Carson 2011.

Study characteristics
Methods	Design: Randomised, unblinded, parallel 2‐group, multicentre, international trial  Setting: 47 clinical sites in the USA and Canada Recruitment: Juyl 2004 to February 2009  Maximum follow‐up: up to 5 years (mean of 3.1 years)
Participants	2016 participants aged 50 years or older who were undergoing surgical repair of a hip fracture, with Hb < 10.0 g/dL within 3 days after surgery, and who had clinical evidence of cardiovascular disease (original protocol, valid until December 2005) or cardiovascular risk factors (from December 2005 onwards, following change to expand criteria to enhance recruitment) Liberal group: n = 1007; M/F = M/F = 250/757; mean (SD) age = 81.8 (8.8) years Restrictive group: n = 1009; M/F = 239/770; mean (SD) age = 81.5 (9.0) years
Interventions	Liberal group: received packed RBCs when Hb dropped to < 10.0 g/dL Restrictive ('symptomatic strategy') group: received transfusion if they developed symptoms of anaemia or if Hb fell to < 8.0 g/dL
Outcomes	Primary outcomes: inability to walk 10 feet (or across a room) without human assistance, death prior to closure of the window for 60‐day mortality Other outcomes: Hb, ACS, in‐hospital myocardial infarction, unstable angina or death, disposition on discharge, survival, functional measures, fatigue/energy, re‐admission to hospital, pneumonia, wound infection, thromboembolism, stroke or transient ischaemic attack, cognition (Gruber‐Baldini), mortality at 30 days  Results for long‐term mortality were published after the main trial report (Carson 2015)
Notes	Trial title: FOCUS Trial registration: NCT00071032. Protocol available on journal website Trial funding: National Heart, Lung, and Blood Institute (USA) COI statement by investigators: apart from participating as investigators in the trial, the paper reported, "No other potential conflict of interest relevant to this article was reported" (Carson 2011 p 1). NCT record states all PIs employed by sponsoring organisation (Rutgers)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Data co‐ordinating centre staff prepared randomisation schedules for each site using randomly ordered block sizes of 2, 4, 6, or 8 (Carson 2011 p 2)
Allocation concealment (selection bias)	Low risk	The trial used an automated telephone randomisation system (Carson 2011 p 2)
Blinding of participants and personnel (performance bias) Objective outcomes	Low risk	Blinding of personnel for this intervention was not feasible, but in our view, for objective outcomes such as mortality (the primary outcome used within this review), we graded risk of bias as 'low'
Blinding of outcome assessment (detection bias) Objective measures	Low risk	"After randomization, clinical‐site staff members, clinicians, and patients were aware of study‐group assignments" (Carson 2011 p 2454) "The primary and secondary outcomes were assessed blinded to treatment assignment in hospital and later, during telephone follow‐up" "Nurses at the clinical coordinating center who were not involved with study implementation and were unaware of study‐group assignments telephoned patients or proxies at or close to 30 days and 60 days after randomization to ascertain outcomes after hospital discharge. They spoke directly to patients who were accessible by telephone or to proxies if patients were cognitively impaired or could not talk on the telephone" (Carson 2011 p 2455)
Blinding of outcome assessment (detection bias) Subjective measures	Low risk	Functional measures, fatigue, and myocardial infarction were assessed blinded to treatment assignment. Other in‐hospital morbidity (i.e. pneumonia) was assessed by clinical staff with knowledge of assignment   "Nurses at the clinical coordinating center who were not involved with study implementation and were unaware of study‐group assignments tele‐ phoned patients or proxies at or close to 30 days and 60 days after randomization to ascertain outcomes after hospital discharge. They spoke directly to patients who were accessible by telephone or to proxies if patients were cognitively impaired or could not talk on the telephone" (Carson 2011 p 2455)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Data for primary outcomes and most secondary outcomes were nearly complete: "there were 14 withdrawals, 2 losses to follow‐up, and 1 incomplete follow‐up ascertainment; follow‐up for the primary analysis was obtained in 99.2% of the patients. Of the 1999 patients included in the primary analysis, we directly interviewed 1075 (53.8%) and obtained data on 923 (46.2%) by proxy; the source of information was missing for 1 patient" (Carson 2011 p 5)
Selective reporting (reporting bias)	Low risk	Reporting was complete (ascertainable by comparison of publications with trial registration and protocol)
Other bias	Low risk	No other biases identified. It is noted that there was a change to inclusion criteria and a reduction in the recruitment target. In the follow‐up trial of long‐term outcomes for mortality, 10% of cases could not be linked to death registries