Jatuworapruk et al reported a model for predicting a flare of gout in an inpatient. The authors are from the University of Otago (New Zealand), Thammasat University (Thailand) and the University of Auckland.
Patient selection: hospital inpatient with a history of gout
Outcome: inpatient flare
Parameters:
(1) pre-admission urate-lowering agent
(2) adjustment of urate-lowering agent during admission and prior to flare
(3) adjustment of diuretics during admission and prior to flare
(4) pre-admission serum urate in mmol/L
(5) tophus
(6) pre-admission gout prophylaxis
(7) gout prophylaxis started or increased during admission and prior to flare
(8) acute kidney injury
Parameter
Finding
beta coefficient
preadmission urate lowering agent
no
1.318
yes
0
adjustment of urate lowering agent during admission
no
0
yes
0.989
adjustment to diuretics during admission
no
0
yes
0.952
pre-admission serum urate
<= 0.36 mmol/L
0
> 0.36 mmol/L
1.079
tophus
absent
0
present
1.301
pre-admission gout prophylaxis
no
1.898
yes
0
gout prophylaxis started or increased during admission
no
0
yes
2.540
acute kidney injury
no
0
yes
0.752
X =
= SUM(beta-coefficients) - 5.393
probability of flare =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.80.
To read more or access our algorithms and calculators, please log in or register.
