Medical Decision Making

Background: Cost-effectiveness analyses (CEA) built on simulation models are usually delivered through project-specific scripts. Each new model, and often each new question, rebuilds the same scaffolding: parameter management, calibration, base-case analysis, sensitivity analysis and export. This is costly, and is where traceability between an estimate and the model version producing it is lost. We present THALASSA, a web application providing this infrastructure once, for any model satisfying a small programming contract. Methods: A simulation model is connected by one file implementing a fixed contract of R functions that determine its behaviour. THALASSA exposes the complete CEA workflow in a browser: a generated overview of the connected model, parameter editing, calibration, base-case comparison with the efficiency curve, and deterministic and probabilistic sensitivity analysis at a willingness-to-pay threshold, each with timestamped file export carrying the parameter values it was run with. Every session is bound to a repository version, so results carry their generating model version. Results: We demonstrate THALASSA on two public models of contrasting difficulty: a simple tutorial model and a generic cancer screening model with two competing natural-history pathways and a 21-dimensional calibration deliberately built to be more complex. Both ran on the same unmodified setup, implementing the same contract, with no code changes. Conclusions: Separating analysis infrastructure from model logic makes the CEA workflow reusable across projects and reproducible by construction. A modelling team can move from a versioned repository to audit-ready, exportable CEA outputs without writing analysis code. The integration effort does not grow with model size and one installation serves unrelated projects.

Importance: The 2026 public charge rule could discourage immigrants from accessing health coverage programs, creating a chilling effect that potentially leads to negative health outcomes; However, its long-term health impact is poorly understood. Objective: To model potential impacts of the 2026 public charge rule on primary care and premature mortality among immigrants in New York City (NYC). Design, Setting, and Participants: The simulation used a deterministic compartmental model with Ordinary Differential Equations (ODEs) using 2023 NYC Vital statistics data and American Community Survey, and estimates obtained from 2 previous studies about effects of healthcare access on primary care and Medicaid expansion on premature mortality. Main Outcomes and Measures: Rates of primary care outcomes (access, doctor's visits) in 5 years, and premature mortality in 5 and 20 years, projected by the model under conservative, moderate, and aggressive scenarios of avoidance/disenrollment due to the public charge rule, known as the 'chilling effect'. Effects of the avoidance/disenrollment on primary care outcomes and premature mortality were obtained from 2 previous studies. Projected rates of the outcomes under each scenario were compared with counterfactuals to estimate the health impacts of the chilling effect. Results: Implementation of the public charge rule was projected to decrease the primary care access rate by 4.1% (conservative) to 9.9% (aggressive) over 5 years, relative to the counterfactual scenario without the rule. The rate of doctors' visits was projected to decrease over 5 years by 5.1% (conservative) to 12.2% (aggressive). Premature mortality was projected to increase by 4.4% (conservative) to 10.6% (aggressive) in 5 years and 7.4% (conservative) to 17.4% (aggressive) in 20 years. Legal noncitizens and Black immigrant New Yorkers were predicted to experience higher burdens of premature mortality attributed to the chilling effect, compared with other immigrant groups and racial/ethnic groups, respectively. Conclusions and Relevance: This study demonstrates adverse health consequences of federal public charge rule changes among immigrants in NYC. The model projected a decrease in primary care visits and increase in premature mortality across various scenarios. These findings suggest urgent reconsideration of a regulatory change that disproportionately increases risk of premature mortality among immigrants in NYC.

Multi-Cancer Detection (MCD) tests can detect more than 50 different types of cancer using a blood test. Recently passed law in the U.S. guarantees that Medicare will pay for these tests when they are FDA approved and show evidence for clinical benefit. This manuscript provides estimates of the cost of MCD tests to Medicare under different assumptions of cost per test, eligibility, and screening uptake in the eligible population. This manuscript additionally estimates the cost of follow-up testing resulting from false positive results, which are considered avoidable costs caused by the screening test.

