Skip to main content
. Author manuscript; available in PMC: 2018 Oct 3.
Published in final edited form as: JAMA Pediatr. 2017 Dec 1;171(12):1151–1159. doi: 10.1001/jamapediatrics.2017.2960

Table 4.

Mediators of Maintenance Treatment Differences in Percentage Overweight Outcomes

Mediator Variable and Treatment Contrast Path Aa Path Bb Path Cc Path Cd Path AB (95% Cl)e % Mediatedf
Monitoring and Goal Setting to Support Child
CONTROL vs LOW 0.29g -3.07g -0.90 -1.79g -0.89 (−1.79 to-0.29) 50
CONTROL vs HIGH 0.48g -2.93g -1.92g -3.33g -1.42 (−2.57 to-0.59) 42

Family and Home Environment to Support Child
CONTROL vs LOW 0.09 -2.42 -1.51 -1.72g -0.21 (−0.78 ta 0.05) NA
CONTROL vs HIGH 0.15g -5.7Sg -2.38g -3.25g -0.88 (−1.66 to-0.34) 27

Healthy Child Behavior With Peers
CONTROL vs LOW 0.14 -1.86 -1.72g -1.98g -0.25 (−1.03 to 0.03) NA
CONTROL vs HIGH 0.26g -3.34g -2.53g -3.39g -0.86 (−1.76 to-0.27) 25

Abbreviations: CONTROL, weight management education condition; HIGH, 32 sessions of enhanced social facilitation maintenance; LOW, 16 sessions of enhanced social facilitation maintenance; NA, not applicable.

a

Path A is the effect of treatment on changes in the mediator variable from month 4 to month 12.

b

Path B is the effect of change in the mediator on change in the weight outcome.

c

Path C’ is the direct effect of treatment on weight outcome, when accounting for the mediated effect.

d

Path C is the total effect of treatment on the weight outcome.

e

Path AB is the product of paths A and B and represents the indirect (mediated) effect

f

Percentage mediated is the ratio of the indirect effect (path AB) to the total effect (path C) and is shown only for statistically significant indirect effects as determined by 95% CI.

g

P<.05.