Pinheiro et al reported the Sao Paulo Osteoporosis Risk Index (SAPORI) to identify a woman with low bone mineral density who is at risk for low-impact bone fractures. Three models were reported - one for low-mineral density in femoral bone, one for low-mineral density in the spine, and one for low-impact fracture. The authors are from Universidade Federal de Sao Paulo and other institutions in Sao Paulo, Brazil.
Patient selection: pre, peri and postmenopausal women in Sao Paulo with mean age 60.1 years
Parameters:
(1) previous fragility fracture
(2) current amoking
(3) age in years
(4) ethnicity
(5) current corticosteroids (> 5 mg per day for >= 3 months)
(6) current physical activity in previous year
(7) current hormone replacement therapy
(8) weight in kilograms
Parameter
|
Finding
|
Beta-Coefficient
|
previous fragility fracture
|
no
|
0
|
|
yes
|
-0.48
|
current smoking
|
no
|
0
|
|
yes
|
-0.21`
|
age
|
|
-0.034 * (age)
|
ethnicity
|
White
|
-0.35
|
|
non-White
|
0
|
current steroids
|
no
|
0
|
|
yes
|
-0.60
|
physical activity
|
no
|
0
|
|
yes
|
0.12
|
hormone replacement
|
no
|
0
|
|
yes
|
0.25
|
weight
|
|
0.04* (weight)
|
where:
• The impact of hormone replacement may pertain to postmenopausal women.
value of X =
= SUM(points for all of the beta=coefficients) - 1.19
probability =
= 1 / (1 + EXP((-1) * X))
When the model is run, low values are returned for data associated with osteoporosis and higher values are returned when data associated with better bone mineral data is entered. It seems that the score gives the probability of normal bone density. Therefor:
probability of low bone mineral density in the femur =
= 1 - (probability form model)