FeePredict is a three-stage random forest machine learning framework to simul-taneously predict whether Medicare reimbursement rates for specific procedures will change, in which direction they will change, and by how much. FeePredict was ap-plied to the four major Medicare fee schedules: the Clinical Laboratory Fee Schedule (CLFS), the Physician Fee Schedule (PFS), the Ambulance Fee Schedule (AFS), and the Durable Medical Equipment, Prosthetics, Orthotics, and Supplies (DMEPOS) fee schedule. Each of these fee schedules contains publicly available data from the Centers for Medicare & Medicaid Services (CMS) for the years 2024, 2025, and 2026, with the number of procedures represented in the data ranging from 3,264 to 2,952,842 observations.FeePredict utilizes lag-1 feature engineering and train-only preprocessing steps to ensure that there is no data leakage into the model. Chronological out-of-time valida-tion was performed on three of the four fee schedules to determine the generalizability of the model over time. FeePredict significantly outperformed the assumption that there would be no changes to Medicare reimbursement rates for procedures (p < 0.001), achieving concordance indices between 0.815 and 0.998, and reducing the mean abso-lute error for predicting changes to reimbursement rates by 29% to 85%. Permutation testing of the model with shuffled reimbursement rate labels indi-cates that there is no evidence of data leakage (AUC values: 0.467-0.515). The model achieved concordance indices of 0.854 and 0.972 for the CLFS and DMEPOS fee sched-ules, respectively, outside of its training period, but performed less well outside of its training period for the PFS, indicating that it generalizes less well to changes to the Medicare policy regime that existed after its training period. Overall, though, these re-sults indicate that it is possible to accurately predict whether Medicare reimbursement rates for medical procedures will change using only data from the historical versions of those fee schedules.

Clinical recommendations are often expressed in narrative form, which limits their direct execution, auditability, and patient-specific interpretation. This paper presents a hybrid decision-support framework that combines Decision Model and Notation (DMN), survey-weighted rule-ensemble learning, and counterfactual sensitivity analysis. The framework is evaluated using an NHANES-derived fasting cohort for classification of documented diabetes status. The full fasting analysis cohort contained 2,582 participants, and a non-diagnostic laboratory subgroup, Gate0, contained 2,111 participants. On untouched test data, the rule-ensemble model achieved ROC-AUC and PR-AUC values of 0.959 and 0.873 in the full fasting cohort and 0.861 and 0.499 in Gate0. Four clinically interpretable candidate rules were selected using validation data only. A nonnegative survey-weighted logistic model removed one redundant rule and converted the remaining three binary activations into an auditable DMN score and model-estimated probability. The final DMN achieved ROC-AUC 0.769, PR-AUC 0.153, and Brier score 0.029 in the untouched Gate0 test set. In small rule-defined test subgroups, hypothetical five-unit BMI reductions lowered mean model-estimated probability by 2.40 to 5.89 percentage points when one or more BMI thresholds were crossed. These findings characterize policy sensitivity rather than causal effects and require external validation.

Large language model (LLM)-based generative agents can reproduce aspects of individual human behavior, but whether they can be scaled to geographically grounded populations that reproduce real-world health behaviors remains unclear. Here, we introduce LLMPopSim, a generative population simulation framework that integrates U.S. Census and Centers for Disease Control and Prevention data to construct synthetic individuals and uses an LLM to simulate individual health behaviors whose aggregate outcomes can be evaluated at the community level. We developed the framework using historical Hawaii data and evaluated temporal and geographic generalizability using held-out 2022 cohorts from Hawaii and New York State, with colorectal cancer screening and mammography as proof-of-concept behaviors. Across the four state-outcome evaluations, mean absolute error ranged from 3.5 to 15.0 percentage points and correlations between simulated and observed ZCTA-level prevalence ranged from 0.26 to 0.69. Performance differed across dimensions of population fidelity: colorectal cancer screening predictions preserved geographic ranking more strongly but systematically overestimated prevalence and compressed geographic variation, whereas mammography achieved lower absolute error but weaker geographic correlation and inconsistent preservation of between-community variability. Prediction error was greatest in communities with lower observed screening prevalence and varied across community characteristics without a uniform socioeconomic gradient. These findings demonstrate that individually represented LLM-based synthetic agents can aggregate into population-level patterns that retain measurable features of real-world health behavior across temporal and geographic transfer, while identifying calibration, distributional fidelity and subgroup performance as key challenges for generative population simulation. LLMPopSim provides an empirical foundation for developing synthetic populations that may ultimately enable simulation of heterogeneous population responses to public health interventions.

