As China rolls out Diagnosis-Related Group (DRG)-based prospective payment, hospitals face mounting pressure to understand exactly where their clinical departments operate efficiently and where resources are wasted. A new study, published in BMC Health Services Research, offers a pilot-tested answer from a single tertiary hospital in Foshan, using 2024 data to build a department-level diagnostic framework that combines efficiency measurement with statistical analysis of what drives performance differences.
The research team, led by Xiaoyi He and Wenxia You, applied a super-efficiency Slack-Based Measure (SBM) model that deliberately incorporates undesirable outputs alongside desirable ones. The undesirable outputs were average length of stay and the 30-day readmission rate, while the inputs were the number of doctors, nurses, and beds. Desirable outputs included bed turnover frequency, total DRG weight, and the Case Mix Index (CMI), a measure of case complexity. This structure allows the model to penalize departments that achieve output gains by prolonging stays or driving up readmissions.
The results revealed substantial room for improvement across the hospital. The mean overall technical efficiency was 0.73, with mean pure technical efficiency at 0.83 and mean scale efficiency at 0.91. The strongest performers were Pediatric Surgery, with a pure technical efficiency score of 2.56, followed by Neonatology at 1.75 and Cardiothoracic Surgery II at 1.58, scores above 1 indicating super-efficiency relative to peers.
As a second stage, the researchers used quantile regression to examine whether the correlates of efficiency differ across the performance distribution, supported by OLS regression and BCC-based robustness checks. This distributional approach matters because factors that drag down weak departments may behave very differently in strong ones.
That heterogeneity proved real. The composite size index showed a significant negative association with efficiency, most pronounced among low-performing departments, with a coefficient of −0.122 at the 0.25 quantile. Average length of stay was consistently negatively associated with efficiency, and the association strengthened across quantiles, moving from −0.351 to −0.452. In other words, longer stays appeared to matter even more among the better-performing departments.
The Case Mix Index told a different story. Its positive association with efficiency grew stronger across quantiles, from 0.336 to 0.705, but was statistically significant only among high-performing departments, with a coefficient of 0.705 at the 0.75 quantile (p = 0.029). This suggests that concentrating on complex cases may distinguish top performers, though the cross-sectional design cannot establish causation. Notably, the 30-day readmission rate showed no independent association with efficiency, and OLS regression found no statistically significant independent effect for any individual input—doctors, nurses, or beds.
From these patterns the authors propose a stratified strategy: downsizing may be appropriate for low performers, process streamlining for medium performers, and a focus on complex cases for high performers. They caution, however, that these are hypotheses requiring validation across multiple hospitals and longer time periods. The findings also suggest that indiscriminate resource expansion is not uniformly associated with higher efficiency and may worsen diseconomies of scale.
The study’s limitations are explicit. It is a single-hospital, cross-sectional analysis from 2024, which the authors acknowledge limits broad policy conclusions. Its value lies in demonstrating a feasible, closed-loop diagnostic framework—integrating efficiency measurement, slack analysis, and quantile regression—that other tertiary hospitals navigating their own DRG transitions can replicate and refine as they seek baseline efficiency diagnoses and tiered resource optimization.
Subject of Research: Operational efficiency of clinical departments under DRG-based payment in a Chinese tertiary hospital
Article Title: Operational efficiency of clinical departments under DRG-based payment: a super-SBM model with undesirable outputs and quantile regression analysis
Article References: He, X., Liang, L., Zhou, X., Chen, F., He, Z., Qiao, Y., Yao, Z., & You, W. (2026). Operational efficiency of clinical departments under DRG-based payment: a super-SBM model with undesirable outputs and quantile regression analysis. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15957-2
Image Credits: AI Generated
DOI: 10.1186/s12913-026-15957-2
Keywords: DRG-based payment, operational efficiency, super-SBM model, quantile regression, tertiary hospital, Case Mix Index, length of stay, readmission rate, China, health services research, Operational, efficiency
News Source: Ophelia Keating. (October 29, 2026). Single-hospital study maps department efficiency under China’s DRG payment shift. Scienmag.
