Abstract
BACKGROUND: National databases are a rich source of epidemiologic data for breast surgical oncology research. However, these databases differ in the demographic, surgical, and oncologic variables provided. This study aimed to compare the strengths and limitations of four national databases in the context of breast surgical oncology research. METHODS: The study comprised a descriptive analysis of four national databases (the National Surgical Quality Improvement Program [NSQIP], the Nationwide Inpatient Sample [NIS], the Surveillance, Epidemiology and End Results [SEER] program, and the National Cancer Database [NCDB]) to assess their strengths and limitations in the context of breast surgical oncology. The study assessed the data available in each database for female patients with a breast cancer diagnosis between 2007 and 2017, and compared patient age, ethnicity, and race distributions. RESULTS: Data from 3.9 million female patients were examined, with most patients being between 60 and 69 years of age, non-Hispanic, and white. Age, ethnicity, and race distributions were similar in the databases. The NSQIP includes data on operative details, comorbidities, and postoperative outcomes. The NIS provides health services and inpatient utilization information, but does not evaluate outpatient procedures. The SEER program provides population-based oncologic detail including stage, histology, and neoadjuvant/adjuvant treatment. The NCDB offers hospital-based oncologic information and the largest population in the study period, with approximately 2.5 million breast cancer patients. CONCLUSION: Epidemiologic datasets offer tremendous potential for the examination of oncologic breast surgery, with each database providing unique data useful for addressing different epidemiologic questions. Understanding the strengths and limitations of each database creates a more efficient and productive research environment.