Background: Descriptive mapping of intensive care unit (ICU) and high care unit (HCU) capacity across Japan's secondary medical areas (SMAs) characterizes where beds exist, but medical planning also requires answers to prospective questions: how likely is a capacity shortfall under demand surge, which assumptions drive that risk, and how much protection do inter-zone transfer arrangements provide. No openly available tool addresses these questions at the SMA level, the geographic unit at which Japanese medical plans are written. Methods: We developed MeshScope-Scenario, a probabilistic capacity-demand framework operating on the MeshScope-Region platform. For a selected SMA, observed inputs (notified ICU/HCU beds from the Hospital Bed Function Reports; resident population) are combined with four explicitly flagged assumption parameters - effective staffed-bed rate, concurrent severe-care demand per 100,000 population, surge multiplier, and net cross-boundary inflow - each with a user-specified distribution. A seeded Monte Carlo engine (deterministic reproduction under a fixed seed) estimates the distribution of bed shortfall; interventions are compared under common random numbers. Parameter dependence is introduced by a Gaussian copula with automatic positive-semidefinite correction; global sensitivity is quantified by Sobol first-order and total-order indices (Saltelli sampling, Jansen estimators) alongside a deterministic one-at-a-time tornado analysis. A two-zone extension transfers unmet demand to the nearest ICU-holding SMA using road-network travel times measured in MeshScope-Region, with a transfer time limit and an acceptance cap; because both zones share the same systemic draws, correlated exhaustion of donor capacity under surge ("shared-fate" risk) is represented structurally. A seasonal layer applies twelve monthly surge multipliers and reports the distribution of annual maximum shortfall and month-specific shortfall probabilities. The engine is a dependency-free pure-function module verified by 40 statistical tests. Results: The framework reproduces identical output under identical seed and input; a flat seasonal profile reproduces the non-seasonal model exactly; copula factorization error is below 1e-9; and 10,000 iterations across three intervention variants complete in approximately 50 ms in a standard browser, permitting fully interactive use. Applied to three archetypal SMAs from the observed FY2024 supply map (seed 42, 20,000 iterations, demand prior 5 per 100,000), an ICU-zero zone with a transfer partner 48 minutes away has shortfall probability 66.0% (P50 4.8, P90 20.7 beds); a median metropolitan zone, 32.7% (P90 5.9), with Sobol indices ranking demand density and surge dominant; a high-supply zone shows zero shortfall up to approximately 2.5x surge. For the ICU-zero zone, independent-donor reasoning credits the transfer arrangement with a 2.27-bed reduction in expected shortfall, of which shared-fate correlation removes 93%; the probability of severe shortfall under the arrangement equals that with no arrangement at all, while a committed pool of five donor beds (6% of donor effective supply) restores a 13-point reduction and more than halves the arrangement's correlation exposure. Conclusions: MeshScope-Scenario extends SMA-level capacity mapping from description to prospective risk assessment. All demand-side inputs are declared assumptions with adjustable distributions rather than estimates presented as fact; the framework's value is to make the consequences of those assumptions, and their interaction with observed supply, explicit, reproducible, and inspectable for planning deliberation.

Improving primary care is a long-standing strategy to constrain health care spending. Yet, evaluations of primary care models focused on payment reform have shown minimal effects on total cost of care. We report the results from a large-scale, real-world evaluation of an advanced primary care model that restructures access through same-day and next-day appointments, on-demand video visits, asynchronous clinician messaging, and extended hours. Using a stacked-cohort difference-in-differences design with entropy balancing and inverse probability of censoring weighting, we analyzed multi-payer claims covering April 2022 through March 2025. Advanced primary care use was associated with an 8.6% reduction in total cost of care (-$729 per patient per year; P = 0.004), driven by lower specialist cost (-$939/year; P < 0.001) and, to a lesser degree, by reductions in inpatient (-$134/year; P < 0.001), urgent care (-$70/year; P < 0.001), and emergency department cost (-$16/year; P = 0.02), partially offset by higher primary care cost (+$350/year; P < 0.001). The specialist reduction was concentrated in knowledge-based consultative encounters (-$663/year; P < 0.001), while procedural specialist cost was largely unchanged (-$276/year; P = 0.09). Cost differences emerged in the first post-index month. These findings suggest that advanced primary care may reduce total health care spending, with observed savings driven primarily by lower spending on consultative specialty care.

