Dormosh et al reported a model for predicting the risk of falls for an older adult in the community. This is based on primary care electronic health records. The authors are from the University of Amsterdam and VU University in The Netherlands.
Patient selection: community-dwelling adult >= 65 years of age
Parameters:
(1) age in years
(2) sex
(3) history of falls
(4) use of proton pump inhibitor
(5) use of opioids
(6) history of previous injury (not elaborated on)
(7) depression
(8) osteoarthritis
(9) urinary incontinence
(10) problems with memory or concentration
Parameter
Finding
Points
age
0.06 * (age)
sex
male
0
female
0.26
history of falls
no
0
yes
0.72
use of proton pump inhibitor
no
0
yes
0.29
use of opioids
no
0
yes
0.24
history of previous injury
no
0
yes
0.35
depression
no
0
yes
0.54
osteoarthritis
no
0
yes
0.20
urinary incontinence
no
0
yes
0.36
memory or concentration problem
no
0
yes
0.41
value of X =
= SUM(points for all of the parameters) - 6.92
probability of falls =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.71.
To read more or access our algorithms and calculators, please log in or register.