Commentary: First seek to understand (the database)

Abstract

An essential part of designing an observational research study or understanding database studies is understanding the strengths and limitations of the data source. Kidane and colleagues 1 provide a detailed review of select databases that are used for large observational research studies in thoracic surgery. The authors clearly convey the specifications of each database they review for the benefit of outcomes researchers.Administrative and clinical databases can be distinguished based on the purpose for which the data are collected. Administrative data are collected primarily for claim, billing, and discharge purposes. They are available at different levels, including national, state, hospital, or by specific payer. The data are organized in a uniform format that contains patient demographic characteristics, discharge status, diagnosis and procedure codes, and hospital identifiers. Examples include Centers for Medicare and Medicaid Services and Healthcare Cost and Utilization Project data. A clinic database, or patient registry, is "an organized system that uses observational study methods to collect uniform data (clinical and other) to evaluate specified outcomes for a population defined by a particular disease, condition, or exposure, and that serves a predetermined scientific, clinical, or policy purpose(s)." 2 The majority of the databases described in this review (American College of Surgeons National Surgical Quality Improvement Project, Society of Thoracic Surgeons General Thoracic Surgery Database, and European Society of Thoracic Surgeons Database), all fall under the category of clinical registries. On the other hand, the databases from Denmark and Canada are administrative data sources.