Across 30 US states and the District of Columbia, eliminating the AIDS Drug Assistance Program is projected to save $6.45 billion in direct costs while generating $14.89 billion in downstream HIV care costs attributable to excess incident infections from 2026-2035. Costs are projected to surpass savings within six years.

Inappropriate antibiotic use presents a major global health challenge, particularly in low-resource settings where access to quality care is limited but antibiotics remain relatively unrestricted. This study estimates the causal effect of frontline primary care quality on inappropriate community antibiotic use, combining detailed community-based data from approximately 100 rural villages in rural China with an instrumental variable (IV) approach embedded within a double/debiased machine learning (DML) framework. We linked objective measures of village doctor clinical practice quality, measured through unannounced standardized patient visits, to household-level antibiotic use data collected from the same villages. To identify the causal effect, we constructed multiple candidate instruments from extensive provider characteristics and used an ensemble of machine learning algorithms within a flexible DML-IV framework to approximate an optimal instrument, addressing a many-weak-instruments problem. We found that improving village provider clinical practice quality reduced both antibiotic receipt during healthcare encounters for common diseases and household antibiotic storage for future self-medication. Our findings suggest that strengthening frontline primary care quality can meaningfully reduce inappropriate community antibiotic use without restricting access to essential treatment. More broadly, this study illustrates how causal machine learning can strengthen conventional causal estimation in complex observational settings in global health economics research.

The US spends more on health care than any other nation, yet tens of millions of Americans are uninsured or underinsured, and coverage retractions enacted in 2025 are widening these gaps. The misalignment between the for-profit insurance architecture and optimal patient care, together with the inefficiencies of a fragmented system, contributes to both unnecessary costs and preventable mortality. We update our previous analyses with the most recent data to project the economic benefits and the number of lives saved that would be achieved by single-payer universal coverage, as proposed in the Medicare for All Act. We estimate that such a system would reduce national health expenditure by $1,041 billion annually. Sources of savings include reductions in administrative overhead, pharmaceutical prices, fraudulent billing, and avoidable emergency care. Combined with the reversal of recent retractions, universal coverage would save over 114,000 lives annually.

Background Conventional cost-effectiveness analysis assumes societal indifference to the distribution of health gains across potential patients, yet evidence suggests public preferences vary between concentrating and dispersing benefits, depending on the size of health gain. The handful of Person Trade-Off (PTO) studies that explored this did not examine the cognitive mechanisms and rationales underlying the PTO tasks. Objectives To qualitatively investigate how individuals conceptualise and justify trade-off decisions within a PTO task designed to elicit distributional preferences regarding healthcare benefits. Methods Following a pilot phase (n=10), online cognitive debrief interviews (n=10) were conducted with UK adults using a think-aloud protocol and concurrent verbal probing. Participants completed an interactive, web-based PTO task that presented choices between concentrating larger health gains among fewer recipients or dispersing smaller gains across a larger population. Data were analysed using codebook thematic analysis. Results Six primary themes emerged across two categories: decision-making processes and decision drivers. Participants displayed varying degrees of moral discomfort, utilizing cognitive mechanisms ranging from strict mathematical logic (QALY maximisation) to emotion-driven heuristics. Key decision drivers included patient age, quality of survival, societal productivity, and strong death aversion. Participants frequently rejected pure QALY maximisation and held strong egalitarian preferences to give every patient a chance to receive treatment. Conclusion Public perspectives on benefit distribution are highly heterogeneous and shaped by complex normative values rather than pure health maximisation. Understanding these qualitative rationales is essential for isolating genuine distributional preferences from cognitive shortcuts, aiding health technology assessment bodies in evaluating equity weights.

Personalized medicine aims to tailor treatments to individual patients, but its implementation is limited by scarce clinical resources and heterogeneous treatment effects. Electronic health records now offer longitudinal data that can support data-driven therapy selection. We focus on SGLT2 inhibitors, medications with broad metabolic and cardiovascular benefits, whose clinical impact varies significantly based on individual patient profiles. We propose a counterfactual Structural Causal Model (SCM) to estimate individual treatment effects (ITE) of SGLT2 inhibitors from observational data. The model assumes counterfactual stability and uses a Gumbel-Max sampling mechanism to capture stochastic treatment assignment while enabling differentiable inference. Optimal treatments are defined by maximizing the counterfactual value of future diagnostic outcomes. We evaluate the Counterfactual Gumbel-Max SCM using a doubly robust value estimator and show that, across diagnostic categories, it outperforms real-world prescriptions in predicted treatment value, highlighting subpopulations where SGLT2 inhibitors yield superior outcomes. By integrating causal inference with electronic health data, this framework provides individualized counterfactual reasoning about SGLT2 treatment effects and can support clinicians in identifying optimal therapies for type 2 diabetes and comorbid conditions.

