Editorial |
Do GPs report diagnostic errors?
Primary Care Clinical Sciences, University of Birmingham, Birmingham B15 2TT, UK; Email: o.kostopoulou{at}bham.ac.uk
Diagnosis is the most important function of a general practitioner (GP). Management decisions, including referrals, are based on diagnosis.1 Missing serious conditions such as cancers can have devastating consequences for patients. It can also have serious adverse consequences for the doctor in terms of trust, self-confidence and litigation. Indeed, most medico-legal claims against GPs are about delay in diagnosis or misdiagnosis (63–66%).2,3 GPs do not perform high-risk, invasive procedures and do not prescribe medications with potentially very serious adverse effects. Treatment errors are therefore unlikely to harm patients seriously and hence unlikely to be the most memorable errors in a GP's career. It is not surprising that GPs remember their diagnostic errors most. What is perhaps surprising is that they report them. In a study published in this issue of the journal, Fisseni and colleagues asked GPs to report anonymously the ‘three most serious errors’ of their careers.4 Seventy-two per cent of the errors reported were about diagnosis. This study succeeded where incident reporting systems, whether confidential or anonymous, have failed: to get clinicians to report diagnostic errors.5–9 There are good reasons for this discrepancy. The scope of incident reporting systems is much larger than diagnostic error and practitioners have ample opportunity to report other types of errors, for example, administrative or errors with a clear ‘system’ cause. These are usually a lot less threatening than diagnostic errors. Incident reporting systems do not ask practitioners to report their most serious errors. In fact, most ask for the reporting of incidents rather than errors. Severity of consequence and being personally involved in the error are not usually conditions for reporting incidents. Finally, incident reporting systems target current rather than past incidents and doctors may be unwilling to draw attention to a recent diagnostic error. GPs in the Fisseni et al. study were reporting past errors, some from several years ago. They were not likely to suffer consequences from reporting these errors. Some of them or their patients had already suffered the consequences.
Inevitably, there are problems when asking people to report past events. First, there is hindsight bias, which occurs when people examine past decisions. Knowledge of the outcome focuses attention on information that is consistent with that outcome. Evidence consistent with the diagnosis is more easily recalled whilst contradictory or ambiguous information may be forgotten. This can make the diagnosis seem more predictable and may even induce false memories of consistent data—not available prior to the diagnosis.10,11
Second, there is the problem of limited memory for the cognitive processes that took place at the time of the error, for example, what diagnostic hypotheses were being considered and how clinical information was interpreted. This information is very important in helping us piece together the ‘diagnostic error path’ but it is unlikely to be obtained by asking people to report diagnostic errors. There is the inevitable memory decay. Diagnostic errors are not immediately noticed and feedback about the outcome of the diagnostic process is often delayed if not missing. More importantly, people have limited access to their cognitive processes and hence are unlikely to report them. Realistic patient scenarios lend themselves to the detailed study of doctors’ thinking processes during diagnosis and management.12–14
The cases illustrating the types of errors GPs reported to Fisseni and colleagues show how important the thinking processes are and how vulnerable they are to the features of the diagnostic problem (case 2) or situational factors (cases 3 and 4). Case 2 is extremely difficult. The patient (female, middle-aged, frequent attender) was labelled as ‘moaning’. Some of her symptoms were explained by a pathophysiological process confirmed by X-ray. The rest were sufficiently non-specific to be attributed to her ‘moaning’ nature. When, some symptoms got increasingly worse, the GP's concerns were put to rest by a specialist. This is where the GP could be faulted—for not acting appropriately on these concerns by referring the patient through the usual channels. Asking for informal advice probably meant that the patient was never properly examined by the specialist. A similar observation can be made about case 3, where the GP reported having ‘bad feelings’ but did not act accordingly. ‘Bad feelings’ reported to the authors by many other GPs in the study could indicate a number of things: a sufficiently specified differential diagnosis that is probably considered quite unlikely, the suspected possibility of a serious but unspecified disease, or simple discomfort resulting from the recognition of significant uncertainty. Whichever the case, many errors would have been avoided had the GPs acted on their ‘bad feelings’. In a recently completed study, we found that simply considering a diagnostic possibility at the end of the consultation will result in appropriate management decisions, even if the GP does not think that it is the most likely diagnosis.1 Identifying barriers to GPs acting on their suspicions is therefore worthwhile.15
Asking people to list their most serious past errors will reveal what is central to their professional role and what had the worst consequences. Diagnosis is central to the GPs’ role and diagnostic error has the potential for the most serious consequences for patients. It is therefore a priority for the research and policy agenda.
