Intended for healthcare professionals

  1. Research
  2. Extent and...
  3. Extent and determinants of error in doctors' prognoses in terminally ill patients: prospective cohort studyCommentary: Why do doctors overestimate?Commentary: Prognoses should be based on proved indices not intuition
Papers

Extent and determinants of error in doctors' prognoses in terminally ill patients: prospective cohort studyCommentary: Why do doctors overestimate?Commentary: Prognoses should be based on proved indices not intuition

BMJ 2000; 320 doi: https://doi.org/10.1136/bmj.320.7233.469 (Published 19 February 2000) Cite this as: BMJ 2000;320:469

Data supplement

  •  
     
     
     
    Table 1  Multinomial logistic regression model of relation of doctor and patient attributes to the accuracy of doctors' objective prognoses in terminally ill hospice patients*
     
    Variable
    Optimistic prognostic error 
    		 
    Pessimistic prognostic error 
     
    Odds ratio
    95% CI
    		 
    Odds ratio
    95% CI
    Patient characteristics
    Age
    1.01
    0.99 to 1.04
    		 
    1.00
    0.97 to 1.03
    Male 
    1.28
    0.70 to 2.34
    		 
    0.42
    0.18 to 0.99†
    Racial background: 
    African-American
    0.61
    0.29 to 1.26
    		 
    1.15
    0.42 to 3.18
    Non-African-American, non-white
    2.59
    0.52 to 12.89
    		 
    8.09
    1.30 to 50.30†
    Unknown
    1.73
    0.74 to 4.06
    		 
    2.49
    0.84 to 7.35
    Diagnosis:
    Cancer
    1.27
    0.57 to 2.80
    		 
    0.39
    0.14 to 1.05
    AIDS
    4.02
    0.82 to 19.57
    		 
    1.30
    0.19 to 9.04
    Duration of disease (years)
    0.93
    0.82 to 1.05
    		 
    1.02
    0.88 to 1.18
    ECOG performance status
    0.87
    0.65 to 1.16
    		 
    0.88
    0.60 to 1.30
    Doctor charateristics
    Upper quartile of practice experience
    0.37
    0.19 to 0.74†
    		 
    0.22
    0.08 to 0.61†
    Male 
    0.72
    0.30 to 1.74
    		 
    0.69
    0.24 to 1.99
    Optimist
    0.85
    0.44 to 1.66
    		 
    0.79
    0.33 to 1.90
    Board certified
    1.03
    0.46 to 2.30
    		 
    1.79
    0.57 to 5.58
    Specialty:
    Oncology
    1.32
    0.59 to 2.96
    		 
    0.55
    0.16 to 1.90
    Other internal medicine subspecialties
    1.91
    0.71 to 5.14
    		 
    3.26
    1.01 to 10.66†
    Surgery and other
    0.65
    0.22 to 1.96
    		 
    0.71
    0.14 to 3.66
    Family practice
    0.67
    0.27 to 1.69
    		 
    1.16
    0.36 to 3.75
    No of similar patients in past 12 months
    1.00
    1.00 to 1.00
    		 
    1.00
    0.99 to 1.00
    Characteristics of the doctor-patient relationship
    Referred 2 patients to hospice in the past quarter
    0.72
    0.39 to 1.34
    		 
    1.80
    0.78 to 4.15
    Duration of the patient-doctor relationship (years)
    1.07
    0.99 to 1.16
    		 
    1.12
    1.02 to 1.22†
    No of contacts in past 3 months
    1.02
    0.99 to 1.05
    		 
    1.02
    0.99 to 1.06
    Days since last examination
    0.99
    0.99 to 1.00
    		 
    0.97
    0.94 to 0.99†

    * Optimistic prognosis and pessimistic prognosis are compared with accurate prognosis. All dichotomous variables are coded as 0 = absent and 1 = present. The omitted (reference) category for racial background is white, for diagnosis is "non-cancer, non-AIDS," and for doctor’s specialty is general medicine and geriatrics. Eastern Cooperative Oncology Group (ECOG) performance status is measured on a continuous 0-4 scale.

    †P 0.05.

Related articles

See more

Back to top