Walking Recovery after a Hip Fracture: A Prospective Follow-Up Study among Community-Dwelling over 60-Year Old Men and Women
Table 2
Characteristics of participants by walking trajectories in walking 500 meters (, mean ± SD, median value/ (%), value) and binary logistic regression model for statistically significant variables as a predictor for catastrophic trajectory.
Trajectory
Binary logistic regression
No-to-minor-difficulty
Catastrophic
value
Model Ie
Model IIf
OR (95% CI)
OR (95% CI)
Demographic and prefracture information
Age
27
0.035
Women (%)
27
23 (85)
54
40 (74)
0.257
Body height (cm)
27
1
Body weight (kg)
27
0.485
Number of chronic diseases
27
0.634
Osteoarthritis
27
0 (0)
54
10 (19)
0.017
Osteoporosis
27
2 (7)
54
8 (15)
0.339
Diabetes
27
1 (4)
54
8 (15)
0.134
Walking aid outdoors before fracture (%)
25
6 (24)
51
33 (65)
<0.001
5.77 (1.83–18.20)
5.49 (1.70–17.77)
Falling indoors, year before the fracture
27
3 (11)
53
17 (32)
0.041
4.49 (1.11–18.10)
4.64 (1.01–21.36)
Falling outdoors, year before the fracture
27
8 (39)
53
20 (38)
0.472
C-statistic of the model
0.79
Hospital information
Collum fracture, S72.0 (%)
27
20 (74)
54
32 (59)
0.190
Type of surgery (%)
27
54
0.454
Fixation
11 (41)
27 (50)
Hemiarthroplasty
11 (41)
22 (41)
Total hip replacement
5 (18)
5 (9)
Lowest haemoglobin after surgery (g/L)
26
51
1
Time from fracture to surgery (days)
27
0.347
Information at discharge
Difficulties in walking at the ward (%)
25
2 (8)
49
3 (6)
0.761
Offending pain at discharge (%)
25
7 (28)
46
22 (48)
0.105
Duration of inpatient period (days)
27
<0.001
1.05 (1.00–1.10)
1.05 (1.00–1.10)
Information 6 weeks after the discharge
0.70
Prescribed pain medication
27
15 (56)
54
35 (65)
0.419
Lower body pain (VAS)d
27
53
0.004
1.09 (1.02–1.16)
1.07 (1.00–1.15)
Functional balance (score)c
26
52
0.004
0.86 (0.77–0.96)
0.90 (0.81–1.01)
Knee extension force, nonfractured side (N)
27
51
0.056
Knee extension force, fractured side (N)
26
48
0.087
Handgrip force (N)
20
42
0.155
Time in home-dwelling (days)
27
54
0.235
C-statistic of the model
0.79
value is adjusted with the body weight using marginal means.
bADL: activities of daily living.
cBBS: range 0–56.
dVAS: range 0–600.
eModel I: OR for prediction in logistic regression for statistically significant variables adjusted for age and gender.
fModel II: OR for prediction in logistic regression for all statistically significant variables from Model I in one time point, adjusted for age and gender.
gVAS and knee extension force were divided by 10 for the regression analysis.
hLogistic regression analysis impossible due to quasicomplete separation and because maximum likelihood parameter does not exist. Statistically significantly different values between the study groups are bolded and values are in italic.