Table 1 Binary features at diagnosis and their predictive value for identifying MODY.

From: Development of a clinical calculator to aid the identification of MODY in pediatric patients at the time of diabetes diagnosis

Criteria

Prevalence of criteria

Sensitivity (proportion with MODY meeting criteria)

Specificity (proportion without MODY not meeting criteria)

ROC AUC

PPV for MODY (proportion of people meeting criteria who have MODY)

NPV (proportion of people not meeting criteria who do not have MODY)

All

(n = 3541)

MODY

(n = 46)

Non-MODY

(n = 3495)

(a) Clinical features

 HbA1c < 58

8%

78%

7%

78.3% [63.6, 89.1]

93.0% [92.1, 93.8]

0.856

12.8% [9.1, 17.3]

99.7% [99.4, 99.9]

 Female Sex

45%

54%

45%

54.4% [39.0, 69.1]

55.3% [53.6, 56.9]

0.548

1.6% [1.0, 2.3]

98.9% [98.4, 99.3]

 Parent affected

14%

63%

13%

63.0% [47.6, 76.8]

87.0% [85.8, 88.1]

0.75

6.0% [4.0, 8.5]

99.4% [99.1, 99.7]

 Absence of Polyuria

7%

72%

6%

71.7% [56.5, 84.0]

93.8% [92.9, 94.5]

0.828

13.2% [9.2, 18.0]

99.6% [99.3, 99.8]

 Absence of polydipsia

8%

72%

7%

71.7% [56.5, 84.0]

92.9% [92.0, 93.7]

0.823

11.7% [8.2, 16.1]

99.6% [99.3, 99.8]

 Absence of DKA

84%

100%

83%

100% [92.3, 100]

16.7% [15.5, 18.0]

0.583

1.6% [1.1, 2.1]

100% [99.4, 100]

 Absence of Weight loss

28%

87%

27%

87.0% [73.7, 95.1]

72.7% [71.2, 74.2]

0.80

4.0% [2.9, 5.4]

99.8% [99.5, 99.9]

(b) Islet autoantibodies

 GAD only tested Negative

46%

100%

45%

100% [92.3, 100]

54.8% [53.1, 56.5]

0.774

2.8% [2.1, 3.8]

100% [99.8, 100]

 IA2 only tested Negative

30%

100%

29%

100% [92.3, 100]

70.9% [69.4, 72.4]

0.855

4.3% [3.2, 5.7]

100 [99.9, 100]

 ZnT8 only tested Negative

39%

100%

38%

100% [92.3, 100]

62.2% [60.6, 63.8]

0.811

3.4% [2.5, 4.5]

100 [99.8, 100]

 All 3 antibodies negative

12%

100%

11%

100% [92.3, 100]

89.4% [88.4, 90.4]

0.947

11.1% [8.2, 14.5]

100 [99.9, 100]

  1. PPV = positive predictive value, NPV = negative predictive value. All features are at time of diagnosis.