Skip to main content

Table 1 Probability conversion table for MODY using the Original method with an adjusted pre-test probability. (A) Model using clinical features only adjusting to population prevalence of MODY (0.6%), (B) model using clinical features but adjusting to population prevalence of MODY based on patients who are C-peptide positive and antibody negative (7.3%). Parentheses are used to signify that an endpoint value is not included. Bracket are used to signify that an endpoint value is included

From: Comparison of Bayesian approaches for developing prediction models in rare disease: application to the identification of patients with Maturity-Onset Diabetes of the Young

(A) Original model adjusting probabilities for prevalence [scenario a)] (pre-test probability = 0.6%)

(B) Original model adjusting probabilities assuming patients have biomarker tests suggesting increased risk of MODY [scenario b)] (pre-test probability=7.3%)

Case-control model probability (%)

Post-recalibration probability (%)

Case-control model probability (%)

Post-recalibration probability (%)

(0, 10)

0.6

(0, 10)

7.3

[10, 20)

1.6

[10, 20)

17.9

[20, 30)

2.3

[20, 30)

23.2

[30, 40)

3.4

[30, 40)

31.6

[40, 50)

4.2

[40, 50)

36.6

[50, 60)

5.5

[50, 60)

43.1

[60, 70)

6.2

[60, 70)

46.5

[70, 80)

7.1

[70, 80)

49.9

[80, 90)

10.9

[80, 90)

61.6

[90, 100]

45.4

[90, 100]

91.6