The impact of insurance status on in-hospital mortality in patients with hyperglycaemic crisis: A propensity score matching analysis

Abstract

Aim: This study was designed to determine the associations between insurance status and clinical outcomes among patients with hyperglycaemic crisis.

Methods: Overall, 1668 patients with hyperglycaemic crisis were recruited from the Chongqing Medical University Medical Data Science Academy's big data platform. In-hospital mortality, length of stay and complications (i.e., hypoglycaemia, hypokalemia, pulmonary infection, multiple systemic organ failure, acute kidney injury and deep venous thrombosis) were assessed. Propensity score matching analysis was used to reduce the confounding bias, and univariate and multivariate logistic regression were used to estimate the effect of insurance status on mortality in patients with hyperglycaemic crisis.

Results: After matching one uninsured patient to two insured patients with a calliper of 0.02, the uninsured group suffered a higher burden of in-hospital mortality than the insured group (16.9% vs. 9.8%); the insured status (odds ratio = 0.216, 95% confidence interval = 0.079-0.587) was a potential protect factor for in-hospital mortality of patients with hyperglycaemic crisis in the multivariate logistic regression analysis.

Conclusions: Insurance status is associated with the outcomes of hospitalisation for hyperglycaemic crisis; uninsured patients with hyperglycaemic crisis face a higher risk of mortality in China.

Keywords: diabetes; hyperglycaemic crisis; insurance; mortality; propensity score matching.