Description

Tabaei and Herman developed an equation to predict diabetes mellitus based on logistic regression analysis. This can help identify a patient who should be a candidate for additional testing and closer monitoring. The authors are from the University of Michigan in Ann Arbor.

Parameters:

(1) age of the patient in years

(2) body mass index in kilograms per square meter

(3) random plasma glucose in mg/dL

(4) postprandial interval in hours at time of plasma sample (from 0 up to maximum of 8 for >= 8 hours)

(5) gender

 Gender Points male 0 female 1

where:

• Equations for converting whole blood glucose to plasma glucose in given in 13.05.07.

X =

= (-10.0382) + (0.0331 * (age in years)) + (0.0308 * (random glucose)) + (0.25 * (postprandial interval)) + (0.5620 * (points for gender)) + (0.0346 * (BMI))

probability of diabetes =

= 1 / (1 - EXP((-1) * X))

NOTE: Almost all of the logistic regression equations I have seen have had the form P = (1 / (1 + EXP((-1) * X))). When the equation was implemented as described, unreasonable values are returned. More reasonable results are returned when the typical equation is used.

alternative equation for probability of diabetes giving more reasonable results =

= 1 / (1 - EXP((-1) * X))