Commentary: core descriptor sets using consensus methods support 'table one' consistency

J Clin Epidemiol. 2024 Oct:174:111470. doi: 10.1016/j.jclinepi.2024.111470. Epub 2024 Jul 20.

Abstract

Background: Inconsistent reporting of patient characteristics in clinical research hampers reproducibility and limits analysis opportunities. This paper proposes condition-specific 'Core Descriptor Sets' comprising key factors like demographics, disease severity, comorbidities, and prognosis to standardize Table 1 reporting.

Methods: Development entails stakeholder involvement, systematic identification of descriptors, value rating, and consensus-building using multiple Delphi rounds. Final agreement comes at an expert meeting.

Conclusion: Benefits include easier cross-study comparison, for example, through individual patient meta-analysis, facilitated by comparison of consistently reported individual data rather than group-level analysis. This may also support routine data analyses, subgroup and risk identification, and reduced research waste. Core Descriptor Sets describe cohorts thoroughly while minimizing research burden. They are intended to enable improved clinical characterization, personalization, reproducibility, data sharing, and knowledge building.

Keywords: Consensus; Data analysis; Documentation; Epidemiologic research design; Guidelines as topic; Prognosis; Publishing; Reproducibility of results; Research design; Risk factors.

MeSH terms

  • Biomedical Research / methods
  • Biomedical Research / standards
  • Consensus*
  • Delphi Technique
  • Humans
  • Reproducibility of Results
  • Research Design / standards