Siggeirsdottir 2005.

Methods	Multi‐site randomised trial.
Participants	50 participants (26 women, 24 men; mean age 68 years, range 28 to 86) undergoing total hip replacement. Intervention group: 14 women, 13 men (mean age 69 years, range 52 to 81); control group: 12 women, 11 men (mean age 66 years, range 28 to 86). Inclusion criteria: participants living in their own home diagnosed with osteoarthritis of the hip, rheumatoid arthritis, primary segmental collapse of the femoral head or sequelae after developmental diseases or hip trauma. Exclusion criteria: primary hip fracture, metastatic tumours, dementia. Location: Reykjavik and Akranes, Iceland.
Interventions	Intervention group (n = 27) received a preoperative education and training programme 1 month before surgery given by a physiotherapist or occupational therapist (or both). Participants were informed about rehabilitation, became familiar with exercises and the devices to be used postoperatively. Participants also received an illustrated brochure on how to exercise postoperatively. Following discharge, participants received regular home visits from an outpatient team. Control group (n = 23) was treated according to clinical procedures already in use.
Outcomes	Function* (Oxford Hip Score (12‐60 points) higher score represented more dysfunction; Harris Hip Score (maximum 100 points) 90‐100: good function and excellent results, 80‐90: good, 70‐80: fair, < 70: poor results; and Nottingham Health Profile (maximum 100 points) higher score represented more dysfunction); length of hospital stay; complications. Participants were assessed preoperatively and 2, 4 and 6 months postoperatively.
Notes	*Only outcome data for the Oxford Hip Score was reported in the trial publication, so no outcome data for the Harris Hip Score and Nottingham Health Profile were included in the review. Oxford Hip Score data at 6 months included in Analysis 1.2. The original trial design was to carry out the trial at one hospital only. The decision to expand to a second site was made for financial reasons, and because of a high initial drop‐out. Treatment group was treated differently postoperatively, in that a physiotherapist or occupational therapist visited them at home after discharge. Only data up to discharge were used in the analyses. A cost analysis of the trial was published as a separate paper (Siggeirsdottir 2005).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Comment: sequence generation not described.
Allocation concealment (selection bias)	Low risk	Quote: "...they were randomized into one of the two groups (SG [study group] or CG [control group]) by opening a sealed envelope containing a note indicating which group the patient was to be allocated."
Blinding (performance bias and detection bias) Self‐reported outcomes	High risk	Comment: no description of any attempt to blind participants and blinding is unlikely to have been possible. Knowledge of being in the intervention group may have influenced participants' pain and function scores.
Blinding (performance bias and detection bias) Objective outcomes	Unclear risk	Comment: not described.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Comment: 2 control group participants missing preoperative pain and function scores. Impact of incomplete data likely to be minimal because outcomes are continuous measures.
Selective reporting (reporting bias)	High risk	Comment: outcome data were not fully reported for the Harris Hip Score or Nottingham Health Profile (trialists only reported P values for the differences between groups rather than means and SDs). Also, without the trial protocol, it is unclear whether any other outcomes were measured but not reported based on the results.
Other bias	Low risk	Comment: no other sources of bias identified.