project awardees (20…
• Faculty ICME research symposium 2025
• MS&E names Commencement Award winners (2024)