Providing information for patients and professionals on research and clinical care in genetic types of diabetes.

MODY Probability Calculator

**Please note work on this model is still in progress and further validation needs to be undertaken**

This is for use in patients diagnosed with diabetes under the age of 35 and was developed on a European Caucasian cohort.
Enter the clinical features of the patient in the form below and press the "Calculate Probability" button.
Age at diagnosis (years)
Sex
 Male Female
Currently treated with insulin or OHA? 
Yes No
Time to Insulin Treatment
(if currently treated with insulin)
Not currently treated with insulin  
Within 6 months of diagnosis               
Over 6 months after diagnosis       
BMI (kg/m2)
HbA1c (%)
  or mmol/mol  
Current Age (yrs)
Parent affected with diabetes?
Yes No

        

 

Based on the clinical features entered into the calculator, the post-test probability (Positive Predictive Value (PPV)) of your patient having MODY is > % i.e. a 1 in chance or lower of testing positive for MODY

As , this is based on a background prevalence level for MODYa of i.e. a 1 in chance of having MODY.

Further screening tests that can be done prior to genetic testing:

If your patient is insulin-treated you can improve the post-test probability by testing for endogenous insulin secretion using urinary C-peptide creatinine ratio (UCPCR) or for pancreatic autoimmunity using GAD/IA2 antibodies to rule out Type 1 diabetes.  Both tests are highly specific and sensitive markers discriminating MODY from Type 1 diabetes (Besser et al. 2011; McDonald et al, 2011).

At what PPV level should I test?

The current pick-up rate for MODY testing in the UK is 25%, but where a diagnosis would make a difference to a patient's management, one might like to consider testing at a lower probability.

The higher the threshold, the less likely you are to pick up false positives (send patients for testing who are not MODY), but the more likely you are to miss a genuine case of MODY (false negatives). 

  • If you use a PPV as the basis to refer for genetic testing you would expect to get  false negatives and false positives for every 100 patients you test

  • If you use a PPV as the basis to refer for genetic testing you would expect to get  false negatives and false positives for every 100 patients you test

  • If you use a PPV as the basis to refer for genetic testing you would expect to get  false negatives and false positives for every 100 patients you test

If you would like to test this patient for MODY, please click here

aWe have assumed background prevalences of 0.7% MODY for those who are insulin treated within 6 months of diagnosis, and 4.6% MODY for those who are not treated insulin treated within 6 months of diagnosis.

Copyright University of Exeter. 

Visits