Skip to main content

Table 2 Between-group differences and adjusted regression analyses

From: Dynamic balance and instrumented gait variables are independent predictors of falls following stroke

 

Non-fallers (n = 58)

Fallers (n = 23)

Between-groupa

25th – 75th quartiles

Regression adjusted for country, prior falls and assistance

Regression adjusted for country, prior falls, assistance and 6mWT

 

Mean ± SD

Mean ± SD

P-value

 

IQR-OR (95% CI)

P-value

IQR-OR (95% CI)

P-value

Gait variables

 6mWT-comfortable, m/s (n = 57:23)

0.78 ± 0.36

0.61 ± 0.32

0.075

0.45–0.98

2.10 (0.90–4.89)b

0.086

N/A

N/A

 Kinect-fast walk, m/s (n = 56:23)

0.93 ± 0.38

0.76 ± 0.37

0.106

0.57–1.16

1.89 (0.78–4.58)b

0.157

N/A

N/A

 Stride length, m (n = 54:20)

1.03 ± 0.26

0.87 ± 0.31

0.043*

0.76-1.22

4.23 (1.14–15.66)b

0.031*

1.97 (0.26–14.79)b

0.511

 Cadence, steps/min (n = 54:20)

107.84 ± 23.50

95.48 ± 31.61

0.114

91.41–122.08

2.48 (0.96–6.41)b

0.060

1.46 (0.45–4.81)b

0.530

 Step width, m (n = 54:21)

0.15 ± 0.04

0.15 ± 0.04

0.943

0.12–0.17

1.19 (0.56–2.52)b

0.649

1.43 (0.65–3.13)b

0.376

 Step length asymmetry, ratio (n = 53:20)c

1.15 ± 0.16

1.54 ± 1.10

0.458

1.06–1.16

1.37 (1.01–1.85)

0.040*

1.28 (0.96–1.69)

0.090

 Gait speed variability, m/s (n = 56:23)c

0.17 ± 0.09

0.13 ± 0.04

0.048*

0.11-0.18

1.82 (0.78–4.26)b

0.169

1.24 (0.46–3.37)b

0.670

 ML pelvic displacement, cm (n = 55:23)

7.00 ± 1.26

5.24 ± 1.39

< 0.001*

5.38-7.22

7.85 (2.56–24.05)b

< 0.001*

6.75 (2.20–20.74)b

0.001*

 Vertical pelvic displacement, cm (n = 55:23)

3.42 ± 1.25

2.83 ± 0.96

0.047*

2.26-4.10

2.03 (0.74–5.56)b

0.168

1.25 (0.36–4.31)b

0.724

Balance variables

 TUG - normal, s (n = 58:23)c

16.49 ± 9.26

27.61 ± 23.84

0.014*

11.30-22.56

4.06 (1.44–11.46)

0.008*

7.84 (1.42–43.26)

0.018*

 TUG - dual task, s (n = 54:19)c

21.47 ± 14.60

24.22 ± 3.06

0.148

11.91–25.44

1.30 (0.50–3.41)

0.589

1.38 (0.35–5.39)b

0.647

 Step test - affected, taps/15 s (n = 58:23)

10.19 ± 5.08

6.35 ± 4.20

0.005*

6.00-12.50

4.50 (1.64–12.35)b

0.004*

5.28 (1.46–19.16)b

0.011*

 Step test - less affected, taps/15 s (n = 58:23)

8.95 ± 4.81

4.74 ± 3.44

< 0.001*

4.50-10.00

6.98 (2.06–23.61)b

0.002*

10.29 (2.26–46.84)b

0.003*

 COP vel EO total, cm/s (n = 57:22)c

1.52 ± 0.65

1.96 ± 1.13

0.186

1.14–1.36

1.26 (0.61–2.61)

0.527

1.29 (0.61–2.76)

0.505

 COP vel EO ML, cm/s (n = 57:22)c

0.68 ± 0.35

0.83 ± 0.50

0.193

0.43–0.86

1.40 (0.70–2.80)

0.345

1.31 (0.64–2.69)

0.458

 COP vel EO AP, cm/s (n = 57:22)c

1.20 ± 0.55

1.59 ± 1.00

0.128

0.89–1.40

1.18 (0.64–2.17)

0.590

1.24 (0.66–2.34)

0.508

 COP vel EC total, cm/s (n = 56:22)c

2.11 ± 0.94

2.75 ± 1.64

0.172

1.45–2.86

1.17 (0.49–2.78)

0.725

1.20 (0.49–2.93)

0.687

 COP vel EC ML, cm/s (n = 56:22)c

0.79 ± 0.37

0.99 ± 0.53

0.127

0.52–1.10

1.63 (0.60–4.44)

0.338

1.48 (0.53–4.14)

0.461

 COP vel EC AP, cm/s (n = 56:22)c

1.78 ± 0.84

2.35 ± 1.54

0.230

1.21–2.32

1.06 (0.49–2.32)

0.882

1.12 (0.50–2.50)

0.788

  1. *Denotes a significance at P < 0.05
  2. a Between-group differences assessed using Mann-Whitney U tests; b an inverse relationship with falls exists, therefore IQR-OR compares the odds of having more falls among individuals with predictor values at the 25th percentile with the odds of having more falls among individuals with predictor values at the 75th percentile; c log-transformed for regression analysis due to significant positive skew
  3. Abbreviations, 6mWT 6-m walk test, AP anteroposterior, COP centre of pressure, EC eyes closed, EO eyes open, IQR interquartile range, ML mediolateral, OR odds ratio, TUG timed up and go
Back to article page

Journal of NeuroEngineering and Rehabilitation

ISSN: 1743-0003

Contact us