When Patients Cross City Lines, Public Hospitals Pay an Efficiency Price

Every year, millions of Chinese patients board trains, buses, and private cars to seek inpatient care far from home. They travel to provincial capitals and megacities where they believe the best surgeons, the newest equipment, and the most trusted diagnoses await. A new study from Guangdong Province suggests that this mass movement of patients is not just a logistical curiosity. It is quietly reshaping the operational performance of the public hospitals left behind, and the hospitals that lose patients may be paying a measurable efficiency price.

The research, published in BMC Public Health by a team from the School of Public Health at Shanghai University of Traditional Chinese Medicine, examined the relationship between the out-of-city hospitalization rate and the efficiency of public hospitals across Guangdong, China’s most populous province. Using balanced panel data covering all 21 prefecture-level cities from 2018 through 2023, the researchers found that cities with higher shares of residents hospitalized outside their home city tended to operate their public hospitals less efficiently. The statistical association was negative and significant, with a coefficient of -2.492 at a p-value below 0.05, meaning that rising patient outflow was reliably linked to declining efficiency scores.

What makes the study technically interesting is how efficiency was measured. The authors turned to data envelopment analysis, a non-parametric frontier method that benchmarks each decision-making unit, in this case the public hospital sector of a city, against an efficiency frontier constructed from the best performers. Rather than the classical Charnes-Cooper-Rhodes radial model, they employed a slacks-based measure, or SBM, which directly accounts for input excesses and output shortfalls. This matters because radial models can overstate efficiency when inputs and outputs do not contract or expand proportionally. The SBM formulation captures slack in every dimension simultaneously, producing efficiency scores that better reflect real-world managerial waste.

The efficiency picture was then decomposed into two conceptually distinct components. Pure technical efficiency reflects how well a hospital system converts its current resources, such as health technicians and medical beds, into services like outpatient visits and discharges, holding scale constant. Scale efficiency, by contrast, captures whether a system operates at its optimal size, where returns to scale are neither increasing nor decreasing. When the researchers ran two-way fixed effects panel regressions, the negative association with out-of-city hospitalization proved far stronger for pure technical efficiency, with a coefficient of -3.322 and a p-value below 0.01, than for scale efficiency. In plain terms, patient outflow appears to erode the day-to-day managerial effectiveness of local hospitals more than it disrupts their optimal size.

This asymmetry carries a compelling mechanistic interpretation. When a city’s sickest and most demanding patients leave, the local hospitals that remain lose case volume, clinical complexity, and the learning opportunities that come with challenging cases. Staff skills may stagnate, equipment utilization falls, and fixed costs spread across fewer services. Yet the physical infrastructure, the beds, the buildings, the payroll, largely stays put. The result is a system whose management efficiency deteriorates even as its scale characteristics remain relatively unchanged. The study’s finding that scale efficiency is comparatively spared fits this picture: the hospitals are not necessarily too big or too small, but what happens inside them runs less smoothly.

The two-way fixed effects design strengthens the causal credibility of the association. By including city fixed effects, the models absorb all time-invariant characteristics of each city, from geography to entrenched referral cultures. Year fixed effects absorb province-wide shocks, such as the COVID-19 pandemic’s disruption of hospital operations across 2020 to 2022, which affected all cities simultaneously. The remaining variation, the year-to-year changes in out-of-city hospitalization within each city, is what drives the estimated coefficient. This within-city identification strategy rules out many confounding explanations that plague cross-sectional comparisons, although the authors are careful to frame the result as an association rather than proof of causation.

Heterogeneity analysis added a crucial nuance. The negative association between patient outflow and hospital efficiency was weaker in cities with relatively abundant healthcare resources, measured through indicators such as the number of public hospitals, health technicians per 1,000 population, and medical beds per 1,000 population. This pattern suggests that well-endowed cities possess a buffer. Their hospitals have the clinical depth, the specialist capacity, and the institutional resilience to remain efficient even when some patients depart. In resource-poorer cities, by contrast, the loss of patients strips away capacity that cannot easily be replaced, and efficiency suffers more visibly.

The broader context is China’s tiered healthcare system, a policy architecture designed to route common conditions to community and county facilities while reserving tertiary hospitals for complex care. Patient mobility across city lines is the system’s pressure valve and its stress test simultaneously. On one hand, allowing patients to seek care where they trust it most respects patient choice and can drive quality competition. On the other hand, when reimbursement policies make out-of-city care financially painless, the tiered system’s local foundations erode. The authors argue that aligning non-local reimbursement policies with tiered healthcare objectives, strengthening regional healthcare coordination, and enhancing local service capacity may help break the negative feedback loop between patient outflow and local efficiency.

Guangdong makes an ideal laboratory for this question. The province spans staggering inequality, from the global manufacturing hub of the Pearl River Delta to rural cities in the mountainous north and east. Patients in less developed cities face long journeys to Guangzhou and Shenzhen for advanced care, and the out-of-city hospitalization rate varies widely across the province’s 21 cities. The panel structure, six years of data across all cities, allowed the researchers to exploit this variation while controlling for economic factors such as per capita GDP, population density, and per capita disposable income, all of which independently shape both healthcare demand and hospital performance.

The study’s implications reach beyond China. Cross-border patient mobility is a growing phenomenon worldwide, from European cross-border care directives to medical tourism in Southeast Asia. If the Guangdong findings generalize, they warn that health systems losing patients to richer neighbors may enter a self-reinforcing spiral: patients leave, efficiency falls, quality perceptions worsen, and more patients leave. Breaking that spiral, the authors suggest, requires not walls but capacity, coordination, and reimbursement rules that give local hospitals both the resources and the incentives to keep and win back the patients they can treat well. For the famous science magazine reader, the message is starkly quantitative: every percentage point of patients crossing city lines leaves a statistical fingerprint on the efficiency frontier of the hospitals they leave behind.

Subject of Research: The association between out-of-city hospitalization rates and public hospital efficiency in Guangdong, China

Article Title: Association between out-of-city hospitalization and public hospital efficiency: a panel study from Guangdong, China

Article References: Ge, Y., Xie, Y., Sun, H., Ai, H., Gu, S., & Lu, Y. (2026). Association between out-of-city hospitalization and public hospital efficiency: a panel study from Guangdong, China. BMC Public Health. https://doi.org/10.1186/s12889-026-29477-5

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29477-5

Keywords: out-of-city hospitalization, public hospital efficiency, data envelopment analysis, slacks-based measure, pure technical efficiency, scale efficiency, patient mobility, tiered healthcare, Guangdong, China, health economics, panel study