Table 1 Research questions to be addressed in CPET study

i) How feasible is the trial

ii) Is recruitment realistic: are families willing to be allocated randomly

iii) What is the extent of sample attrition

iv) To what extent is the trial affected by missing data, perhaps arising from inconvenience of home visits or the necessity to make hospital visits for outcome measures including DXA scans

i) How feasible is the dog-based intervention

ii) Is process evaluation favourable

iii) How should intervention content and delivery be modified for future interventions in the light of study findings

iv) What sample size would be required for a more definitive trial for each of the outcomes tested

i) Is there preliminary evidence of favourable outcomes for primary and/or secondary outcomes

ii) Is there any indication that increases in physical activity during the intervention are compensated (it is not known whether increased physical activity in dog-walking would reduce physical activity at other times; 'compensation' (15,17)

iii) Does physical activity undertaken as part of the dog walking intervention displace other forms of physical activity (e.g. going to the gym)