Model of Baek et al for Predicting Relapse of Graves' Disease

Description

Baek et al reported a model for predicting the risk of relapse for a patient with Graves' disease. This can help to identify a patient who may require more aggressive management. The authors are from the Catholic University of Korea and multiple associated hospitals.

Patient selection: history of Graves' disease

Outcome: relapse within 2 years of stopping antithyroid drugs

Parameters:

(1) age in years

(2) sex

(3) thyrotropin-binding inhibitory immunoglobulin (TBII; thyrotropin receptor antibody)

(4) thyroid stimulatory immunoglobulin (TSI) bioassay

Parameter	Finding	Points
age in years	< 46 years	62.5
	>= 46 years	0
sex	male	100
	female	0
thyrotropin-binding inhibitory immunoglobulin	negative	0
	positive	22.5
thyroid stimulatory immunoglobulin	negative	0
	positive	67

total score =

= SUM(points for all of the parameters)

Interpretation:

• minimum score: 0

• maximum score: 252.1

• The higher the score the greater the risk of relapse.

value of X =

= (0.01003 * (score)) - 1.423

probability of relapse =

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

Performance:

• The area under the ROC curve is 0.67.

• A cutoff of 111.4 has a sensitivity of 57% and specificity of 71%.

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