基于干预分析模型的三级医院重点监控高值耗材影响因素研究

Study on influencing factors of high-value consumables in tertiary hospitals based on intervention analysis model

目的 利用漳州市某三甲医院人均重点监控高值耗材收入数据建立干预分析ARIMAX模型,寻找影响人均重点监控高值耗材的影响因素及影响程度。方法 收集2018年1月—2024年1月按月为单位的人均重点监控高值耗材数据和三四级手术占比数据,并分别设为因变量y和自变量x_1,同时分别设定哑变量x_2和x_3,分别代表国家耗材集采政策变量和DIP支付改革政策变量,其中x_2以2021年1月为界,在此之前未执行国家耗材集采,赋值为0,在此之后实行了国家耗材集采,赋值为1,x_3变量以2022年10月为界,在此之前未实行医保支付政策改革,赋值为0,之后实行了医保政策改革,赋值为1。利用干预分析模型建立人均重点监控高值耗材数据分析模型,确定影响人均重点监控高值耗材的影响因素及影响程度。结果 以BIC最低为原则确定的最佳模型为ARIMA(1,0,0)模型,国家集采可以为患者降低10.75%重点监控高值耗材,医院三四级手术占比每上升1个百分点,将会使人均重点监控高值耗材提高2.16个百分点,医保支付政策改变并没有降低重点监控高值耗材使用,该政策在模型中没有统计学意义。结论 国家集采将是今后进行耗材管控的重要途径,三四级手术占比对人均重点监控高值耗材的影响显著。

Objective To establish an intervention analysis ARIMAX model using the per capita key monitored high-value consumables revenue data of a tertiary hospital in Zhangzhou City,and to find the influencing factors of per capita key monitored high-value consumables and their impact levels.Methods The data of monthly per capita key monitored high-value consumables and the proportion of third-and fourth-level surgeries those were created during January 2018 and January 2024 were collected,and the data were set as dependent variable y and independent variable x 1.Meanwhile,dummy variables x 2 and x 3 were also defined,representing the national consumables centralized procurement policy variable and the DIP payment reform policy variable,respectively.Taking January 2021 as the boundary,at that time,the boundary the national consumables centralized procurement was not implemented,thus x 2 was given the value of 0,and then the national consumables centralized procurement was implemented,so the x 2 was given the value of 1.The x 3 variable was set to 0 before October 2022,when the medical insurance payment policy reform was not implemented,and to 1 after October 2022,when the medical insurance payment policy reform was implemented.The intervention analysis model was used to establish a per capita key monitored high-value consumables data analysis model,and determine the influencing factors of per capita key monitored high-value consumables and their impact levels.Results Based on the principle that the BIC was the lowest,ARIMA(1,0,0)was the best model.Adopting national centralized procurement of consumables could reduce the use of key monitored high-value consumables by 19.8%for patients,but every 1%increase in the proportion of third-and fourth-level surgeries in the hospital would increase the use of key monitored high-value consumables by 2.16%.The change in medical insurance payment policy had not reduced the use of per capita high-value consumables,and this policy had no statistical significance in the model.Conclusions National centralized procurement will be an important way for consumables control in the future,and the proportion of third-and fourth-level surgeries has a significant impact on per capita key monitored high-value consumables.