Skip to main content
. 2021 Apr 5;181(6):747–755. doi: 10.1001/jamainternmed.2021.0269

Table 3. Estimates of Probability of Disease Before Testing and After Positive or Negative Test Results for 5 Testing Questions .

Clinical scenario Scientific evidence range, % Median (IQR)
Resident physician estimate, % (n = 290) Attending physician estimate, % (n = 202) Nurse practitioner or physician assistant estimate, % (n = 61)
Pneumonia
Pretest probability 25-42 80 (75-90) 85 (80-90) 80 (70-90)
After positive test result 46-65 95 (90-99) 95 (95-100) 95 (90-100)
After negative test result 10-19 60 (40-80) 50 (20-80) 50 (20-50)
Breast cancer
Pretest probability 0.2-0.3 5 (1-10) 2 (1-10) 10 (5-20)
After positive test result 3-9 60 (35-75) 50 (20-80) 60 (50-80)
After negative test result <0.05 5 (1-10) 1 (1-10) 10 (2-20)
Cardiac ischemia
Pretest probability 1-4.4 10 (5-20) 5 (3-10) 15 (6.25-30)
After positive test result 2-11 75 (50-90) 60 (25-80) 90 (60-95)
After negative test result 0.43-2.5 5 (1-10) 5 (1-10) 10 (5-20)
Urinary tract infection
Pretest probability 0-1 25 (10-60) 20 (5-50) 30 (10-50)
After positive test result 0-8.3 77.5 (25-95) 90 (40-98) 90 (75-100)
After negative test result 0-0.11 5 (0.1-20) 5 (0-10) 5 (0-10)
Hypothetical testing situation
After positive test result 2 95 (95-95) 95 (80-100) 95 (95-100)
After negative test result 0 2 (0-10) 5 (0-5) 5 (5-75)

Abbreviation: IQR, interquartile range.