2008 Annual Meeting Abstracts

---

Identification of Quality Measures for Breast Cancer Surgery and Assessment of Surgeon Variation at the University of Vermont
Alicia R. Privette, MD, Laurence McCahill, MD, Ted James, MD, Seth Harlow, MD, Mary Stanley, MD, David Krag, MD, Johanna Sheehey-Jones, RN.
University of Vermont/Fletcher Allen Health Care, Burlington, VT, USA.

1. Objective: To identify and quantify surgical quality measures for breast cancer surgery and determine the feasibility of assessing variation among surgeons.
2. Design: Descriptive analysis of prospectively collected patient outcome measures
3. Setting: Rural university-based hospital with NCI designated breast cancer center
4. Patients: All patients with pre-operative diagnosis of ductal carcinoma in situ or invasive breast cancer treated by a breast center surgeon (March 2003-2008) (n=885). Exclusion: excisional biopsy.
5. Intervention(s): N/A
6. Main Outcome Measure(s): Six quality outcome measures identified: ¹total mastectomy (TM) rate, ²partial mastectomy (PM) close (<1mm) and positive margin rate, ³mean number of operations per initial PM patient to achieve definitive care, ⁴number of lymph nodes from axillary dissection (AD), ⁵proportion AD following positive sentinel lymph node biopsy (SLNB), and ⁶use of intra-operative lymph node assessment. Data analysis: Chi-square and non-parametric tests.
7. Results: 885 operations (DCIS =214, invasive = 671) performed by five surgeons (March 2003-2008). Significant variation (p<0.05): combined close and positive margin rate, close margin rate alone, and use of intra-operative lymph node assessment. Wide range and trend toward significance: AD following SLNB (0.065). No statistically significant variation: TM rate, mean number of operations, positive margin rate, number of lymph nodes from AD.

Quality Measure	Institutional Average	Surgeon Range	p-value
Total mastectomy rate (%) (n=885)	25.6%	23.6-37.3%	0.086
PM combined close/positive margin rate (%) (n=658) - Positive alone - Close alone	29.8% 9.4% 20.4%	17.9-37.1% 5.8-11.2% 10.7-25.8%	<0.001 0.3 0.006
Mean number operations per initial PM patient (n=658)	1.19	1.11-1.23	0.5
Number lymph nodes obtained from AD (n=205)	17.1	14-18.5	0.088
AD following positive SLNB (%) (n=103)	57.3%	43.2-100%	0.065
Intra-operative lymph node assessment (%) (n=554)	60.5%	37.3-83.9%	<0.001

8. Conclusions: Quality measures for breast cancer surgery are identifiable and quantifiable. Significant variation exists in surgical practice patterns even among surgeons at a high volume breast center. Further study into the causes of surgical practice variability and correlation to long term disease outcomes is needed.

---