Introduction: Research has documented that obesity and morbidity are associated. Black persons in the United States (U.S.) incur higher financial costs of obesity-related multimorbidity (ORM). However, lifetime healthcare costs (LHCs) remain underexamined for these populations. Objective: We quantified racial differences in 1) LHCs and 2) lifetime healthcare cost differential (LCD) associated with ORM for ages > 40 years. Methods: We used the 2008- 2012 Medical Expenditure Panel Survey Household Component to examine unique obesity-related diseases (ORDs): high blood sugar, hypertension, coronary heart disease, and stroke. We used a prior published Markov model to simulate a person's life history of ORDs and compute LHCs among ages > 40 years. We computed LCD-associated ORM as the difference in LHC for those with ORM and LHC for members without ORDs. We quantified differences in race as the difference between LHC or LCD among White and Black men and women. Results: Our analytic sample included 53,035 Black and White persons representing 97,229,611 (S.E., 2,104,365), 12.4% as Black and 87.6% as White persons. ORM was more prevalent in the Black (21.2%) than the White group (13.4%). LHCs by race (Black/White) for women/men with ORM and LCDs associated with ORM (2012$) were $3 1,035/43,595 and $11,350/26,948 for age 40-49, $2 1,567/25,6 115 and $3,846/9,808 for 50-59, $9,863/18,515 and -$2,566/7,426 for 60-69, -$8,220/16,285 and -$11,524/3,865 for 70-79. Conclusions: Racial Differences in LHCs and LCDs related to ORM persist and vary across subpopulations. Future interventions designed to prevent/manage ORM are crucial for prioritizing populations with high LHCs and advancing health equity.

Ambient artificial intelligence scribes are being increasingly used in healthcare to improve efficiency and reduce provider clinical documentation burden, yet their performance across linguistically diverse patient populations is not well characterized. We conducted a retrospective analysis of 54,160 outpatient encounters within a U.S. safety net health system to evaluate the performance of an artificial intelligence documentation tool in English and non-English clinical encounters, and in encounters where an interpreter or bilingual provider was present. Documentation performance was measured by the percentage of words in the final note that were generated by the ambient AI documentation tool and not edited by the provider. Associations between language factors and documentation performance were measured using Generalized Estimating Equations with exchangeable correlation structures to account for clustering of multiple encounters within unique patients. Univariable models were fitted to estimate the odds of adequate performance by language and interpreter modality, and a multivariable interaction model was used to evaluate within-language differences between bilingual providers and interpreter-mediated encounters. Non-English encounters were 21% to 25% less likely than English encounters to achieve the same performance threshold. There was no significant difference in generative documentation performance between interpreter-mediated and bilingual provider encounters. These findings underscore the importance of equity-focused evaluation and multilingual model refinement to ensure that artificial intelligence documentation benefits are distributed fairly across diverse patient populations.

Background: While health system costs of COVID-19 are well documented, evidence on out-of-pocket costs to patients and families, including non-medical and time burden, is limited. This study aimed to estimate out-of-pocket medical, non-medical, and time burden of COVID-19 illness directly from patients and caregivers. Methods: Two cross-sectional surveys were fielded to national nonprobability samples of U.S. adults in Fall 2023: a personal burden survey (adult sample) and a caregiver burden survey (child sample). Surveys collected data on care setting, visit duration including travel and wait time, and medical and non-medical out-of-pocket costs. After applying quality-based exclusion criteria, 409 adult sample and 52 child sample responses were analyzed. Results: Over 87% of respondents were insured. Median out-of-pocket medical costs per visit were low, ranging from $5 (emergency department) to $10 (urgent care) in the adult sample, and from $0 (emergency department) to $75 (hospitalization) in the child sample. Median non-medical costs were $28 or less across care settings. Median time spent receiving care ranged from 90 minutes (primary care, child sample) to 300 minutes (emergency department, adult sample). Despite the relatively low reported out-of-pocket costs, COVID-related financial challenges were reported by 34% of the adult sample and 29% of the child sample, with the most common difficulty being paying non-medical bills (11% and 15%, respectively). Twelve percent in both groups reported problems paying COVID-19-related medical bills. Conclusions: In this post-pandemic non-probability sample, individual out-of-pocket costs for COVID-19 illness were relatively modest. However, time burden was substantial and one-third of respondents reported financial hardship. These estimates, including non-medical and time burden not captured in administrative data, can support more robust economic evaluations of COVID-19 prevention and treatment interventions.

