Chernoff 2002.

Study characteristics
Methods	Parallel RCT. Comparing the effect for carers of children with a chronic illness of a comprehensive community‐based support programme with having a contact telephone number for an experienced mother of a child with a similar illness. For children it was hypothesised that those in the experimental group would have better mental health at the end of the intervention than controls. For the mothers it was hypothesised that those in the intervention group would report fewer psychological symptoms than control mothers.
Participants	Population of interest N = 193. Sample agreeing to participate N = 161 children with a chronic illness and their mothers. Children with diabetes, sickle cell, CF or moderate to severe asthma were included. For CF subgroup: n = 7 assigned to the intervention group and n = 6 controls. Included children aged 7 ‐ 11 years living within 80 km radius of Baltimore, with no learning disability, with a mother as a primary carer, on the telephone, speaking English, and diagnosed > 6 months ago.
Interventions	1. Community‐based support programme over 15 months ‐ trained network mother + child life specialist pairing. a. 7 home visits b. twice‐weekly telephone calls c. 3 special family events 2. Control group are given a contact number for an experienced mother.
Outcomes	Children a. personal adjustment and role skills b. depression c. anxiety d. self‐perception Mothers a. anxiety b. depression c. stressful life events
Notes	POI: anxiety subscale ‐ 11 items self‐report from 'Psychiatric Symptom Index'. The groups were equivalent at baseline although these data not reported for individual disease subgroups. Network mothers were trained and supervised and a manual was produced. Treatment fidelity: the authors reported that weekly meetings took place to ensure that intervention was provided as planned (Chernoff 2002, p.534/35).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The authors did not describe the randomisation process in detail. The stated that of the 161 families being eligible and who agreed to participate were 'randomly assigned to the experimental group and [...] to the control group (Chernoff 2002; p. 773).
Allocation concealment (selection bias)	Unclear risk	The method of concealment is not described.
Blinding (performance bias and detection bias) All outcomes	Low risk	Due to the nature of the intervention blinding of participants is not possible adequately, but the authors provided the information that 'interviews were completed by paid interviewers who had undergone extensive training and were blind to group assignment' (Chernoff 2002; p. 534).
Incomplete outcome data (attrition bias) All outcomes	High risk	The authors reported that about 15.5% of the randomised participants were not available for follow‐up analysis. 'of the 161 families randomised, 86 were randomised to the experimental group and 75 to the control group. Overall, 25 families, 14 and 11 from the experimental and the control group, respectively, were lost to follow up [...] (Chernoff 2002; p. 536).
Selective reporting (reporting bias)	High risk	The presentation of the results was selective. The subgroup analysis reported on p. 537 and figure 2 e.g. was not pre‐specified.