Description

Glumer et al developed a risk score to help identify patients who should be screened for diabetes mellitus. This can aid in the early recognition of undiagnosed diabetic patients. The authors are from the Steno Diabetes Center in Gentofte, Glostrup University Hospital and the University of Aarhus in Denmark.


Patient selection: The study subjects were all Danish (European Caucasian) with age range apparently 30 to 60 years.

 

Parameters:

(1) age

(2) gender

(3) body mass index (BMI)

(4) known hypertension

(5) physical activity during leisure time

(6) parental history of diabetes mellitus

 

Parameter

Finding

Points

age of the patient

<= 40

0

 

45

7

 

50

13

 

55 - 60

18

gender of the patient

male

4

 

female

0

BMI

< 25

0

 

25 - 29

7

 

>= 30

15

hypertension

absent

0

 

present

10

leisure time activity

inactive

6

 

active

0

parental diabetes

no

0

 

yes

7

 

where:

• The point assignment for patients in their 40's and early 50's is ambiguous. If the data points are entered into JMP the following equation is generated:

 

points for age from 40 to 55 =

= (-0.02 * ((age in years)^2)) + (3.1 * (age in years)) - 92

 

total score =

= SUM(points for all 6 parameters)

 

Interpretation:

• minimum score: 0

• maximum score: 60

• The selected cutoff for targeted screening was 31. I will assume that this means >= 31.

 

Performance at selected cutoff:

• The sensitivity was 73% and specificity was 74% (Youden Index 0.46).


To read more or access our algorithms and calculators, please log in or register.