Purpose/Objectives: To identify previously unstudied factors predicting re-excision following breast-conserving surgery (BCS) and to assess the feasibility of obtaining data about breast density for predictive modeling.
Design: Retrospective secondary data analysis.
Setting: Data were obtained from the cancer registry and electronic health records (EHRs) at Texas Health Harris Methodist Hospital, a large, urban, private, nonprofit hospital in North Texas.
Sample: 244 patients choosing BCS from 2011–2012.
Methods: Data were subjected to univariate analyses (chi-square) followed by logistic regression.
Main Research Variables: The primary dependent variable was re-excision following BCS. Predictors of interest included lifestyle factors, time from diagnosis to surgery, surgical approach, patient age, and breast density, and controlled for covariates, such as assay results.
Findings: Three factors predicted re-excision with 87% accuracy: time from diagnosis to surgery, needle localization, and age. Missing data precluded using breast density as a predictor.
Conclusions: Women younger than 60 years whose surgery included placement of a wire for localization of tissue to be removed and who underwent surgery soon after diagnosis are the least likely to require reoperation after BCS. Data integrity is critical to the success of research using EHRs and registry information.
Implications for Nursing: Nurses may improve patient outcomes by helping women considering BCS solve problems that may delay surgery. Nurses can contribute to the success of nursing research by thoroughly and accurately recording patient information in EHRs.