The nonlinear relationship between provincial hospital bed scale and the pure technical efficiency of the healthcare system and its regional heterogeneity: evidence from pure technical efficiency, DEA-Malmquist, and a two-way fixed-effects model

Background Hospital beds constitute an important material resource of healthcare systems and profoundly affect regional medical supply and service efficiency. Since the 2009 healthcare reform, hospital bed scale has continued to expand; however, supporting human resources and technical capabilities have lagged behind, highlighting prominent regional allocation-related contradictions. Existing empirical studies lack systematic provincial-level evidence to support quantitative references for differentiated bed allocation policies. Accordingly, this study analyzes the nonlinear relationship between the number of hospital beds and the pure technical efficiency (PTE) of provincial healthcare systems, as well as their regional heterogeneity. Methods Using balanced panel data covering 31 Chinese provinces from 2014 to 2023, hospital beds were excluded from input indicators in Data Envelopment Analysis (DEA) to calculate the pure technical efficiency of provincial healthcare systems; this indicator represents the relative efficiency level obtained under the input-output assumptions specified in this study. The Malmquist index was adopted to analyze dynamic changes in total factor productivity change (TFP). A two-way fixed-effects model containing the quadratic term of the logged bed scale was constructed. The Lind-Mehlum test was used to identify nonlinear relationships. Subgroup regressions for eastern, central, and western regions, as well as robustness tests, were performed. Results National healthcare TFP declined by an average of 2.2 % annually from 2014 to 2023, with an annual mean technical change (TC) of 0.980 and technical efficiency change (EC) of 0.998. Under the baseline model using the provincial absolute bed scale, the provincial bed scale exhibits a U-shaped association with pure technical efficiency, with a sample-based statistical turning point of approximately 190,200 beds. This turning point is an in-sample statistical estimate and cannot be directly used as a universal policy threshold. Observations show that an increase in the number of hospital beds on the left side of the inflection point is accompanied by a decline in pure technical efficiency; after passing the inflection point, efficiency rises, but the marginal gain is slight. Based on a full-sample interaction analysis, regional heterogeneity was statistically significant overall, with the heterogeneity primarily observed between the western region and the eastern and central regions. Conclusion The identified U-shaped association and sample-estimated turning point challenge conventional linear assumptions and provide quantitative references for provincial-level hospital bed planning. Combining the results of static efficiency regression with those of dynamic total factor productivity decomposition reveals that the decline in national total factor productivity in the healthcare sector is primarily driven by insufficient technological progress; extensive bed expansion struggles to simultaneously ensure both current system allocation efficiency and long-term productivity gains. Given such regional heterogeneity, uniform national bed allocation policies cannot fit diverse local resource endowments, and region-specific differentiated governance should be implemented.

Authors

Institutions

Publication Details

Journal
Frontiers in Public Health
Published
2026-11-18
DOI
https://doi.org/10.3389/fpubh.2026.1991772
Primary Topic
Efficiency Analysis Using DEA
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

The nonlinear relationship between provincial hospital bed scale and the pure technical efficiency of the healthcare system and its regional heterogeneity: evidence from pure technical efficiency, DEA-Malmquist, and a two-way fixed-effects model

Xiaoling Wei, Yong Liu, Qingwei Guo, Huifang Liang et al.
Frontiers in Public Health
Efficiency Analysis Using DEA
article

The nonlinear relationship between provincial hospital bed scale and the pure technical efficiency of the healthcare system and its regional heterogeneity: evidence from pure technical efficiency, DEA-Malmquist, and a two-way fixed-effects model

Xiaoling Wei, Yong Liu, Qingwei Guo, Huifang Liang, Wenjin Cai
article en

Abstract

Background Hospital beds constitute an important material resource of healthcare systems and profoundly affect regional medical supply and service efficiency. Since the 2009 healthcare reform, hospital bed scale has continued to expand; however, supporting human resources and technical capabilities have lagged behind, highlighting prominent regional allocation-related contradictions. Existing empirical studies lack systematic provincial-level evidence to support quantitative references for differentiated bed allocation policies. Accordingly, this study analyzes the nonlinear relationship between the number of hospital beds and the pure technical efficiency (PTE) of provincial healthcare systems, as well as their regional heterogeneity. Methods Using balanced panel data covering 31 Chinese provinces from 2014 to 2023, hospital beds were excluded from input indicators in Data Envelopment Analysis (DEA) to calculate the pure technical efficiency of provincial healthcare systems; this indicator represents the relative efficiency level obtained under the input-output assumptions specified in this study. The Malmquist index was adopted to analyze dynamic changes in total factor productivity change (TFP). A two-way fixed-effects model containing the quadratic term of the logged bed scale was constructed. The Lind-Mehlum test was used to identify nonlinear relationships. Subgroup regressions for eastern, central, and western regions, as well as robustness tests, were performed. Results National healthcare TFP declined by an average of 2.2 % annually from 2014 to 2023, with an annual mean technical change (TC) of 0.980 and technical efficiency change (EC) of 0.998. Under the baseline model using the provincial absolute bed scale, the provincial bed scale exhibits a U-shaped association with pure technical efficiency, with a sample-based statistical turning point of approximately 190,200 beds. This turning point is an in-sample statistical estimate and cannot be directly used as a universal policy threshold. Observations show that an increase in the number of hospital beds on the left side of the inflection point is accompanied by a decline in pure technical efficiency; after passing the inflection point, efficiency rises, but the marginal gain is slight. Based on a full-sample interaction analysis, regional heterogeneity was statistically significant overall, with the heterogeneity primarily observed between the western region and the eastern and central regions. Conclusion The identified U-shaped association and sample-estimated turning point challenge conventional linear assumptions and provide quantitative references for provincial-level hospital bed planning. Combining the results of static efficiency regression with those of dynamic total factor productivity decomposition reveals that the decline in national total factor productivity in the healthcare sector is primarily driven by insufficient technological progress; extensive bed expansion struggles to simultaneously ensure both current system allocation efficiency and long-term productivity gains. Given such regional heterogeneity, uniform national bed allocation policies cannot fit diverse local resource endowments, and region-specific differentiated governance should be implemented.

Frontiers in Public HealthVol. 14
Shandong University (CN), Shandong Provincial Hospital (CN)
Good health and well-being, Industry, innovation and infrastructure
Efficiency Analysis Using DEA
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon