Bao-Rong Yu1. 1. Health Care Economics, Graduate School, Tokyo Medical and Dental University, Japan. yuhce@tmd.ac.jp
Abstract
BACKGROUND: It is unknown whether insurance status influences care provided and patients' prognosis, in China. METHODS: This retrospective cohort study included medical records of 4,714 patients with acute myocardial infarction aged 20 and older, discharged consecutively from 14 Chinese hospitals between January 2000 and February 2003. Uni-variate analysis, multivariate logistic regression and linear regression were used to compare differences in patients' characteristics, care provided and prognosis between insured patients and the uninsured. RESULTS: The uninsured were more likely to be older, female, have transfer admissions, and less likely to be hospitalized to institutions with cardiac interventional facilities, intensive care units or coronary care units. The uninsured were also less likely to undergo diagnostic procedures, interventions and to receive medications, and stayed shorter in hospital and consumed less health care resources. In-hospital mortality in the uninsured, the non-government insured and the government insured was 10.5%, 12.2% and 8.4% respectively. After adjusting for potential confounders, odd ratio in hospital mortality was 1.079 (95% CI, 0.836-1.392) and 0.763 (95% CI, 0.559-1.041) for the non-government insured and the government insured, compared to the uninsured. At significant level of 0.05, we could not assert insurance status is a significant factor to in-hospital mortality.
BACKGROUND: It is unknown whether insurance status influences care provided and patients' prognosis, in China. METHODS: This retrospective cohort study included medical records of 4,714 patients with acute myocardial infarction aged 20 and older, discharged consecutively from 14 Chinese hospitals between January 2000 and February 2003. Uni-variate analysis, multivariate logistic regression and linear regression were used to compare differences in patients' characteristics, care provided and prognosis between insured patients and the uninsured. RESULTS: The uninsured were more likely to be older, female, have transfer admissions, and less likely to be hospitalized to institutions with cardiac interventional facilities, intensive care units or coronary care units. The uninsured were also less likely to undergo diagnostic procedures, interventions and to receive medications, and stayed shorter in hospital and consumed less health care resources. In-hospital mortality in the uninsured, the non-government insured and the government insured was 10.5%, 12.2% and 8.4% respectively. After adjusting for potential confounders, odd ratio in hospital mortality was 1.079 (95% CI, 0.836-1.392) and 0.763 (95% CI, 0.559-1.041) for the non-government insured and the government insured, compared to the uninsured. At significant level of 0.05, we could not assert insurance status is a significant factor to in-hospital mortality.