Table 4 Quality scores and results of included studies for evaluating effectiveness of biofeedback-based interventions.

Hatzitaki et al[38]

1 | 5

rANOVA & post-hoc testing.

Significant interactions between group and time,

in favor of experimental group (E1^c),

for 2 of 4 asymmetry and 4 of 4 sway outcomes for tandem standing.

Asymmetry = 1.40 & 1.08

Sway = 0.38, 0.56, 0.69, 0.78

Heiden & Lajoie[39]

1 | 5

rANOVA & post-hoc testing.

Significant interactions between group and time,

in favor of experimental group, for RT and CB&M.

RT, CB&M = - (values are given in bar charts)

Lajoie[42]

1 | 4

rANOVA & post-hoc testing.

Significant between-group differences for RT and

BBS at posttest in favor of experimental group.

RT, BBS = - (values are given in bar charts)

Rose & Clark[46]

1 | 2

Doubly multivariate rANOVA & post-hoc testing.

Significant interactions between group and time in

favor of experimental group.

Sway = .51

Weight-shifting = .38 & .79 & .85

BBS = .46; TUG = .55

Sihvonen et al[48, 49]

2 | 6

rANOVA & Friedman's test.

Significant interactions between group and time, in

favor of experimental group, for 2 of 6 weight-shifting,

4 of 18 sway outcomes and BBS. Significant improvement in

activity level in experimental group.

Sway = .56 & .86 to 1.12

Weight-shifting = .77 & 1.29

BBS = .34

Activity level = - (categorical variable)

Wolf et al[52]

0 | 4

rANOVA with baseline characteristics and sway as

covariates & post-hoc testing.

Significant between-group differences in improvement

for 5 of 12 sway outcomes in favor of experimental group.

Sway = .43 & .89 to 1.71

Ref

Quality ^a

EV PEDro

Analysis ^b

Main short-term results

Effect sizes (absolute numbers)

Cheng et al[30]

1 | 4

rANOVA & post-hoc testing.

Significant between-group differences in

weight-shifting at posttest in favor of experimental group.

Weight-shifting = .59 & .78 to .90

Grant et al[35]

2 | 5

rANOVA & post-hoc testing.

No significant between-group difference.

Sackley & Lincoln[47]

2 | 6

Student's t-test & Mann-Whitney U-test.

Significant between-group differences in weight- distribution,

ADL and motor function at post-test in favor of experimental group.

Weight-distribution = .99

ADL = 1.21

Motor function = .99

Shumway et al[57]

2 | 4

Chi-square test.

Significant between-group difference in change score

for weight-distribution in favor of experimental group.

Weight-distribution = - (values are given in box plots)

Walker et al[51]

2 | 6

rANOVA & post-hoc testing.

No significant between-group differences.

Yavuzer et al[55]

2 | 6

Mann-Whitney U-test.

Significant between-group differences in change scores

for 2 of 17 gait outcomes in favor of experimental group.

Pelvic obliquity = .55^d

Peak vGRF paretic side = .54

Ref

Quality ^a

EV PEDro

Analysis ^b

Main short-term results

Effect sizes (absolute numbers)

Aruin et al[26]

0 | 4

rANOVA

Significant between-group difference after

the intervention in favor of experimental group.

Step width = - (mean (SE) are given: .09 m(.003) to .16 m(.006) vs. 10 m(.004) to .13 m(.003))

Bradley et al[28]

2 | 5

Mixed model rANOVA (sign. if ?).

No significant between-group differences.

Montoya et al[44]

1 | 3

Factorial rANOVA^e.

Significant between-group difference, interaction

between beginning/end and group, interaction between

session and group, all in favor of experimental group.

Step length = 3.33

Morris et al[45]

2 | 7

Mann-Whitney U-test.

Significant between-group difference for reduction in peak

knee extension during phase 2 in favor of experimental group.

Peak knee extension = - (mean reduction (SD) are given: 1.7°(1.8) vs. 4°(3.1) (phase 2))

Ref

Quality ^a

EV PEDro

Analysis ^b Main short-term results

Effect sizes (absolute numbers)

Cheng et al[29]

1 | 5

- (only long-term results)

- (only long-term results)

Engardt et al[32]

2 | 5

Student's t-test (sign. if p < .01) & Mann-Whitney U-test.

Significant between-group differences in improvement

for weight-distribution and functional sit-to-stand

in favor of experimental group.

Weight-distribution = 1.16 & 1.47

Functional sit-to-stand = - (median (range) are given: 2(2) to 6(2-6) vs 2(2) to 4(2-6))

Ref

Quality ^a

EV PEDro

Analysis ^b

Main short-term results

Effect sizes (absolute numbers)

Gauthier et al[56]

2 | 4

Mann-Whitney U-test.

No significant between-group differences^f.

Hershko et al[40]

2 | 5

Student's t-test & Chi-square test.

The experimental groups improved significantly

in PWB, whereas the control groups did not.

PWB = 1.22 (groups with Touch WB instruction) & 1.40 (groups with Partial WB instruction)

Isakov[41]

1 | 4

Student's t-test.

Significant between-group difference in

improvement in favor of experimental group.

FWB = - (mean improvement (SD) are given: 7.9 kg(5.3) vs. 7 kg(2.4)