The Learning Curve for Coronary Surgery: An Stochastic Frontier Model Approach
Posted: 21 Jun 2007
Abstract
Objective: To study the learning curve associated with independent practice in coronary artery surgery.
Participants: 123 patients undergoing coronary artery surgery for the first time between January 2001 and April 2002, who were operated on by 6 surgeons.
Methods: Stochastic frontier cost funtion and an inefficiency model. The analysis of the data was carried out as an individual approach in order to estimate the influence of individual effects. Considering each patient as if it was the result of a small production line enables us to evaluate variations between different inputs. We assume that each patient enters in the hospital and faces a cost function that not only does it depend on resources used, but also on patient-specific characteristic and circumstances that might arise during surgery. The principal added value of this paper consists in the stochastic frontier model approach. The stochastic frontier approach enables us to measure the efficiency derived from the learning process.
Results: The results obtained also suggest the following factors have an important impact on technical efficiency: the presence of surgery complications, be a diabetic patients and occurrences of retrospective myocardial infarction.
Conclusions: Costs in patients operated on by newly appointed consultant sugeons is similar to costs in patients operated on by established surgeons. Assuming the availability of appropriate sources of data, stochastic frontier approach might prove fruitful in other areas of health care evaluative research.
Keywords: Learning curve, stochastic bayesian frontiers, coronary surgery
Suggested Citation: Suggested Citation