A detailed analysis of the reduction mammaplasty learning curve: a statistical process model for approaching surgical performance improvement

Abstract

Background: The increased focus on quality and efficiency improvement within academic surgery has met with variable success among plastic surgeons. Traditional surgical performance metrics, such as morbidity and mortality, are insufficient to improve the majority of today's plastic surgical procedures. In-process analyses that allow rapid feedback to the surgeon based on surrogate markers may provide a powerful method for quality improvement.

Methods: The authors reviewed performance data from all bilateral reduction mammaplasties performed at their institution by eight surgeons between 1995 and 2007. Multiple linear regression analyses were conducted to determine the relative impact of key factors on operative time. Explanatory learning curve models were generated, and complication data were analyzed to elucidate clinical outcomes and trends.

Results: A total of 1068 procedures were analyzed. The mean operative time for bilateral reduction mammaplasty was 134 +/- 34 minutes, with a mean operative experience of 11 +/- 4.7 years and total resection volume of 1680 +/- 930 g. Multiple linear regression analyses showed that operative time (R = 0.57) was most closely related to surgeon experience and resection volume. The complication rate diminished in a logarithmic fashion with increasing surgeon experience and in a linear fashion with declining operative time.

Conclusions: The results of this study suggest a three-phase learning curve in which complication rates, variance in operative time, and operative time all decrease with surgeon experience. In-process statistical analyses may represent the beginning of a new paradigm in academic surgical quality and efficiency improvement in low-risk surgical procedures.