(S-158) Glance, L.G., Sunday 9:15

TITLE: THE COST-EFFECTIVENESS OF ANESTHESIA WORKFORCE MODELS: A SIMULATION APPROACH USING DECISION- ANALYSIS MODELING

AUTHORS: Laurent G. Glance, MD
AFFILIATION: University of Rochester Medical Center, Rochester, NY.

INTRODUCTION: Anesthesia costs are estimated to account for 3-5% of total health care expenditures. A recent proposal by HCVA to eliminate the requirement for medical supervision of CRNAs may lead to policies encouraging the use of lower-paid CRNAs as substitutes for anesthesiologists in an effort to lower costs. The objective of this study was to evaluate the incremental cost-effectiveness (incr. CE), using the technique of decision analysis, of anesthesia work force staffing scenarios as a function of skill mix.
METHODS: A decision tree model was constructed to compare the incremental cost-effectiveness of alternative delivery systems for anesthesia care from the perspective of the health care system. Five different staffing scenarios were modeled - ranging from a 'physician-intensive' to 'nurse-intensive' - based on the Abt study. In the 'nurse-intensive' model, low- and intermediate-risk patients were cared for by solo CRNAs; high-risk patients by physicians. In the 'physician-intensive' model, all patients received anesthesia from a physician. For the 1^st-, 2^nd, and 3^rd- team models, all high-risk patients were cared by physicians working alone and all intermediate-risk patients were cared using an anesthesia care team approach (ACT) with a ratio of 1 physician to 2 CRNAs. The low-risk patients were managed using an ACT approach with ratios of 1:2, 1:4 and 1:8 in the 1^st, 2^nd, and 3^rd-team models, respectively. The perioperative death rate was based on the CEPOD database. Differences in outcomes for anesthesiologists and CRNAs were based on the literature; their effect on the model output was explored using sensitivity analysis. For survivors, the numbers of years of life saved (YLS) was calculated using Markov trees. Cost and patient demographics were based on a data set of 24,970 patients from a major teaching hospital.
RESULTS: In the baseline analysis, the incr CE of the 3^rd-team model versus the 'nurse-intensive' model was $4900 per YLS; the incr CE of the 2^nd-team model versus the 3^rd-team model was $31,000/YLS. The 'physician-intensive' and the 1^st-team model were not cost-effective. Sensitivity analysis demonstrated that the 3^rd-team model was always cost-effective.
CONCLUSION: An ACT approach with a physician to nurse ratio of 1:2 appears to be cost-effective for intermediate-risk patients versus CRNAs working alone without medical super- vision. Medical direction of CRNAs caring for low-risk patients is also cost-effective. The small improvement in outcome resulting from increasing the level of supervision of low-risk patients from a physician to CRNA ratio of 1:8 to 1:4 may not be justified by the added cost.