Zeng et al reported a model for predicting discharge to a post-acute care facility after a total knee arthroplasty (TKA). This can help to identify a patient with need for ongoing assistance in the recovery period. The authors are from Kaiser Permanente Colorado and University of Colorado.
Patient selection: status post total knee arthroplasty (TKA)
Parameters:
(1) age in years
(2) any help needed for basic hygiene and needs
(3) how much mobility affects the ability to care for home and to run errands
(4) number of days per week of moderate-to-strenuous exercise
(5) someone staying on hand to help during the recovery
(6) reliable source of transportation
(7) dementia
(8) other neurological disorder
Parameter |
Finding |
Points |
age in years |
|
0.072 * (age) |
help needed |
no |
0 |
|
yes |
1.169 |
impact mobility status |
none |
0 |
|
sometimes |
0.473 |
|
most of the time or cane |
1.302 |
|
all of the time with assist device |
0.724 |
|
all of the time plus balance issues |
1.917 |
days with exercise |
0 |
1.31 |
|
1 or 2 |
0.75 |
|
3 or more |
0 |
someone staying with |
all of the time |
0 |
|
most of the time |
0.514 |
|
sometimes |
1.364 |
|
none |
1.340 |
reliable transportation |
all of the time |
0 |
|
most of the time |
0.786 |
|
sometimes but difficult to find |
0.741 |
|
no one |
0.905 |
dementia |
no |
0 |
|
yes |
1.202 |
other neurologic disorder |
no |
0 |
|
yes |
2.005 |
total score =
= SUM(points for all of the parameters) - 10.067
probability of discharge to a post-acute care facility =
= 1 / (1 + EXP((-1) * (score)))
Performance:
• The area under the ROC curve is 0.87.
Specialty: Surgery, orthopedic