Does comprehensive public hospital reform influence the healthcare costs of inpatients with pulmonary tuberculosis? An interrupted time series analysis in western China

In December 2016, the Chinese government launched a comprehensive reform of public hospitals, including the zero-markup drug policy (ZMDP) and adjustments to the price compensation mechanism for medical services. This study aimed to assess the impact of the reform on hospitalization and out-of-pocket (OOP) costs for pulmonary tuberculosis (PTB) patients. This study collected data on healthcare costs from the medical insurance database of Luzhou City, China from 2015 to 2018. An interrupted time series (ITS) analysis was used to assess monthly changes in hospitalization and OOP costs across different basic medical insurance (BMI) groups before and after the reform. After the reform, the median total cost per visit, median cost of medicines and materials per visit, median OOP cost per visit, and proportions of medicine cost and OOP cost decreased significantly, whereas the median medical service cost per visit increased significantly ( P < 0.001). The total cost per visit, costs of medicines, examinations, materials, and medical services per visit, and OOP cost per visit showed long-term monthly downward trends. Compared with the Urban and Rural Resident Basic Medical Insurance (URRBMI) group, hospitalization costs in the Urban Employee Basic Medical Insurance (UEBMI) group showed larger variation. The OOP cost proportion in the URRBMI group decreased by 3.94% immediately after the reform was implemented ( P < 0.01), whereas the level change in the UEBMI group was insignificant ( P > 0.05). The comprehensive reform of public hospitals played a positive role in optimizing the structure of medical costs and reducing the total costs for PTB patients. However, the long-term policy effects were limited by deficient price assessment and compensation mechanisms for medical services. OOP costs and OOP proportion decreased significantly but fell short of the expected target. The inconsistent responses of different BMI beneficiaries may be associated with inequalities in the BMI benefits.