Checklist Improves Diagnosis over Physician Recall

Posted on by cradmin
0

Introduction

Diagnostic error in medicine is conservatively estimated at 15% (Newman-Toker) while the actual number may be considerably higher(Winters, Newman). Checklists have been demonstrated to reduce error in procedural  areas of medicine yet, utilization remains limited (Ely, Gaw 2009 ). To measure usefulness in diagnosis in neurologic disease, a computer generated differential diagnostic checklist utilizing NeurologicDx.com, an open access internet based program (www.NeurologicDx.com) is compared to physician recall in third year neurology residents and practicing neurologists. This study demonstrates checklist programs  can impact neurologic diagnosis when compared to physician recall alone.

Background

Checklists have been demonstrated effective in procedural aspects of medicine, yet implementation has not been fully embraced and diagnostic capability not systematically studied.

Objective

To assess the potential of a diagnostic checklist by comparing physician recall to a computer generated differential diagnostic program, NeurologicDx.

Methods

A group of 12 PGY4 resident and practicing neurologists were each given a sheet of paper which contained the single instruction “List as many neurologic diseases associated with each term below”. There were three terms, one at the top of each of three columns with 10 blank lines drawn below.  Participants were asked to indicate their time in practice as either PGY4 for third year residents or decades in practice for attending physicians. The terms, “jaundice”, “night sweats”, and “oval pupil” were used without a time limit for completion. Each of the answers provided by participants would be validated by Google search.

Results for combined method, NeurologicDx, group total unique terms and individual mean

Term Total List Database Generated (%) Group (%) Individual Mean (%)
Jaundice 35 23 (65.7%) 17 (48.5%) 3.5 (10.0%)
Night Sweats 41 33 (80.4%) 15 (36.5%) 4.4 (10.7%)
Oval Pupil 13 11 (84.6%) 5 (38.5%) 1.1 (11%)


Summary

Computer generated checklists can be used to expand the information available at point of care to remind physicians of significant items to add to their differential diagnosis. The overall suggestion is that computer generated checklists will significantly improve diagnostic capability, shorten time for appropriate care and reduce error.

References

  1. Winters BD, Aswani MS, Pronovost PJ. Diagnostic Errors: Another Role for Checklist? Acad Med 2011; 86: 279-281.
  2. Newman-Toker DE, Pronavost PJ. Diagnostic Errors- The next frontier for patient safety. JAMA 2009; 301: 1060-1062.
  3. Ely JW, Graber ML, Crosberry P. Checklists to Reduce Diagnostic Errors.  Acad Med 2011; 86: 307-313.
  4. Gawande A. The Checklist Manifesto-How to Get Things Right. New York, NY: Metropolitan Books; 2009.
  5. www.NeurologicDx.com
  6. Pronovost P, Needham DE, Berenholtz S, et al. An Intervention to Decrease Catherter-related bloodstream infections in the ICU. New Englal J Med. 2006; 355: 2725-2732.
  7. Gawandi A. The Checklist. The New Yorker. December 10, 2007; 86-95.
  8. Haynes AB, Weiser TG,Berry WR, et al. A surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population. New Engl J Med 2009; 360; 491-499.
  9. Finelli PF, McCabe AL. NeurologicDx: Internet Based Differential Diagnostic Checklist in Clinical Neurology. Ann Neurol (abstract) 2013; 74: S5.