Al-Rubaie et al reported a model for predicting pre-eclampsia in a pregnant woman in Australia. The authors are from the University of Notre Dame Australia, University of Sydney, Macquarie University, University of Toronto and affiliated hospitals.
Patient selection: pregnant woman in Australia
Parameters:
(1) maternal age in years
(2) body mass index (BMI) in kg per square meter
(3) Australian/New Zealand born, English speaker
(4) multiple pregnancy
(5) family history of pre-eclampsia
(6) autoimmune disease
(7) chronic hypertension
(8) chronic renal disease
Parameter
|
Finding
|
Points
|
maternal age
|
|
0.052 * ((age) - 27)
|
BMI
|
|
0.078 * ((BMI) - 26.3)
|
Australian/New Zealand born English speaker
|
no
|
0
|
|
yes
|
0.525
|
multiple pregnancy
|
no
|
0
|
|
yes
|
1.318
|
family history of pre-eclampsia
|
no
|
0
|
|
yes
|
1.74
|
autoimmune disease
|
no
|
0
|
|
yes
|
1.512
|
chronic hypertension
|
no
|
0
|
|
yes
|
1.545
|
chronic renal disease
|
no
|
0
|
|
yes
|
1.494
|
• The parameter Australian/New Zealand born English speaker would be difficult to apply elsewhere. The Australian/New Zealand born indicates non-immigrant and this probably includes mostly people who are Caucasian.
value of X =
= SUM(points for all of the parameters) - 7.786
probability of pre-eclampsia =
= 1 / (1 + EXP((-1) * X))
Interpretation:
• A probability of pre-eclampsia >= 8% indicates need to consider aspirin prophylaxis.
Performance:
• The area under the ROC curve is 0.66.