Bennell 2018.

Study characteristics
Methods	RCT; 2 arms; assessed pretreatment, post‐treatment, and 12 months FU. Only post‐treatment data were extracted as there was > 30% interaction (face to face) with the health professional and participants and physiotherapy input from post‐treatment to 52 weeks.
Participants	Start of treatment: N = 144 Post‐treatment: N = 137 Sex: 82 F, 62 M Mean age: 61.25 years (SD 7.15) Diagnosis: hip osteoarthritis Mean years of pain: reported as the number and percentage of participants falling into three symptom duration categories: < 2 years (n = 64), 2 to 10 years (n = 66), > 10 years (n = 18). We note that there is discrepancy within the article report between the overall sample size and this breakdown of symptom duration (with the latter, apparently erroneously, equalling 148).
Interventions	Intervention name: pain coping skills training Psychological approach: CBT Duration: 8 weeks, 1 module per week, 35‐ to 45‐minute module, then 16‐week exercise programme Control type: active
Outcomes	Primary pain outcome: pain intensity (NRS) Primary disability outcome: WOMAC – Physical Function subscale Primary quality‐of‐life outcome: AQOL Adverse events: number of people experiencing adverse events Primary anxiety outcome: DASS – Anxiety subscale Primary depression outcome: DASS – Depression subscale Intervention satisfaction: none Intervention engagement: adherence, number of education information sheets accessed, number of physiotherapy sessions attended, and home exercise adherence Other outcome measures: self‐efficacy for pain and function (Arthritis Self‐Efficacy Scale), coping skills (Coping Attempts Scale of the Coping Strategies Questionnaire), Pain Catastrophizing Scale, and physical activity (Physical Activity Scale for the Elderly), WOMAC Pain subscale
Notes	Funding: National Health and Medical Research Council Programme Grant Conflicts of interest: KLB received grants from National Health and Medical Research Council during the conduct of the study; personal fees from Physitrack, ASICS Oceania, Peking University, and Brigham and Women’s Hospital outside submitted work; and other support from ASICS Oceania outside the submitted work. RKN received grants from National Health and Medical Research Council during conduct of study. FD received grants from National Health and Medical Research Council during conduct of study and personal fees from Elsevier Oracle outside submitted work. PWH received grants from the National Health and Medical Research Council during the conduct of the study. RSH received grants from National Health and Medical Research Council and Australian Research Council Future Fellowship during conduct of study; grant from Medibank Better Health Foundation; and other support from ASICS Oceania outside submitted work. Remaining authors have no conflicts of interest. Country: Germany Trial registration: Australian and New Zealand Clinical Trials Registry 12614000230651 Intervention development, intervention delivery, and percentage interaction of healthcare professional obtained and confirmed by study authors upon contact
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	An independent researcher prepared a computer‐generated randomisation schedule (randomised permuted blocks of varying sizes) stratified by physiotherapist and sex.
Allocation concealment (selection bias)	Low risk	The schedule was concealed using opaque, sealed envelopes.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Participants, who were themselves blinded, self‐reported outcome assessments. The statistician was blinded to group allocation.
Incomplete outcome data (attrition bias) All outcomes	Low risk	< 10% of participants dropped out during the whole study.
Selective reporting (reporting bias)	Low risk	Registry details are available; all outcomes were reported as planned.