Article Text

SAT0453 Why do we make diagnostic errors? A delphy exercise to identify potential causes of diagnostic errors
  1. C. Rosca1,
  2. F. Berghea2,
  3. C. Constantinescu2,
  4. M. Abobului2,
  5. V. Vlad1,
  6. I. Saulescu2,
  7. L. Groseanu2,
  8. M. Negru2,
  9. D. Opris2,
  10. V. Bojinca2,
  11. M. Bojinca2,
  12. A. Balanescu2,
  13. D. Predeteanu2,
  14. R. Ionescu2
  15. and Romanian Society of Rheumatology Working Group
  1. 1Rheumatology, Sf. Maria Hospital
  2. 2Carol Davila University of Medicine and Pharmacy, Bucharest, Romania


Background Diagnostic errors in medical practice are neither rare nor minor. Diagnoses that are delayed, wrong or missed entirely result in 40,000 to 80,000 U.S. hospital deaths annually [1]. In USA, about 40% of medical liability lawsuits involve diagnostic errors. Rheumatic diseases need an early diagnostic in order to avoid permanent damages. Unfortunately this is not always possible and many cases receive wrong diagnostics.

Objectives Our objective was to create an inventory of the most important causes of diagnostic errors in rheumatology

Methods Six highly experienced rheumatologists (with more than 15 years of practice) have been involved in a focus group exercise with the aim of developing a draft inventory of the most commune causes of diagnostic errors. 27 such causes equally divided in three groups have been identified. 98 Rheumatologists have been invited to participate in a two stage internet based Delphi exercise to rank these items according their susceptibility to generate diagnostic errors. In the first stage all participants have been instructed to evaluate the potential effect of each item on a 10 point Likert-type scale (with 0 for “this item is definitely not involved in diagnostic errors” and 10 for “this is a major cause of diagnostic errors”). In the 2nd stage we involved just the responders from the 1 stage; the participants were also asked whether mean value obtained for each item is appropriate or should be smaller or higher.

Results The draft inventory of possible causes of diagnostic errors included 27 causes divided in: causes related to doctors, related to patients and to medical system. 49 responders (50%) of the invited rheumatologists participated in the first stage and 42 in the final. Their clinical experience was between 1 and 45 years with a mean of 22 years and a SD of 8.6. 81.6% of responders were from an academic city, 71.4% were employed by public hospitals and 53% were active in private practice (full or part-time). 15 items passed the two stages: four related to doctors, four related to patients and seven related to health system. The highest ranked was computed for: Relevant investigations are missing (7.6), Incorrect clinical examination (7.5), Incorrect evaluation of differential diagnostics (7.4), Clinicians are overloaded with non-clinical activities (teaching, research, management) (7.3), Clinicians are overloaded with clinical activities (7.2). As a general rule, patient related items received lower ranks. Significant difference between inpatient and outpatient doctors exists.

Conclusions The responsibility for diagnostic errors should be shared between doctors, patients and medical system. Not a single cause can’t be removed or fixed – but first is important to be identified. Our study offers an instrument to audit the potential of developing of such errors in one setting, to identify active causes and to propose potential solutions.

  1. Kevin B. O’Reilly Diagnostic errors: Why they happen? American Medical News. Posted Dec. 6, 2010.

  2. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003;78(8):775–80

Disclosure of Interest C. Rosca: None Declared, F. Berghea Grant/Research support from: Pfizer, Consultant for: Amgen, Abbott, RG, C. Constantinescu: None Declared, M. Abobului: None Declared, V. Vlad: None Declared, I. Saulescu: None Declared, L. Groseanu: None Declared, M. Negru: None Declared, D. Opris: None Declared, V. Bojinca: None Declared, M. Bojinca: None Declared, A. Balanescu: None Declared, D. Predeteanu: None Declared, R. Ionescu: None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.