Table 2 Summary of analysis 1 – effectiveness at 12 months

Determine the effect of the AFLY5 intervention on primary outcomes

ITT analysis

DAL

Feb –March 2013

Aug 2013

Multivariable multi-level linear regression (continuously measured outcomes), with adjustment for baseline variables

LH

RRK

TJP

Determine the effect of the AFLY5 intervention on secondary outcomes

ITT analysis

DAL

Feb – March 2013

Aug 2013

Multivariable multi-level linear regression (continuously measured outcomes) and multivariable multi-level logistic regression (for the two binary – general and central overweight/obese – outcomes), with adjustment for baseline variables

LH

RRK

TJP

Complete secondary analyses to determine effect in those who completed the intervention as intended

Per-protocol analysis, excluding children from the intervention schools in which fewer than 11 lessons were taught. Multivariable multi-level linear or logistic regression (as above), with adjustment for baseline variables

DAL

Mar – Apr 2013

Aug 2013

LH

TJP

Sensitivity analyses to determine whether any effect of the intervention on primary outcomes based on accelerometer data vary by weekend or weekday

ITT analysis

DAL

Mar – Apr 2013

Aug 2013

Multivariable multi-level linear regression (continuously measured outcomes), with adjustment for baseline variables

LH

TJP

Additional secondary analyses

Complete secondary analyses explore whether associations differ by gender and area deprivation

Stratified analyses (by gender and separately by school area deprivation)

DAL

Mar – Apr 2013

Not for journal but will be reported to funder (see below)

LH

ITT analysis

TJP

Multivariable multi-level linear or logistic regression (as above), with adjustment for baseline variables.

Test of interaction between gender × intervention and deprivation × intervention