Assessing Racial Bias Within American Society of Anesthesiologists Classification.

INTRODUCTION

The American Society of Anesthesiologists (ASA) classification system is widely used to evaluate surgical risk, make patient care decisions, and determine reimbursement. However, race/ethnicity bias in ASA class is largely unexplored, with potential ramifications for healthcare disparities.

METHODS

In a retrospective cohort study of 350,187 Veterans Health Administration surgical cases and 4,051,185 non-Veterans Health Administration cases, race/ethnicity groups were matched on the components of the Risk Analysis Index, a surgical frailty measure, and preoperative acute serious conditions, conditions associated with urgent/emergent surgeries, to evaluate differences in ASA class.

RESULTS

After matching, Black cases had higher ASA class versus White cases (adjusted odds ratios = 1.37 and 1.07, 95% confidence intervals = 1.37-1.38 and 1.04-1.09, P < 0.001). Hispanic cases had lower ASA class versus White cases (adjusted odds ratios = 0.96 and 0.96, 95% confidence intervals = 0.96-0.97 and 0.93-0.99, P < 0.001 and 0.008), though the effect changed sign in a sensitivity analysis.

CONCLUSIONS

ASA class varies by race/ethnicity, representing bias that could drive health disparities. Given the matching strategy used, such bias may have practical implications for patient outcomes and quality improvement. For Black patients, a higher ASA class may drive increased preoperative testing, surgical delays, and increased costs. For Hispanic patients, a lower ASA class may drive reduced preoperative screening and riskier outpatient procedures. For some purposes, the ASA class may need to be replaced with other measures of patient morbidity. For others, ASA class may need to be revised and assigned by automated systems with safeguards against race/ethnicity bias.