Crowe 2003.

Methods	Randomised parallel group trial.
Participants	133 participants undergoing total hip replacement (65 participants) or total knee replacement (68 participants). Intervention group: 51 women, 14 men (mean age 67 years, SD 12); control group: 55 women, 13 men (mean age 71 years, SD 11). Inclusion criteria: participants who were not functioning well; had limited social support or co‐existing medical conditions, or both. Exclusion criteria: participants who were functioning well despite their joint dysfunction, who were managing activities of daily living and had good carer support. General exclusions included people with limited English and who were undergoing a revision or second joint replacement within 2 years. Location: Ontario, Canada.
Interventions	Intervention group (n = 65) received a preoperative education package (video, booklet plus information on length of stay, discharge criteria, respite care and diet). Some participants given tour of the hospital unit, demonstration of equipment, dietician counselling and social work input. All received individualised counselling from an occupational therapist. Control group (n = 68) received 1 standard preoperative clinic visit (lasting about 7 hours) 1‐2 weeks before surgery. Participants were informed about the hospital stay and the immediate postoperative phase.
Outcomes	Days to eligibility for discharge; preoperative anxiety (Spielberger State‐Trait Anxiety Inventory (20‐80 points) higher score represented greater anxiety); length of hospital stay; days to mobilisation (out of bed, walking and climbing stairs).
Notes	Randomisation resulted in uneven numbers of hip and knee replacements in each group. Data for hip and knee replacement groups were obtained from the study author.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Subjects were allocated to one of the two groups by means of a random number table and using a system of sealed envelopes."
Allocation concealment (selection bias)	Low risk	Quote: "Subjects were allocated to one of the two groups by means of a random number table and using a system of sealed envelopes."
Blinding (performance bias and detection bias) Self‐reported outcomes	High risk	Quote:"[Participants] were telephoned by the research coordinator who explained the trial. Oral assent was obtained, and the client signed an informed consent at his/her initial trial visit." Comment: participants were unlikely to be blinded and this may impact self reported outcomes like anxiety. It was not possible to blind the in‐hospital PT and OT staff as participants often choose to discuss previous rehabilitation with staff.
Blinding (performance bias and detection bias) Objective outcomes	Low risk	Quote: "Outcomes were measured in‐hospital by an investigator who was blinded to group allocation."
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Comment: incomplete outcome data not reported. It is unclear if there were missing data and if so, how this was handled. Results tables report continuous measures but omit the denominator.
Selective reporting (reporting bias)	Unclear risk	Comment: all outcomes listed in the methods were reported in the results, but without the trial protocol, it is unclear whether any other outcomes were measured but not reported based on the results.
Other bias	High risk	The control group had significantly poorer overall function prior to surgery (i.e. a pre‐existing difference that favoured the treatment group).