Email Alert | RSS

Chinese Journal of Antituberculosis ›› 2023, Vol. 45 ›› Issue (11): 1084-1089.doi: 10.19982/j.issn.1000-6621.20230127

• Original Articles • Previous Articles     Next Articles

Research on the factors influencing tuberculosis hospitalization costs and diagnosis related grouping

Yan Xiaojing1, Nie Wenjuan2, Liu Haiyan3, An Jun4, Chi Yuqing3, Chu Naihui2()   

  1. 1Quality Control Office, Medical Service, Beijing Chest Hospital, Capital Medical University, Beijing 101149, China
    2Department Ⅰ of Tuberculosis, Beijing Chest Hospital, Capital Medical University, Beijing 101149, China
    3Medicare Office,Beijing Chest Hospital, Capital Medical University, Beijing 101149, China
    4Medical Records Department,Beijing Chest Hospital, Capital Medical University, Beijing 101149, China
  • Received:2023-04-21 Online:2023-11-10 Published:2023-11-03
  • Contact: Chu Naihui, Email: dongchu1994@sina.com

Abstract:

Objective: Based on the research on the influencing factors of hospitalization costs, explore the scheme of diagnosis related groups (DRG) of tuberculosis under the classification of tuberculosis, determine the grouping according to the cost, and provide a basis for reasonable grouping of DRG. Methods: We collected 1812 confirmed patients from Beijing Chest Hospital affiliated to Capital Medical University from January 1, 2018 to December 31, 2020. According to the classification of pulmonary tuberculosis, the basic group level was divided into two groups: sensitive pulmonary tuberculosis and drug-resistant pulmonary tuberculosis. There were 109 cases in the sensitive pulmonary tuberculosis group and 1703 cases in the drug-resistant pulmonary tuberculosis group. Statistical description of the median and quartile of hospitalization expenses for patients under different factors was done. Kruskal Wallis H-test, Mann Whitney U-test, and multiple linear regression methods were used to analyze the factors affecting hospitalization expenses. P<0.05 indicates a statistically significant difference. Using decision trees to process data to complete the division of sub-groups. The coefficient of variation (CV) and reduction in variance (RIV) were used to evaluate intra group homogeneity and inter group heterogeneity between the groups, and the Kruskal Wallis H test was used to test the group results. Results: We analyzed the hospitalization expenses of patient groups under different factors. It was found that the median (quartile) hospitalization expenses for patients aged ≤30 years old were 20586.66 (13656.41,90163.58) yuan, 23256.47 (17490.26,82919.32) yuan for those aged 31-40 years old, 21154.19 (13615.14,76240.28) yuan for those aged 41-50 years old, 23544.59 (16325.46,87608.46) yuan for those aged 51-60 years old, 25190.27 (20528.01,78321.02) yuan for those aged 61-70 years old, and 24890.24 (18424.86,91662.89) yuan for those aged ≥71 years old. The median (quartile) hospitalization cost for drug sensitive patients was 21693.82 (16567.26,91662.89) yuan, 21866.32 (14390.17,87896.56) yuan for multidrug-resistant patients, 23703.70 (18021.55,90163.58) yuan for monodrug-resistant patients, 18846.98 (11400.11,78321.02) yuan for extensively resistant patients, and 23900.67 (16006.08,58834.03) yuan for multidrug-resistant patients. The median (quartile) hospitalization cost for patients with 0 complications was 9730.37 (5361.33,28053.18) yuan, 20378.65 (14110.72,90163.60) yuan for patients with 1-5 complications, 24072.81 (17966.19,87608.46) yuan for patients with 6-10 complications, 24953.26 (18411.59,87896.56) yuan for patients with 11-15 complications, 32530.70 (23904.41,74195.40) yuan for patients with 16-20 complications, and 29647.02 (25707.11,91662.89) yuan for patients with ≥21 complications. The median (quartile) hospitalization cost for newly treated patients was 23594.24 (17890.74,91662.89) yuan, while for re-treated patients it was 21662.63 (14369.58,87896.56) yuan. Kruskal Wallis H-test showed that there was statistical significance in hospitalization expenses between groups after grouping according to age, drug-resistance type, and different complications (H-values were 58.510, 19.556, and 92.424, respectively, P<0.05). The Mann Whitney U-test showed that there was a statistically significant difference in hospitalization expenses between groups after initial/recurrent treatment (U=-4.422, P<0.05). The presence of complications, type of initial and recurrent treatment, length of hospital stay, and age were all included in the regression model. At the sub-group level, the DRG group of pulmonary tuberculosis was divided into 5 groups: drug sensitive pulmonary tuberculosis, drug-resistant pulmonary tuberculosis with hospitalization time ≤7 days, drug-resistant pulmonary tuberculosis with hospitalization time 8-14 days, drug-resistant pulmonary tuberculosis with hospitalization time 15-20 days, and drug-resistant pulmonary tuberculosis with hospitalization time ≥21 days. The median (quartile) hospitalization cost for drug sensitive pulmonary tuberculosis group was 21693.82 (16567.26,91662.89) yuan. While for drug-resistant pulmonary tuberculosis with hospitalization time ≤7 days group, the median (quartile) hospitalization cost was 12334.88 (7141.91,59741.83) yuan. For drug-resistant pulmonary tuberculosis with hospitalization time 8-14 days, the median (quartile) hospitalization cost was 20842.28 (16459.27,80496.43) yuan. The median (quartile) hospitalization cost for drug-resistant pulmonary tuberculosis with hospitalization time 15-20 days was 26612.13 (21683.16,90163.58) yuan. The median (quartile) hospitalization cost for drug-resistant pulmonary tuberculosis with hospitalization time ≥21 days was 33326.87 (25680.93,87896.56) yuan. The Kruskal Wallis H-test results showed an H-value of 660.241, with a P<0.05, indicating a statistically significant difference between the groups. The CV values of each group were less than 1, which met the DRG grouping criteria. From the basic group divided by sensitive drug-resistance type to the sub-group level, the RIV value increased from 0.01% to 33.82%, with a significant increase in differences. Conclusion: Introducing elements of drug-resistance type and treatment type into grouping elements was conducive to closely combining clinical similarity and resource consumption similarity, improving the accuracy and scientificity of grouping. It is recommended to group according to different types of pulmonary tuberculosis and determine payment standards based on this to improve the scientificity of DRG grouping and payment.

Key words: Tuberculosis, pulmonary, Diagnosis-related groups, Decision trees

CLC Number: