Methods and Dilemmas of Using Patient Registries in Evidence-Based Research

• Activity Number: 240
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Carolyn Carroll*+ and Jan F. Nygard
• Companies: Stat Tech Inc and Norwegian Cancer Registry
• Keywords: categorical data ; nonparametric methods ; analysis ; evidence-based ; ethics ; sampling

Abstract

Some of the earliest patient registries began in the 1950's and contain information about patients' diagnosed with cancer. Patient records for those with other disease areas have also formed the basis of registries..In a sample of evidence-based research abstracts, the number articles indicates indirectly their uses: counts by key word were international registry (5560), registry and accident (1902), registry and outcomes (9378), registry and design (8504), registry and design and objectives or rationale (52566), trauma registry and military (213), accidents and military trauma (711). There are benefits to scrutinizing the collection of registries and their uses in research about treatments and diseases. Not only are researchers interested in the scope, design, maintenance and data quality of registries but also about the methodological and statistical issues and dilemmas they present. The purpose of this paper is to provide a critical review of patient registries, describe best practices, identify areas for improvement in the design and development of evidence-based research registries and present examples of methodologically sound statistical analyses and outcomes.