cost-benefit analysis

This study addresses confounding bias in cost-effectiveness analysis (CEA) using observational healthcare data. We propose a joint nonparametric causal inference framework based on Bayesian additive regression trees (BART), which simultaneously estimates heterogeneous causal effects of treatment on both health outcomes and medical costs—departing from conventional linear or unidimensional modeling approaches. By jointly modeling these dual endpoints, our method enables robust, unbiased estimation of the incremental cost-effectiveness ratio (ICER). The framework is implemented in an end-to-end, fully reproducible software package and validated empirically. Results demonstrate substantial improvements in accuracy and credibility of CEA conducted on observational data, thereby strengthening the validity of causal evidence for budget-constrained, evidence-based decision-making in healthcare.

This study addresses time-varying confounding bias in observational healthcare data—arising from non-random treatment assignment, administrative censoring, and irregular follow-up, exemplified by heterogeneous timing of adjuvant radiotherapy initiation among high-risk early-stage endometrial cancer patients—by proposing the first continuous-time Bayesian framework that jointly models medical costs and time-to-event outcomes to support dynamic treatment strategies. The approach employs Bayesian g-computation to estimate cost-effectiveness measures with causal interpretation and enables posterior comparisons across treatment regimens. By operating in continuous time, it circumvents the data inflation and zero-inflation issues inherent in discrete-time models, while accommodating censoring and dynamic decision-making under minimal parametric assumptions. Simulations demonstrate superior performance over existing methods under various censoring scenarios, and application to SEER-Medicare data yields a robust evaluation of the cost-effectiveness of initiating adjuvant radiotherapy within six months post-surgery.

This study addresses the challenge of jointly weighing loss aversion, adverse effects, and implementation costs in clinical or public health intervention decisions. Methodologically, it proposes a formal Bayesian utility-theoretic decision framework that integrates posterior distribution inference with a structured, non-arbitrary 1–9 value scale to map effect size, inter-individual response variability (standard deviation), and multidimensional costs onto a unified expected utility score. The framework further incorporates credible intervals, sensitivity analyses, and estimation of response-type proportions (benefit, no effect, harm). Its primary contributions are: (1) establishing the first comparable, interpretable, and operationally actionable quantitative utility paradigm for intervention evaluation; (2) generating both a single composite utility score and a full population-level response distribution; and (3) implementing the framework in an accessible spreadsheet tool, thereby providing a transparent, robust, and user-friendly aid for evidence-informed decision-making.

This study addresses the limitations of existing cost-effectiveness analyses in longitudinal cluster randomized trials, which often fail to adequately model the complex correlation structure between clinical and cost outcomes, thereby hindering accurate sample size determination. To overcome this, the authors develop a unified sample size calculation framework based on a bivariate linear mixed model for three common designs: parallel, crossover, and stepped wedge. They derive, for the first time, closed-form variance expressions under cost-effectiveness objectives and introduce a standardized ceiling ratio to calibrate willingness-to-pay thresholds. By employing generalized least squares estimation, the approach jointly accounts for within-period, between-period, and inter-outcome correlations. Integrating both locally optimal and MaxiMin robust design strategies, the method substantially enhances the precision and robustness of sample size estimates. Empirical validation using real stepped wedge trial data demonstrates its applicability and efficiency across multiple designs.

Local deployment of large language models (LLMs) in data-sensitive domains (e.g., healthcare, finance) faces prohibitive hardware costs and benchmarking redundancy due to frequent model iterations; existing evaluation tools neglect economic metrics, hindering informed deployment decisions. Method: We propose the first multi-objective benchmarking framework tailored for LLM local deployment, systematically incorporating cost per query ($) alongside accuracy, latency, throughput, and inference overhead modeling. It enables configuration-driven, reproducible economic evaluation via a modular, open-source Python toolkit integrating hardware telemetry and cross-model (Llama-3, Phi-3, Qwen) and cross-hardware (A10, A100, RTX 4090) benchmarking. Contribution/Results: Experiments demonstrate that our framework significantly improves deployment decision efficiency and enhances predictability of cost savings, thereby filling a critical gap in industrial-grade, economically grounded LLM evaluation.

Health economic evaluations often struggle to accurately estimate the joint distribution of potential outcomes under interventions due to the absence of a single comprehensive data source, leading to structural and parametric biases in decision-analytic models. This study proposes a unified framework that integrates decision-analytic modeling within causal inference by leveraging the potential outcomes paradigm to synthesize causal parameters from multiple data sources and approximate intervention effects. It systematically decomposes estimation bias into structural bias and target parameter bias, elucidating how nonstandard parameter specifications induce bias and demonstrating that model credibility critically depends on underlying causal assumptions. By clarifying these mechanisms, the work enhances the transparency and rigor of health policy decision-making.

This work addresses the limitations of conventional experimental evaluation methods, which treat multiple metrics in isolation, ignore their interdependencies, and rely on subjective judgments when objectives conflict—hindering scalability. To overcome these challenges, the authors propose the first framework that integrates Bayesian decision theory with hierarchical priors. By designing a custom loss function that incorporates business preferences alongside observed evidence, and leveraging historical experiment data to construct informative priors, the approach enables automated and systematic trade-offs among multiple objectives. Evaluated on both real-world and simulated supply chain experiments at Amazon, the method significantly improves estimation efficiency, streamlines complex decision-making processes, and transcends the constraints of traditional hypothesis testing.

This study systematically evaluates the economic value and its heterogeneity of multi-cancer early detection (MCED) technologies in population-based screening from the perspective of the UK National Health Service (NHS). The authors develop a decision-analytic model that innovatively decomposes overall health benefit—measured in quality-adjusted life years (QALYs)—into distinct components attributable to cancer detection, false positives, overdiagnosis, and misclassification, with analyses further stratified by cancer type. Simulations of annual Galleri screening yield a net population benefit of 0.135 QALYs per person, with colorectal, lung, and ovarian cancers collectively accounting for more than 50% of this gain. This framework elucidates the heterogeneous sources and distribution of health value, offering a robust foundation for priority setting and informed health policy decisions.

Traditional health evaluation metrics, such as QALYs and PALYs, struggle to simultaneously account for equity and productivity contributions. This work proposes a unified framework that integrates welfare economics and health measurement theory through a normative axiomatic approach. Within this framework, a new class of evaluation functions is developed, satisfying scale invariance and the Pigou-Dalton transfer principle. The authors derive a tractable power-form representation of these functions, offering a coherent basis for assessing interventions that jointly affect health and productive capacity. By doing so, the framework overcomes key limitations of existing metrics, providing a principled balance between efficiency and fairness in health policy evaluation.