Surgical treatment of liver cancer and pancreatic cancer under the China Healthcare Security Diagnosis Related Groups payment system

Abstract

BACKGROUND

Data from the World Health Organization’s International Agency for Research on Cancer reported that China had the highest prevalence of cancer and cancer deaths in 2022. Liver and pancreatic cancers accounted for the highest number of new cases. Real-world data (RWD) is now widely preferred to traditional clinical trials in various fields of medicine and healthcare, as the traditional research approach often involves highly selected populations and interventions and controls that are strictly regulated. Additionally, research results from the RWD match global reality better than those from traditional clinical trials.

AIM

To analyze the cost disparity between surgical treatments for liver and pancreatic cancer under various factors.

METHODS

This study analyzed RWD 1137 cases within the HB1 group (patients who underwent pancreatectomy, hepatectomy, and/or shunt surgery) in 2023. It distinguished different expenditure categories, including medical, nursing, technical, management, drug, and consumable costs. Additionally, it assessed the contribution of each expenditure category to total hospital costs and performed cross-group comparisons using the non-parametric Kruskal–Wallis test. This study used the Steel–Dwass test for post-hoc multiple comparisons and the Spearman correlation coefficient to examine the relationships between variables.

RESULTS

The study found that in HB11 and HB13, the total hospitalization costs were significantly higher for pancreaticoduodenectomy than for pancreatectomy and hepatectomy. Although no significant difference was observed in the length of hospital stay between patients who underwent pancreaticoduodenectomy and pancreatectomy, both were significantly longer than those who underwent liver resection. In HB15, no significant difference was observed in the total cost of hospitalization between pancreaticoduodenectomy and pancreatectomy; however, both were significantly higher than those in hepatectomy. Additionally, the length of hospital stay was significantly longer for patients who underwent pancreaticoduodenectomy than for those who underwent pancreatectomy or liver resection.

CONCLUSION

China Healthcare Security Diagnosis Related Groups payment system positively impacts liver and pancreatic cancer surgeries by improving medical quality and controlling costs. Further research could refine this grouping system and ensure continuous effectiveness and sustainability.

Key Words: China health care security diagnosis-related groups; Real-world study; Liver cancer surgical treatment; Pancreatic cancer surgical treatment; Hospitalization costs; Cost structure; Average length of stay

Core Tip: To analyze cost differences and their influencing factors in surgical treatments for liver and pancreatic cancer within the China Healthcare Security Diagnosis Related Groups (CHS-DRG) payment system, this study aims to examine the appropriateness of the current diagnosis-related groups groupings and provide recommendations for improvement. The CHS-DRG payment system has shown a positive impact on the surgical treatment of liver and pancreatic cancer by enhancing medical quality and managing costs. However, additional research is necessary to refine the grouping system and ensure its continued effectiveness and sustainability.

INTRODUCTION

The China Healthcare Security Diagnosis Related Groups (CHS-DRG) is a tool for clinical management based on the first page of medical records, considering individual characteristics such as disease diagnosis, treatment methods, comorbidities, complications and age of cases, and then dividing diseases into several disease diagnosis groups according to the principle of similar clinical pathways and similar resource consumption. The CHS-DRG payment system plays a vital role in reforming the payment modalities for health insurance. Its main objectives include controlling medical expenditures, improving the efficiency of medical services, standardizing medical service practices, and optimizing the use of health insurance funds[1].

Real-world data (RWD) are research data that originate from the actual medical environment and reflect real diagnoses, treatment processes, and patient health statuses. This type of data differs from traditional clinical trials, in which populations are often highly selected and interventions and controls are strictly regulated. Apparent differences exist in various aspects, such as follow-up procedures and real-world conditions. Based on the data source, RWD can be divided into research and non-research data[2]. Owing to research results from the RWD matching reality and their extrapolation potential, RWD is now widely used in various fields of medicine and healthcare[3].

The latest data from the World Health Organization’s International Agency for Research on Cancer on the global cancer burden indicate that China had the highest number of new cancer cases and deaths in 2022. Liver and pancreatic cancers were the most common cancers in China and constituted the top list of new cases[4]. These two cancers also have the lowest 5-year survival rates[5]. The surgical treatment of hepatobiliary and pancreatic tumors is complicated, the overall cost is relatively high, and the economic burden of patients is relatively heavy. Establishing a scientific and reasonable diagnosis-related groups (DRGs) grouping scheme for hepatobiliary and pancreatic tumors is particularly important.

