Gregersen 2015.

Study characteristics
Methods	Design: RCT, parallel 2‐arm, single‐site trial Setting: orthopaedic ward in a university teaching hospital, Denmark Recruitment: January 2010 to June 2013 Maximum follow‐up: 12 months
Participants	284 participants aged 65 years or older undergoing hip fracture surgery who had postoperative Hb between 9.7 g/dL and 11.3 g/dL during the first 6 days postoperatively Liberal group: n = 140; mean age (SD) = 88 (6.9) years Restrictive group: n = 144; mean age (SD) = 86 (6.8) years
Interventions	Liberal group: received transfusion when Hb was < 11.3 g/dL Restrictive group: received transfusion when Hb was < 9.7 g/dL
Outcomes	Primary outcomes: recovery from physical disabilities (3 tools were used to measure physical performance: the Modified Barthel index, New Mobility score, and Cumulated Ambulation Score); total number of infections (pneumonia, urinary tract infection, other); cognition; delirium (10 days); depression; quality of life; modified Barthel Index; comprehensive frailty index Death within 30 days provided (together with causes, e.g. sepsis, cardiovascular disease, pneumonia)
Notes	Four publications reported results Trial title: Transfusion Requirements In Frail Elderly (TRIFE) Trial registration: NCT01102010 Trial funding: "we have received grants from the Helga and Peter Korning Foundation for medical equipment (HemoCue portable photometer). Costs of data collection, analyses, and article writing were borne by the Department of Geriatrics at Aarhus University Hospital" (Gregersen 2015 p 7) COI statement by investigators: "the authors declare that there is no conflict of interest" (Gregersen 2015 p 7)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The randomization procedure was provided by an allocation concealment process and conducted electronically in the web‐based clinical‐trial‐support‐system ‘TrialPartner’ from Public Health and Quality Improvement in Central Denmark Region. This central computer program using permuted block‐sizes stratified the randomization according to gender and type of residence. Results of randomization were available at the electronic patient record for the hospital staff in the Orthopedic and Geriatric wards since the staff should administer the transfusions" (Gregersen 2015 p 2)
Allocation concealment (selection bias)	Low risk	See above
Blinding of participants and personnel (performance bias) Objective outcomes	Low risk	Blinding of personnel for this intervention is 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	Blinding of mortality (the primary outcome used within this review) is not relevant, and we graded risk of bias as 'low'
Blinding of outcome assessment (detection bias) Subjective measures	Low risk	"The participants, their relatives, and the endpoint assessors were blinded to the result of randomization" (Gregersen 2015 p 2)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Of the 284 participants, there were 8 dropouts (4 from each group ‐ reasons cited as refusing transfusion and 1 acute bleeding ulcer) and 8 further protocol deviations per transfusion group
Selective reporting (reporting bias)	Low risk	Trial was prospectively registered (NCT01102010). No reporting bias was apparent
Other bias	Low risk	No other biases were apparent