References
1 Kostopoulou O, Harries C, Delaney BC, Oudhoff J, Nath R, Munro CW. Predictors of diagnostic accuracy and patient management in difficult diagnostic problems in family medicine. Presented at the 29th annual meeting of the Society for Medical Decision Making (24/10/2007). Abstract published on web: http://smdm.confex.com/smdm/2007pa/techprogram/P3590.HTM.
2 The Medical Defence Union. Training and education: Primary care development programme – Risk management and delay in diagnosis. The Medical Defence Union 2004 [cited 2003 Dec. 2]; Available from URL:http://www.the-mdu.com/gp/services/training_and_education/primary_care_development_programme/overview_of_the_programme/index.asp.
3 Silk N. What went wrong in 1000 negligence claims. Health Care Risk Report (2000).
4 Fisseni G, Pentzek M, Abholz H-H. Responding to serious medical error in general practice – consequences for the GPs involved: analysis of 75 cases from Germany. Fam Pract (2008) 25:9–13.
5 Dovey S, Meyers D, Phillips RJ, Green L, Fryer G, Galliher J, et al. A preliminary taxonomy of medical errors in family practice. Qual Saf Health Care (2002) 11:233–238.
6 Makeham M, Dovey S, County M, Kidd M. An international taxonomy for errors in general practice: a pilot study. MJA (2002) 177:68–72.[Medline]
7 Rubin G, George A, Chinn D, Richardson C. Errors in general practice: development of an error classification and pilot study of a method for detecting errors. Qual Saf Health Care (2003) 12:443–447.
8 Kostopoulou O, Delaney B. Confidential reporting of patient safety events in primary care: results from a multilevel classification of cognitive and system factors. Qual Saf Health Care (2007) 16:95–100.
9 Shaw R, Drever F, Hughes H, Osborn S, Williams S. Adverse events and near miss reporting in the NHS. Qual Saf Health Care (2005) 14:279–283.
10 Arkes H, Harkness A. Effect of making a diagnosis on subsequent recognition of symptoms. J Exp Psychol Hum Learn Mem (1980) 6:568–575.[CrossRef]
11 Sacchi S, Cherubini P. The effect of outcome information on doctors’ evaluations of their own diagnostic decisions. Med Educ (2004) 38:1028–1034.[CrossRef][Web of Science][Medline]
12 Elstein A, Shulman L, Sprafka S. Medical Problem Solving: An Analysis of Clinical Reasoning (1978) Cambridge, MA: Harvard University Press.
13 Denig P, Witteman C, Schouten H. Scope and nature of prescribing decisions made by general practitioners. Qual Saf Health Care (2002) 11:137–143.
14 Kostopoulou O, Wildman M. Sources of variability in uncertain medical decisions in the ICU: a process tracing study. Qual Saf Health Care (2004) 13:272–280.
15 Dempsey O, Bekker H. ‘Heads you win, tails I lose’: a critical incident study of GPs’ decisions about emergency admission referrals. Fam Pract (2002) 19:611–616.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  O. Kostopoulou, B. C Delaney, and C. W Munro Diagnostic difficulty and error in primary care--a systematic review Fam. Pract., December 1, 2008; 25(6): 400 - 413. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||