Cognitive errors in diagnosis: Instantiation, classification, and consequences

Kassirer, Jerome P.; Kopelman, Richard I.

doi:10.1016/0002-9343(89)90342-2

Cited by 175 publications

(105 citation statements)

References 49 publications

Supporting

Mentioning

101

Contrasting

Unclassified

Order By: Relevance

“…However, diagnostic errors remain common, as they range from <5% in radiological or pathological specialties up to 10-15% in most other clinical fields 24 . Inaccurate probability estimation is considered a substantial contributor to these errors 4,10,23,26,27 . In light of our results and of previous work, increased risk of diagnostic error and unnecessary testing may result either from the overestimation of positive predictive value when pre-test probability is low (as in our scenario) or the overestimation of negative predictive value when pre-test probability is high.…”

Section: Interpretation Of the Results And Implicationsmentioning

confidence: 99%

Does Prevalence Matter to Physicians in Estimating Post-test Probability of Disease? A Randomized Trial

et al. 2010

View full text Add to dashboard Cite

BACKGROUND: The probability of a disease following a diagnostic test depends on the sensitivity and specificity of the test, but also on the prevalence of the disease in the population of interest (or pre-test probability). How physicians use this information is not well known. OBJECTIVE:To assess whether physicians correctly estimate post-test probability according to various levels of prevalence and explore this skill across respondent groups. DESIGN: Randomized trial. PARTICIPANTS: Population-based sample of 1,361 physicians of all clinical specialties. INTERVENTION: We described a scenario of a highly accurate screening test (sensitivity 99% and specificity 99%) in which we randomly manipulated the prevalence of the disease (1%, 2%, 10%, 25%, 95%, or no information). MAIN MEASURES: We asked physicians to estimate the probability of disease following a positive test (categorized as <60%, 60-79%, 80-94%, 95-99.9%, and >99.9%). Each answer was correct for a different version of the scenario, and no answer was possible in the "no information" scenario. We estimated the proportion of physicians proficient in assessing post-test probability as the proportion of correct answers beyond the distribution of answers attributable to guessing. KEY RESULTS: Most respondents in each of the six groups (67%-82%) selected a post-test probability of 95-99.9%, regardless of the prevalence of disease and even when no information on prevalence was provided. This answer was correct only for a prevalence of 25%. We estimated that 9.1% (95% CI 6.0-14.0) of respondents knew how to assess correctly the post-test probability. This proportion did not vary with clinical experience or practice setting. CONCLUSIONS: Most physicians do not take into account the prevalence of disease when interpreting a positive test result. This may cause unnecessary testing and diagnostic errors.KEY WORDS: Bayes' theorem; predictive value of tests; prevalence; sensitivity and specificity; diagnosis; risk assessment; probability; evidence-based medicine.

show abstract

Section: Interpretation Of the Results And Implicationsmentioning

confidence: 99%

Does Prevalence Matter to Physicians in Estimating Post-test Probability of Disease? A Randomized Trial

et al. 2010

View full text Add to dashboard Cite

show abstract

“…Representativeness is the cognitive bias that uses resemblance as a quick way to assess risk. 25,26 To the extent that older BRCA1/2 mutation carriers are representative of older women in general, physicians may have interpreted their 10-year risk of breast cancer as being greater than the risk for younger mutation carriers. This erroneous interpretation was made despite the incidence data, which showed that the probability that 60-year-old BRCA1/2-positive women will develop breast cancer during the next decade is less than the 10-year incidence of breast cancer among 40-year-old BRCA1/2-positive women.…”

Section: Discussionmentioning

confidence: 99%

Reporting BRCA test results to primary care physicians

Sandhaus

Singer

Dawson

et al. 2001

Genetics in Medicine

View full text Add to dashboard Cite

Purpose: The main purpose of this study was to determine if comprehension of the cancer risk information presented in a hypothetical report for BRCA1/2 gene analyses was influenced by the format in which the information was presented. A secondary objective was to determine physician characteristics that might influence comprehension of the report. Methods: A survey was conducted in which a case vignette describing a young woman at high risk for carrying a BRCA mutation was presented. Survey participants, all primary care practitioners, were asked to interpret a laboratory report that provided the patient's BRCA1/2 test result and accompanying data about the cumulative risk and incidence rates of breast cancer for BRCA1/2 mutation carriers and the general population.These data were presented in the report in either a tabular or a graphic format. The main outcome was measured by the responses to four questions that addressed the probabilistic cancer risk information. Physician predictor variables included medical specialty, practice setting, years in practice, continuing medical education in genetics, and knowledge of cumulative risk. Results: Knowledge of cumulative risk was the only physician variable that influenced comprehension of the cancer risk information (OR ϭ 31.9; P Ͻ 0.001). After adjusting for this variable, the graphic format tended to perform better than the tabular format in conveying breast cancer risk information (OR ϭ 3.1; P ϭ 0.102). Conclusions: Many physicians may be unprepared to interpret genetic risk information, due to lack of understanding of basic epidemiologic terms used to express the risk of disease. Genet Med 2001:3(5): 327-334.

show abstract

“…Several studies have shown deficiencies in using pre-test probability when interpreting test results. [7][8][9][10][11][12][13][14][15][16] "Affected until proved otherwise . .…”

Section: The Basics Of Bayesian Logicmentioning

confidence: 99%

Evidence based diagnosis: does the language reflect the theory?

2006

BMJ

View full text Add to dashboard Cite

Cognitive errors in diagnosis: Instantiation, classification, and consequences

Cited by 175 publications

References 49 publications

Does Prevalence Matter to Physicians in Estimating Post-test Probability of Disease? A Randomized Trial

Does Prevalence Matter to Physicians in Estimating Post-test Probability of Disease? A Randomized Trial

Reporting BRCA test results to primary care physicians

Evidence based diagnosis: does the language reflect the theory?

Contact Info

Product

Resources

About