Cox 2005.

Methods	Design: Parallel RCT Setting: Single site paediatric oncology unit, USA
Participants	Number Intervention: n = 131 (4 lost to follow‐up) Control: n = 135 (1 lost to follow‐up) Age at study entry Group:12‐18 years Intervention (mean ± SD): 15.09 ± 1.90 years Control (mean ± SD): 14.96 ± 1.97 years Sex Intervention: 57 males: 74 females Control: 61 males: 74 females Diagnosis Leukaemia/lymphoma: Intervention: 73 Control: 72 Solid tumour: Intervention: 58 Control: 63 Treatment Information not available Age at diagnosis Information not available Time since diagnosis Intervention (mean ± SD): 15.09 (1.90) years Control (mean ± SD): 10.31 (2.94) years Inclusion criteria 12‐18 years In remission 2+ years from completion of therapy Adequate cognitive functioning English as a primary language Exclusion criteria Not USA residents English not their primary language
Interventions	Intervention The intervention consisted of standard care plus a single multi‐behavioural intervention provided by a clinical physician or nurse practitioner during a routine visit to the long‐term follow‐up clinic. The multi‐behavioural intervention consisted of: discussion of after therapy clinical summary health behaviour training of health goal health goal commitment to practice Telephone reinforcement of the education was provided at 3 and 6 months after their initial clinic visit Cointerventions: Changing the health behaviour practices of smoking cessation, sun protection and exercise Contraindications: None Control Group Standard care consists of: breast or testicular self‐examination targeted late effects screening clinical assessment late effects risk counselling Cointerventions Other health behaviour practices were targeted during the intervention. These included; smoking cessation, sun protection and exercise
Outcomes	The outcomes were measured at baseline and 12 months postintervention for both the intervention and control groups Outcome measure: Behavioural change 1) Frequency of nutrition as a health protective behaviour 2) Frequency of junk food consumption as a health risk behaviour
Notes	Study sponsors Oncology Nursing Society Foundation (2003‐2005) American Lebanese Syrian Associated Charities (ALSAC)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: " The randomisation was stratified by gender and age because of the clinical impression that risk perception could carry by gender or age"
Allocation concealment (selection bias)	Unclear risk	Comment: Although randomisation was performed using the procedure as set out by Zelen 1974, it was unclear which actual randomisation technique was used
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Comment: The study reported that five participants (four in the intervention group and one in the control group) were lost to follow‐up. There was no discussion on how this data were handled. We were unable to assess how this would influence the outcome or whether this would have a clinically relevant effect
Selective reporting (reporting bias)	High risk	Comment: This study presented a secondary analysis of data. This analysis was not in the original publication of the results
Other bias	Low risk	Comment: Minimal baseline imbalance: At baseline, there was no significant difference between the intervention and the control group for demographic and other reported characteristics No differential diagnostic activity: All assessments were performed at baseline and follow‐up for both the intervention and the control group
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: This study does not discuss whether participants or personnel were blinded. Due to the nature of the study and the form of the intervention, it would be impossible for the participants and personnel to be blinded
Blinding of outcome assessment (detection bias) All outcomes	High risk	Comment: The outcome is subjective (a self‐reported outcome) and the participants are not blinded