This study used RWD to focus on a tertiary cancer specialty hospital in Beijing. The analysis included using the non-parametric Kruskal–Wallis test for comparing data between multiple groups and the non-parametric Tukey Steel–Dwass test for post-hoc multiple comparisons. Spearman’s correlation coefficient was used to assess the relationships between variables. This study investigated the differences in hospital costs and influencing factors for patients undergoing surgical treatment for liver and pancreatic cancer within the pancreatic and liver resection and/or shunt surgery groups under the CHS-DRG payment system. It evaluated the current DRG grouping scheme and proposed recommendations to improve the scientific and rational aspects of DRG payment standards. By aligning with clinical practice and promoting the efficient allocation and utilization of medical resources, this study aimed to facilitate collaboration among health insurance providers, hospitals, and healthcare providers, thereby enabling patients to benefit from the implementation of DRGs.

MATERIALS AND METHODS

Research objects

This study utilized data from the Beijing Region Inpatient Medical Service Performance Evaluation Platform, which collected information from the first page of patient records from January to December 2023. This information included details such as the length of stay, type of surgery, complications, and costs. Following the CHS-DRG 1.1 version of the grouping scheme, the study focused on 1137 patients in the HB1 group who underwent pancreatic and liver resection and/or shunt surgery. Surgical methods, including liver resection, pancreatectomy, and pancreaticoduodenectomy, were categorized based on inclusion criteria and surgical site. Patients were further categorized into subgroups based on complications/comorbidities during hospitalization: 462 HB11 patients with severe complications or comorbidities, 497 HB13 patients with severe complications or comorbidities, and 178 HB15 patients without complications or comorbidities.

This study analyzed the cost structure of inpatients by distinguishing between medical, nursing, medical technology, administrative, drug, and consumption costs. It also evaluated the contribution of various cost categories to the total hospitalization costs. Strict data encryption and privacy protection measures were taken to ensure the confidentiality of patients’ personal information. The variables were carefully controlled to enhance the scientific validity and reliability of the results. Measures were taken to avoid potential blind spots in data collection and ensure high standards of data quality.

Statistical analysis

Variables such as length of stay and hospital costs demonstrated a skewed distribution. The non-parametric Kruskal–Wallis test was employed to compare multiple groups, followed by the non-parametric Steel–Dwass test for post-hoc multiple comparisons. The Spearman’s correlation coefficient was used to analyze the relationships among variables. All tests were two-sided, with a 5% significance level.

RESULTS

Overall situation

Table 1 compares the total cost of hospitalization and length of stay for the different surgical categories within the various DRG groups. In HB11, the total hospitalization cost for pancreaticoduodenectomy was significantly higher than that for pancreatectomy, which in turn was significantly higher than that for hepatectomy. The duration of hospitalization for pancreaticoduodenectomy was not significantly different from that for pancreatectomy; however, both were significantly longer than those for liver resection. In HB13, the total hospitalization costs for the three operations followed a similar pattern as in HB11, with pancreaticoduodenectomy being significantly higher than pancreatectomy and hepatectomy. A comparison of the hospitalization days among the three groups showed a similar trend. In HB15, no significant difference was observed in the total cost of hospitalization between pancreaticoduodenectomy and pancreatectomy; however, both were significantly higher than those in hepatectomy. Regarding the length of stay, pancreaticoduodenectomy was significantly longer than pancreatectomy and hepatectomy, whereas no significant difference was found between pancreatectomy and hepatectomy.

Table 1 Comparison of total hospitalization costs and length of stay for different surgical procedures in the same diagnosis-related group.

DRG	Surgery classification	Total hospitalization expenses			The number of days in hospital
		Median (quartiles)	χ2	P value	Median (upper and lower quartiles)	χ2	P value
HB11			158.75	< 0.0001		125.9	< 0.0001
	Liver resection	74463.5 (60025.5, 94356)3			9 (7, 12)2
	Pancreaticoduodenectomy	142663 (133720, 179562.25)1			20 (14.25, 24.75)1
	Pancreatectomy	102252.5 (95235, 119203.25)2			13.5 (11, 18.75)1
HB13			234.83	< 0.0001		131.83	< 0.0001
	Liver resection	63929 (51004.75, 78147)3			7 (7, 10.25)3
	Pancreaticoduodenectomy	136335 (120212, 165313.5)1			16 (12.5, 22)1
	Pancreatectomy	92788.5 (79258.5, 105594.25)2			10.5 (8, 14.25)2
HB15			44.60	< 0.0001		15.06	0.0005
	Liver resection	67274.5 (54285.25, 81836.25)2			9.5 (7, 13)2
	Pancreaticoduodenectomy	134561 (114775, 150848)1			19 (11, 22)1
	Pancreatectomy	82528 (76816, 95098)1			11 (8, 13.5)2

1, 2, and 3 are pairwise comparison result indicators; 1 > 2 > 3. If the letters are the same, the difference between the groups is statistically significant; if the letters are different, the difference between the groups is not statistically significant.

