Skip to main content

Table 7 Results of the mixed-effects meta-regressions for the continuous moderator variables on the effect sizes obtained from measures of knowledge in the posttest

From: Preventive physiotherapy interventions for back care in children and adolescents: a meta-analysis

Cluster / Moderator variable

k

Min.

Max.

Mean

SD

b j

Q R

Q E(df)

R 2

(A) Treatment characteristics:

         

Treatment duration (weeks)

15

1.0

96.0

17.5

32.1

−0.012

6.60**

144.46(13)***

0.430

Treatment intensity (hours/week)

12

0.2

2.4

0.9

0.6

−0.059

0.03

142.28(10)***

0.0

Treatment magnitude (total hours)

12

0.2

19.0

5.0

5.0

−0.001

0.00

128.72(10)***

0.026

(B) Subject characteristics:

         

Mean age (years)

14

8.0

12.0

1.2

1.2

0.130

0.48

345.18(12)***

0.0

Gender (% male)

14

45.8

58.2

3.7

3.7

0.029

0.20

384.63(12)***

0.0

(C) Methodological characteristics:

         

Differential attrition

16

0.0

0.033

0.004

0.009

−24.205

1.14

355.39(14)***

0.0

Quality score

16

3.4

7.5

6.0

1.1

0.282

2.71

267.05(14)***

0.177

(D) Extrinsic characteristic:

         

Publication year

16

1984

2011

2003

8.4

−0.014

0.28

370.92(14)***

0.0

  1. * p < .05. ** p < .01. *** p < .001. k: number of studies. Min. and Max.: minimum and maximum values of the moderator variable. SD: standard deviation of the moderator variable. b j: unstandardized regression coefficient associated to the moderator variable. Q R: statistic for testing the significance of the moderator variable. Q E: statistic for testing the model misspecification. df: degrees of freedom. R 2: proportion of variance accounted for by the moderator variable.