| Objective: The basic information and medical expenditures for stroke inpatients with medical insurance in Tianjin from 2003 to 2013 were obtained from the Tianjin Health Insurance Bureau. To understand the different types of stroke, the hospital costs of stroke inpatients were compared. Multiple linear regression method was applied to analyze the associations between the total hospital charge and the possible influencing factors of the patients with stroke. In addition, we used the time series model to explore and forecast the hospitalization costs trend. The results can be used as a reference for the government to control the growth of hospital costs in patients with stroke reasonably and for the related departments to develop the health insurance policy and improve the medical insurance system.Methods: All the information of inpatients was collected from the Tianjin Health insurance database with 50 percent random sampling from 2003 to 2013. 200,561 stroke patients were included. The basic characteristics and medical expenses of the different types of patients were described and compared. Multiple linear regression method was used to analyze the possible factors which may have influence on the total hospital costs. Autoregressive integrated moving average(ARIMA) models were fitted with the monthly per capita cost of hospitalization from January 2003 to December 2013. The predictive effect was evaluated.And the models were finally applied to predict the costs of hospitalization in patients with stroke.Results: Among the 200,561 patients, 130,653 were male with the mean age 65.65±11.25 years and 69,908 were female with the mean age 66.88±10.46 years. 178,708 patients suffered from ischemic stroke, which accounted for nearly 90% and 21,853 patients suffered from hemorrhagic stroke. Most patients were older than 60 years and tended to visit tertiary hospitals. The majority of patients were at hospital length for 8-14 days. Compared with patients suffered from ischemic stroke, more patients with hemorrhagic stroke were choosing the surgical treatment. The hospitalization costs of insured patients increased in the past 11 years, and the costs of patients with hemorrhagic stroke were higher than ischemic stroke patients. During that period, the hospital costs almost no changed in the structure in which frees for drugs took the highest proportion and followed by treatment and examination costs. The results of multiple linear regression analysis showed that length of stay, whether surgery, hospital level, personnel category and gender were the common influencing factors of hospital costs of ischemic and hemorrhagic stroke patients. While age and insurance counties were the independent factors of the hospitalization expenses of patients with ischemic stroke. The ARIMA model was used to fit the hospitalization expenses of all stroke patients and different types of stroke patients, and the annual average relative errors were 2.27%, 2.37% and 7.44% respectively.Conclusions: There were more patients with ischemic stroke than patients with hemorrhagic stroke in Tianjin. More attentions should be paid to the elderly patients with ischemic stroke. Most patients tended to visit the high-level medical institutions. Community medical institutions should be further developed to alleviate the pressure of tertiary hospitals. The structure of hospitalization costs was unreasonable. It was a key to take effective measures to control the drug costs, thus reining the total hospitalized expense. Corresponding policies should be made according to the disease characteristics. Hospital costs of patients with stroke presented upward trend overall and the trend would not change in the short term, which provided basis and reference for the relevant departments to monitor and control health care costs effectively. And it also contributed to reduce health care costs, lighten the economic burden on patients and improve the social medical insurance system. |
PDF Full Text Request