The Determinants of Health Care Costs in Older Adults Undergoing Non-Elective Abdominal Surgery

Abstract

Health care spending in Canada has been increasing faster than the rate of gross domestic product (GDP). A disproportionate amount of the health care spending is allocated to care of older adults. Non-elective abdominal surgery is an expensive area of care for older adults. Despite this, the factors associated with cost in this patient population remain unclear. OBJECTIVES The primary objective of this study was to estimate the association between perioperative factors (age, American Society of Anesthetists (ASA) classification, operative severity (OS), frailty index (FI), complication severity) and health care costs among older adults undergoing non-elective abdominal surgery. The secondary objectives were: 1. to provide a comprehensive description of costs based on patient-level resource utilization; and 2. to examine the relationship between hospital costs and adverse events (non-fatal complication severity, mortality, and change in living arrangement). METHODS This study was an observational prospective cohort study. Over a 15 month period all patients 70 years or older who underwent non-elective abdominal surgery at the QEII Health Sciences Centre, Nova Scotia, were enrolled. Data were collected on patient demographics, investigations, treatments, and outcomes. Direct hospital health care costs (2012 $CAD) were calculated by tabulating patient-level resource use and assigning specific costs. The association between five perioperative factors and costs were analyzed using univariate non-parametric tests and multiple linear regression. The associations between adverse events and costs were assessed using univariate non-parametric tests and multiple linear regression. RESULTS During the study period, 212 patients who underwent abdominal surgery (median age 78 years (range 70-97)) were enrolled. The median costs of care were $9,166 (range $1,993-$104,403). The largest proportions of spending were non-procedural costs (65% [$2,176,875]) and intensive care costs (16% [$554,523]). The perioperative factors ASA classification (p=0.0010), OS (p<0.0001), FI (p=0.0002) and complication severity (p<0.0001) were all independently associated with health care costs, while age was not (p=0.5330). The following adverse events were independently associated with health care costs: non-fatal complication severity (p<0.0001), change in living arrangement (p=0.0002), and mortality (p=0.0337). Non-fatal complications had the strongest association with hospital costs (standardized β coefficient = 0.3931). CONCLUSION Four perioperative factors (ASA, OS, FI and complication severity) are associated with costs; therefore, representing a potential cost prediction model for this patient group. This study is important for health care administrators, identifying targets for cost reduction. Cost reduction strategies and research should concentrate on mitigating or preventing complications and high cost areas, such as non-procedural costs and intensive care, in order to achieve cost savings.