Predicting different factors that affect hospital utilization and outcomes Among diabetic patients admitted with hypoglycemia using structural equation modeling.

Background: Hypoglycemia (HG) is a common complication among diabetic patients. Many diabetics who experience HG are admitted to hospitals and usually utilize more resources. While plenty of studies examined multiple HG risk factors, there is limited knowledge about the correlation between different risk factors of HG and their impact on utilization.

Objective: To identify key factors influencing utilization among diabetic HG patients and to examine the mechanisms and interactions between those factors.

Design: A quantitative, non-experimental, and retrospective design that is based on the selection of the study subjects from the Healthcare Cost and Utilization Project National database for the years of 2012-2014. We employed Andersen Behavioral Model of Health Services Use as the main framework for this study.

Results: Structural Equation Modeling was used as the main multivariate statistical method for the analysis. Total sample size was 4822 patients. We found that diabetes complications, renal disease, hypertension, and high Charlson comorbidity index score had the strongest impact on length of stay (LoS) as well as total charge. Geographical location of patients strongly influenced total charge. Age had an indirect impact on LoS and total charge.

Limitations: The use of secondary data seems to be the primary limitation for this study as some relevant risk factors for hypoglycemia were not available in the database.

Conclusions: This study examined the multilevel character of different factors leading to high utilization of healthcare services among HG patients admitted to hospitals. Findings of this study help clinicians and policy makers to formulate policies and protocols that aid in providing efficient care to HG patients with less utilization of resources.