Quinn et al reported a model for predicting a patient with multiple sclerosis who is likely to fall. This can help to identify a patient who may benefit from interventions to reduce fall injury. The authors are from University of Limerick, St. James's Hospital, and St. Vincent's University Hospital.
Patient selection: multiple sclerosis
Parameters:
(1) fall in the past 3 months
(2) visual problems
(3) bladder control problems
(4) timed up and go test in seconds
Parameter
Finding
Points
fall in past 3 months
no
0
yes
1.2
visual problems
no
1.06
yes
0
bladder control problems
no
0
yes
0.90
timed up and go
0.02 * (seconds)
value of X =
= SUM(points for all of the parameters) - 2.04
probability of falls =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.72.
To read more or access our algorithms and calculators, please log in or register.