Hospital finances and inpatient sepsis mortality

INTRODUCTION: A handful of studies have reported relationships between measures of hospitals' financial health and patient outcomes. Others have observed that financial measures may have differential effects depending on ownership type (profit or non-profit). However, there is little consistency across studies as to the best financial measures. California is instructive given its size, diversity, and mandatory submission of detailed financial data by most hospitals to the Office of Statewide Health Planning and Development (OSHPD). Much of that financial data is published on the State website and includes 12 financial ratios comparing debt, equity, income and profit margins. This study examines the relationship of those ratios to mortality of patients diagnosed with sepsis, emphasizing hospital ownership type. METHODS: Annual 2003-2005 financial reports and public-use 2003-2005 general acute care hospital discharges were obtained from OSHPD. Discharge and financial data were merged based on unique hospital identifier and year. ICD-9 CM diagnosis and procedure codes were used not only to extract discharges having sepsis, severe sepsis, and/or septic shock; but also to identify organ failure and relevant comorbid conditions for the sepsis discharges. Due to high correlation, only nine ratios were considered, with analysis restricted to hospitals having at least 100 sepsis discharges and complete financial data. Covariates included other hospital characteristics (such as bed size) and patient-level variables (demographics, insurance, and numbers of organ failures and comorbities) in logistic regression models, clustered by hospital identifier, using STATA S/E 9.2. RESULTS: This study examined 294 hospitals: 86 for-profits, 169 non-profits, and 43 public/other hospitals. In the analysis of 248,497 discharges with any sepsis (of which 22.9% ended with inpatient mortality), the ratio of net return on total assets (net income / total assets) was significantly associated with mortality in the overall and the for-profit hospital models. Operating margin (net from operations / net patient revenue + other operating revenue) was significant only in the public hospital model. In the analysis of 133,514 severe sepsis or septic shock discharges (36.6% mortality), no financial variables were significant in the full model. However, the current ratio (current assets / current liabilities) and operating margin were significant in the public hospital model. CONCLUSION: The modest association of income, equity, assets and liability ratios with inpatient sepsis mortality, after adjusting for many other factors, underscores the importance of hospital financial viability, regardless of ownership type. Further research is needed to determine how finances result in differential mortality risks.