Sullivan 2004.

Methods	Randomised clinical trial, USA
Participants	57 hospitalised adults older than 64 who underwent surgical repair of an acute hip fracutre, at nutritonal risk due to being frail elderly Male:Female = 39:18 Mean age = 78.8 years Exclusion criteria: incapable of giving informed consent and did not have a legal guardian; pathological fracture (due to cancer or other non‐osteoporotic pathologies), trauma to other organ systems (e.g. multi‐trauma from a motor vehicle accident); metastatic cancer, cirrhosis of the liver, a contraindication to the use of enteral feedings (e.g. severe short‐bowel syndrome), or organ failure which rendered the proposed intervention inappropriate
Interventions	Experimental group: The participants' ‘nutrient deficit’ for the day (‘target intake’ minus ‘volitional intake’) was calculated each evening. Nightly enteral feedings were initiated with a nutritionally complete, lactose‐free, polymeric enteral formula (Pro‐mote®, Ross Laboratories) that contained 1000 Kcal (4187kJ), 62.5 g protein (25% of calories), 26 g fat (23% of calories), and 130 grams carbohydrates (52% of calories) per litre. On the 1st night after the feeding tube was placed, the participant was provided enteral feedings at a rate of 50 cc/hr over an 11‐hr period beginning at 7 p.m. (i.e. a total of 550 cc of enteral formula, 34.5 g protein). If the participant tolerated the tube‐feedings, the rate was increased by 25 cc/hr each night to either: (a) a maximum of 125 cc/hr over an 11‐hr period beginning at 7 p.m.; or (b) the ‘nutrient deficit’ was reached. For example, if the participants' ‘target intake’ was calculated to be 2100 Kcal and his ‘volitional intake’ was 1400 Kcal, the enteral feeding rate that night was set to 64 cc/hr for a total of 700 cc over 11 hrs, which equalled his ‘nutrient deficit’. (n = 27) Control group: No intervention (n = 30) Co‐interventions: standard postoperative care
Outcomes	Complications, life‐threatening complications, discharge data, mortality, length of stay, MMSE, ADL, albumin, pre‐albumin, cholesterol
Study dates	Not stated
Notes	Notes taken from Avanell 2010. We contacted the authors on 8th February 2016 by email: sullivandennish@uams.edu. We received no reply.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The trial was described as being randomised, but it was unclear how the sequence was generated.
Allocation concealment (selection bias)	Low risk	"The randomisation process was prepared by the biostatistician, using a series of sealed envelopes. Security (lined) envelopes were used to assure that the assignment could not be read without opening the envelope. After consent had been obtained and the baseline assessment was completed, the next envelope in order was opened to reveal the group assignment. Each envelope contained a card. The card had the assignment for treatment or control pre‐printed. Space was provided to enter the patient name and ID as well as the date, time and person responsible for randomization. The study nurse completed the card, photocopied it, and returned the original to the biostatistician as a check that the randomization process was progressing appropriately. Subjects were randomized to either treatment or control within blocks to assure that there were roughly equal numbers of subjects in each group at the end of the study. The block sizes were randomly varied to minimize the ability to deduce the assignment for a particular patient before opening the envelope" Quote taken from (Avenell 2016).
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not described
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not described
Incomplete outcome data (attrition bias) All outcomes	Low risk	There were no dropouts.
Selective reporting (reporting bias)	Low risk	The trial reported mortality and serious adverse events.
For‐profit bias	High risk	The trial was funded by Ross Laboratories: "We also wish to express our appreciation to Ross Laboratories for supplying the nutritional supplements and the nasogastric feeding tubes".
Other bias	Low risk	The trial appeared to be free of other components that could put it at risk of bias.