DRG: Diagnosis-related group.

For each patient, the proportion of individual expenditure categories in the total expenditure was calculated, and the mean value of the proportion in each group was determined. Table 2 presents the study’s results. In each DRG and surgical category, consumables and medicines accounted for the highest proportion of total expenditure, followed by medical expenditure. Nursing, medical technology, and management fees were the three categories with the lowest percentages.

Table 2 Proportion of various expenses under same diagnosis-related group group and same surgery, %.

DRG	Surgery classification	Average medical expense ratio	Average proportion of nursing expenses	Average proportion of medical technology expenses	Average management fee ratio	Average proportion of consumables expenses	Average proportion of drug expenses
HB11	Liver resection	12.39	1.09	3.05	3.05	38.59	37.45
	Pancreaticoduodenectomy	10.69	1.57	2.85	2.85	34.80	43.05
	Pancreatectomy	11.38	1.40	3.01	3.01	36.45	42.32
HB13	Liver resection	14.13	1.07	2.88	2.88	43.04	31.87
	Pancreaticoduodenectomy	10.95	1.49	2.65	2.65	38.23	40.81
	Pancreatectomy	12.66	1.25	2.69	2.69	40.03	38.44
HB15	Liver resection	13.55	1.15	3.00	3.00	43.13	31.94
	Pancreaticoduodenectomy	11.10	1.52	3.83	3.83	35.91	41.90
	Pancreatectomy	13.04	1.34	2.73	2.73	40.33	37.35

DRG: Diagnosis-related group.

The correlation analysis results indicated a significant positive correlation between total hospitalization costs, sub-item costs, and length of stay (Figure 1). The strongest correlation (0.94) was observed between total hospitalization expenses and drug costs. Nursing, medical, and consumable expenses exhibited correlations of 0.88, 0.83, and 0.80, respectively, with total hospitalization expenses. Medical technology fees exhibited a correlation of 0.79, whereas management fees demonstrated the lowest correlation at 0.74 with total hospitalization expenses. The correlation between the total hospitalization cost and length of stay was 0.76.

Figure 1 Hospitalization cost correlation matrix.

DISCUSSION

The DRG transforms medical service output into a relatively measurable model, thereby aiming to reduce medical expenses and resource consumption. It is currently recognized as a payment method that provides better control over the impact of hospitalization[6]. This study performed multidimensional factor analysis on big data from hospitals to reveal the composition and determinants of hospitalization costs for patients undergoing surgical treatment for liver and pancreatic cancers. These findings highlight the significant impact of the CHS-DRG system on China’s medical care system.

Recommendations on optimization of CHS-DRG pancreatic and hepatic resection and/or shunt surgery grouping plans

Statistical analysis revealed significant differences in hospital costs and length of stay between liver resection and pancreatectomy within the HB1 subgroups (HB11, HB13, and HB15). Particularly, the cost of pancreatic surgery was found to exceed that of liver surgery. Pancreatic surgery is widely recognized in clinical practice as one of the most challenging general surgeries, with a high risk of complications. Surgical approaches vary depending on the location of the disease and range from complex radical pancreaticoduodenectomy (such as the Whipple operation) to robotic procedures. These operations are associated with high technical complexity and high treatment costs.

The current CHS-DRG grouping scheme has limitations because it combines pancreatic and liver surgery, making it challenging to accurately assess the complexity of pancreatic surgery and its associated clinical resource consumption. This grouping also results in pancreatic surgery being categorized as a more cost-effective liver surgery, which affects the cost standards for pancreatic surgery and hinders progress in clinical technology. To address this issue, it has been proposed that liver and pancreatic surgeries be grouped separately based on their characteristics. Alternatively, pancreatic surgery can be grouped separately, with further refinement based on specific surgical methods to better capture the complexity and resource requirements of each procedure. By accurately reflecting the treatment complexity and resource requirements of different surgeries, the value of clinical services can be maximized, ultimately increasing the enthusiasm for clinical services. Regular review, adjustment of details concerning the grouping plan, and holding discussions about it is recommended to ensure alignment with evolving medical technology and clinical practice. Additionally, clinical experts should be involved in assessing its adaptability and scientific validity. This will help ensure that the CHS-DRG grouping plan remains in sync with clinical progress.

