Background Fractures are among the most common traumatic injuries in China, with rising incidence driven by population aging, traffic accidents, and sports injuries. They impose a heavy economic burden due to high treatment costs and prolonged rehabilitation. To improve cost efficiency, China launched a nationwide Diagnosis-Related Groups (DRG) payment reform in 2019. While DRG has shown positive effects in controlling costs for chronic diseases, its impact on trauma-related conditions like fractures remains unclear due to clinical complexity and treatment variability. Existing research mostly focuses on epidemiology, lacking economic evaluations under the DRG system. This study aims to assess how DRG reform influences the inpatient cost structure of fracture patients and explore differential effects across comorbidity and fracture types, providing evidence for more refined payment strategies in trauma care. Methods Using data from 12,101 hospitalized fracture patients (ICD-10 codes S22–S92) admitted to a tertiary hospital in Anshan, Liaoning Province between 2018 and 2024, we conducted a structural change analysis to assess shifts in the composition of inpatient costs before and after the introduction of the DRG payment system. An interrupted time series (ITS) model was applied to estimate both the immediate impact and the longitudinal trend changes associated with the DRG reform initiated on July 1, 2019. In addition, gray relational analysis was employed to further examine the relative contribution of different cost categories to overall expenditure. Results A total of 12,101 fracture inpatients were included, covering the entire period before and after the DRG reform. Significant changes in the cost structure were observed post-implementation. The median drug cost decreased from 3,416.06 CNY to 2,796.74 CNY (a reduction of 18.1%). Although the proportion of consumables costs slightly increased (median rose from 7,358.12 CNY to 7,465.64 CNY), the growth rate significantly slowed ( p 0.05). Meanwhile, therapeutic costs (median increased from 1,015.38 CNY to 1,200.91 CNY) and the proportion of surgical fees rose, indicating a shift of medical resources toward technical services under DRG. Rehabilitation costs declined in certain fracture types (e.g., femoral fractures, S72), but increased in others (e.g., lower leg fractures, S82), reflecting DRG’s differential effects on treatment stages. Structural variation analysis showed the greatest fluctuation in consumables costs in spinal fracture cases (S32 group, DsV = 2170.42%), while drug costs significantly declined in the S72 group (DsV = −39.78%). Patients with comorbidities experienced more pronounced structural adjustments—for example, the structural variation in the hypertensive group was 15.3% higher than that in the non-comorbidity group ( p 0.01), suggesting stronger cost-control effects of DRG in complex cases. ITS analysis revealed that the DRG reform had a significant impact on costs across various fracture types ( p 0.05). For total costs, S32 fractures exhibited a reversal from a pre-policy increasing trend (β₁ = 1247.93) to a rapid decreasing trend (β₃ = −2467.0). After the DRG implementation, diagnostic costs showed an increasing trend in most fracture types, while decreasing significantly in S32 fractures (β₃ = −227.16); in contrast, S72 fractures demonstrated a notable increase (β₃ = 52.86). Treatment and medication costs generally displayed decreasing trends, with the most pronounced decline observed in medication costs for S32 fractures (β₃ = −355.1). Consumables costs exhibited a divergent pattern, characterized by an anomalous increasing trend in S42 fractures (β₃ = 1578.62). Rehabilitation costs showed a universal decreasing trend, with the most significant control effect seen in S32 fractures (β₃ = −483.58). Gray relational analysis indicated that, before and after DRG implementation, the cost struetures of different fracture types exhibited distinct patterns of change. In S22 and S42 fractures, the correlation coefficients of diagnostic and drug-related costs increased notably, with all categories in S42 rising to 0.89–0.90, reflecting a highly concentrated cost structure. S32 and S82 fractures showed overall stability or slight increases across cost categories. In contrast, S52 and S62 fractures demonstrated a general decline, particularly in therapeutic, consumable, and rehabilitation costs. S72 fractures remained relatively stable, with minimal fluctuations in correlation coefficients. S92 fractures displayed increases across all cost categories with balanced magnitudes, indicating a comprehensive enhancement. Overall, different fracture types exhibited distinct patterns of cost structure adjustment following DRG implementation. Conclusion The DRG-based payment reform effectively controlled pharmaceutical expenditures while increasing diagnostic costs, leading to fracture-type-specific shifts in treatment structure and highlighting the need for differentiated management strategies. This study provides empirical evidence to support the optimization of DRG payment standards and the advancement of healthcare payment reform.
Wang et al. (Mon,) studied this question.