Five steps to develop checklists for evaluating clinical performance: an integrative approach

Acad Med. 2014 Jul;89(7):996-1005. doi: 10.1097/ACM.0000000000000289.

Abstract

Purpose: The process of developing checklists to rate clinical performance is essential for ensuring their quality; thus, the authors applied an integrative approach for designing checklists that evaluate clinical performance.

Method: The approach consisted of five predefined steps (taken 2012-2013). Step 1: On the basis of the relevant literature and their clinical experience, the authors drafted a preliminary checklist. Step 2: The authors sent the draft checklist to five experts who reviewed it using an adapted Delphi technique. Step 3: The authors devised three scoring categories for items after pilot testing. Step 4: To ensure the changes made after pilot testing were valid, the checklist was submitted to an additional Delphi review round. Step 5: To weight items needed for accurate performance assessment, 10 pediatricians rated all checklist items in terms of their importance on a scale from 1 (not important) to 5 (essential).

Results: The authors have illustrated their approach using the example of a checklist for a simulation scenario of infant septic shock. The five-step approach resulted in a valid, reliable tool and proved to be an effective method to design evaluation checklists. It resulted in 33 items, most consisting of three scoring categories.

Conclusions: This approach integrates published evidence and the knowledge of domain experts. A robust development process is a necessary prerequisite of valid performance checklists. Establishing a widely recognized standard for developing evaluation checklists will likely support the design of appropriate measurement tools and move the field of performance assessment in health care forward.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Checklist / methods*
  • Clinical Competence*
  • Delphi Technique
  • Humans
  • Quality Assurance, Health Care / methods*
  • Reproducibility of Results