Factors affecting hospitalization costs for pancreatic and hepatic resection and/or shunt surgeries

With the continuous development of new clinical diagnostic and treatment technologies, using medical devices has become increasingly widespread. Strengthening the oversight of healthcare materials is essential to manage increasing costs[7]. According to surveys, DRG payments can reduce the overall cost of drugs for patients[8]. Statistical analysis of the composition of hospital costs in various subgroups of HB1 cases showed that drugs and supplies accounted for over 70% of the total costs. Specific analysis revealed that pancreaticoduodenectomy had the highest proportion of drug costs within each subgroup (HB11, HB13, and HB15 disease groups). The complexity of the procedure appears to be closely related to severe complications and the subsequent increase in drug costs; the group with severe complications (HB11) had the highest drug costs and a higher proportion of consumable costs. Drug costs for cases of serious surgical complications has significantly increased. Patients who underwent liver resection had the highest consumption costs. However, medical, administrative, and medical technology expenses account for a relatively low proportion, while consumable expenses such as drugs and consumables account for a relatively high proportion. Consequently, differences exist between the original intentions of the current medical care reform and the medical insurance benefit reform policies. The CHS-DRG aims to save medical resources, reduce treatment costs, and reduce unnecessary resource waste.

At the hospital’s internal management level, it is suggested that medical and insurance departments work with clinical departments to develop clinical pathways for specific diseases based on DRG disease groups and various surgical techniques. The implementation of clinical pathways has been proven to reduce patient hospitalization costs, reduce length of stay, and enhance medical care efficiency[9]. Improving the monitoring of the consumption of drugs and supplies, promoting the appropriate clinical use of these resources, and discouraging unnecessary tests and treatments to reduce hospitalization costs, is recommended to optimize medical resource allocation, and maximize benefits. It is also recommended that payment standards for complex pancreaticoduodenectomy and/or shunt surgery within the DRG group for pancreatic procedures should improve, particularly for pancreaticoduodenal surgery, and that a flexible adjustment mechanism for payment standards be established. Furthermore, it is recommended that the medical service department and medical insurance collaborate with clinical departments to develop clinical pathway management for relevant diseases. They must monitor the use of drugs and supplies, promote rational clinical practices, reduce unnecessary tests and treatments, reduce hospitalization costs, and optimize the allocation of medical resources to maximize efficiency and benefits.

Impact of severe complications or comorbidities on hospitalization costs for pancreatic and hepatic resection and/or bypass surgery

Analysis of the total hospitalization costs and length of stay in the three HB1 subgroups revealed that severe complications were more common after pancreatic and liver resection and/or shunt surgery than in patients with general complications or those without complications (subgroups HB3 and HB5). Patients with HB11 disease (subgroup HB11) incurred significantly higher costs and had longer hospital stays. Although advances in surgical techniques have decreased perioperative mortality following pancreatectomy, the incidence of complications remains high[10]. Studies have reported perioperative complication rates of 25%–40%[11]. Preventing and effectively managing serious complications are essential for reducing patients’ length of hospital stay, costs, and postoperative recovery quality. Medical departments should establish appropriate standards and indicators for patients undergoing pancreatectomy and liver resection, conduct thorough preoperative assessments, execute precise surgical procedures, closely monitor postoperative progress, and implement robust perioperative management strategies. Tailored cost estimates and prognoses are critical for patients with severe complications or comorbidities.

To reduce the costs of drugs and medical consumables, emphasis should be placed on the use of centrally procured medicines and consumables by the state while ensuring clinical effectiveness. Implementing a system of evaluation indicators, introducing performance appraisal, adopting a risk-adjustment model, and considering the impact of serious complications and comorbidities on costs can help achieve a more equitable and rational cost structure. Through these measures, clinical departments can raise their awareness of its importance, improve their level of management, and better plan their medical resources.

CONCLUSION

Under the CHS-DRG payment system, hospital costs are influenced by factors such as the complexity of surgery and the presence of serious complications or comorbidities. By refining the CHS-DRG grouping scheme with the help of improved management, technological advancements, and personalized assessments, we can more effectively address the specific requirements for treating liver and pancreatic cancers. This approach is expected to help manage hospitalization costs appropriately without compromising medical quality. Continued research and practical implementation will contribute to enhancing the CHS-DRG payment system and ultimately improving the efficiency and equity of medical services.