Objectives: The North Carolina Integrated Care for Kids (NC InCK) pediatric demonstration model created a whole-child risk stratification approach to identify children for enhanced care management. Unlike traditional approaches that rely primarily on healthcare data, NC InCK integrates health, education, social, and juvenile justice data. We described the NC InCK risk stratification approach and compared its output to Medicaid managed care organization (MCO)-assigned risk levels, with the hypothesis that these approaches would identify distinct populations. Methods: This was a cross-sectional study including NC Medicaid participants aged 0-20 years residing in a five-county region in central NC in October 2022. The NC InCK risk stratification algorithm was used to assign participants to Service Integration Levels (SILs) 1, 2, or 3, reflecting increasing anticipated need for integrated cross-sector support. Descriptive analyses compared participant baseline characteristics by NC InCK and Medicaid MCO risk level. Results: Of 99,564 children, 90,151 (90.5%) were assigned to SIL 1, 5,482 (5.5%) to SIL 2, and 3,931 (4%) to SIL 3. About 24% and 20% of children assigned to SILs 2 or 3 by NC InCK were identified as low risk by their Medicaid MCO. Conclusions: The NC InCK risk stratification approach assigns children to SILs 2 and 3 using health, education, social, and juvenile justice data. These populations differ from those identified by Medicaid MCO risk stratification, suggesting that inclusion of non-clinical variables adds unique value. This approach is a step forward in demonstrating the feasibility of methods that proactively identify populations for preventive support by including upstream risk factors.

Objective: To estimate the impact of removing race-adjusted antihypertensive treatment recommendations for Black adults on health outcomes, healthcare utilization, and racial/ethnic disparities. Materials and methods: A discrete-time state-transition microsimulation model of hypertension natural history and health system utilization was developed to simulate lifetime trajectories among 100,000 non-Hispanic Black and non-Hispanic white 40-year-old adults. We evaluated the effect of removing race-adjusted antihypertensive treatment recommendations under two scenarios: one assuming racial heterogeneity in treatment effectiveness, reflecting evidence behind prior race-specific guidelines, and the other assuming no racial heterogeneity. Our main outcomes included life expectancy, cumulative incidence and duration of hypertension, and blood pressure control with antihypertensive treatment. Results: Under assumptions of heterogeneous treatment effectiveness by race, guidelines without race adjustment decreased life expectancy among non-Hispanic Black adults by 0.4 years, increasing the life expectancy gap with non-Hispanic white adults from 4.1 to 4.5 years. Hypertension control among non-Hispanic Black adults also worsened. Years with uncontrolled hypertension increased by 0.8 years, widening the disparity with non-Hispanic white adults from 1.7 to 2.5 years. Assuming no racial heterogeneity, these effects were diminished: life expectancy decreased by 0.1 years and years with uncontrolled hypertension increased by 0.2 years, less than one-half and one-quarter of the effects observed under assumptions of heterogeneous treatment effectiveness by race. Discussion: The potential effects of removing race-adjusted antihypertensive treatment guidelines depended on assumptions of racial heterogeneity in treatment effectiveness. Conclusion: Our findings highlight the importance of understanding drivers of previously observed racial differences when considering modifying antihypertensive treatment guidelines.

Importance: The Medicare Physician Fee Schedule values ambulatory services using survey-based estimates of physician time, a costly process vulnerable to inaccuracy. EHR audit logs are passively and universally collected and likely meet the American Medical Association Relative Value Scale Update Committee (AMA RUC) Extant Database Standards for alternative data to inform service valuation estimates. However, audit logs' validity for estimating face-to-face time, a core component of many of the most used procedural codes, is unknown. Objective: To assess whether EHR audit logs can facilitate accurate estimation of physician-patient face-to-face time during ambulatory visits. Design: Single-site validation study pairing EHR audit logs with direct audio/video recordings of ambulatory primary care visits (reference standard). We assessed agreement between audit-log estimates and recorded visit duration with Bland-Altman analysis and mean absolute error. Setting: University of Pennsylvania Observer data repository, a multi-modal database of ambulatory visits to Penn Medicine primary care clinics including video and audio recordings, transcripts, clinical EHR data, and EHR audit-log data. Participants: 75 visits from 12 physicians with data collected between August 2023 and February 2025. Exposures: A rule-based algorithm estimated face-to-face time via clusters of physician EHR activity relative to patient check-in time. Inactivity thresholds were applied to differentiate intraservice time with no EHR activity (e.g., a physical exam) from the end of a visit, with longer thresholds used for longer visits. Main Outcomes and Measures: Duration of recorded face-to-face visit time. Results: Recorded face-to-face time averaged 31.6 minutes (SD, 13.9). The audit-log estimator was unbiased (mean bias: -0.63 minutes; 95% CI: -6.0 to 2.6; mean absolute error: 9.0 minutes) and estimated face-to-face time within approximately 2% (38 seconds) of the recorded visit time with no proportional bias (slope: 0.13; P = .25). The 95% limits of agreement were -25.7 to 24.4 minutes, and 37% of estimates fell within 10% of recorded duration. Conclusions and Relevance: In this validation study, EHR audit logs recovered unbiased estimates of ambulatory face-to-face time. Larger-scale studies are required to verify generalizability, but EHR audit logs offer a scalable means to improve the rigor of retrospective physician work relative value unit estimation. This study represents a first step in a potential paradigm shift for measuring components of the fee schedule across procedures, settings, and provider types.

Background: Digital reminders reduce outpatient no-shows, but the optimal timing and frequency of messages remain unclear, particularly in Latin American public health systems. We emulated a target trial to evaluate the comparative effectiveness of four WhatsApp reminder strategies on appointment absenteeism and patient-initiated cancellations. Methods: We analyzed administrative and electronic health-record data from the public health system of the Autonomous City of Buenos Aires, Argentina (June 2023-May 2024). Eligible individuals had scheduled an in-person outpatient appointment in one of 15 prioritized specialties at least 75 hours in advance and had a mobile phone on record. We compared four strategies: (1) dual reminders at ~72 and ~24 hours before the appointment; (2) a single reminder at ~72 hours; (3) a single reminder at ~24 hours; and (4) no reminders. The primary outcome was the proportion of no-shows by the end of follow-up. Secondary outcomes were the cumulative incidence of patient-initiated cancellations overall, within 12 hours of the appointment, and followed by rebooking. We emulated the target trial using a cloning-censoring-weighting approach to estimate per-protocol controlled direct effects, with inverse-probability weights to address time-varying confounding and selection bias. Cumulative incidence of secondary outcomes was estimated using weighted Kaplan-Meier curves. Three pre-specified sensitivity analyses and standardized mean differences assessed robustness and covariate balance. Results: A total of 475,214 first eligible person-appointments were included; baseline no-show risk in the control arm was 34.6%. All three active strategies reduced no-shows compared with no reminders. The single 24-hour reminder produced the largest reduction (Risk Ratio [RR] 0.76, 95% CI 0.72, 0.81; Risk Difference [RD] -8.21 percentage points [pp], 95% CI -9.68, -6.54), followed by the dual-reminder strategy (RR 0.80, 95% CI 0.79,0.81; RD -7.05 pp, 95% CI -7.41, -6.71) and the single 72-hour reminder (RR 0.91, 95% CI 0.84,0.99; RD -3.16 pp, 95% CI -5.69, -0.49). All active strategies increased patient-initiated cancellations relative to control, with the dual-reminder strategy producing the largest increase. Sensitivity analyses preserved the qualitative ranking of strategies across all specifications. Conclusions: In this large target trial emulation, a single just-in-time WhatsApp reminder sent ~24 hours before the appointment was as effective as a dual-reminder schedule in preventing no-shows and superior to a distal 72-hour reminder alone. Adding a second, distal reminder provided no measurable benefit for attendance but substantially increased patient-initiated cancellations, which may be operationally valuable when active slot reallocation is a goal. These findings support timing, rather than frequency, as the primary lever of digital-reminder effectiveness, and favor the deployment of a single proximal reminder as the default strategy in resource-constrained outpatient settings.