Pedersen 2016b.

Study characteristics
Methods	RCT. Parallel design with 2 intervention groups and 1 control group. Duration: 8 weeks (randomization took place upon discharge from hospital and follow‐up interventions lasted 8 weeks). Location: single centre in Denmark.
Participants	Inclusion criteria: malnourished or at risk of malnutrition, 75 years and older, living independently and alone in the area served by the hospital, able to speak the Danish language, and to communicate over the telephone. Exclusion criteria: nursing home residents, terminal illnesses or cognitive impairment. Number randomised: 208 participants (intervention group 1, n = 73; intervention group 2, n = 68; control group, n = 67). Gender split: intervention group 1, 57% females; intervention group 2, 51% females; control group, 65% females. Age: mean 86.1 years. Nutritional status: mean (SD) MNA score, intervention group 1 17.1 (3.2); intervention group 2 17.3(3.7); control group 17.0 (3.9).
Interventions	Intervention group 1: participants received dietary advice plus ONS if required in the form of standard care during hospital stay followed by nutritional counselling* during home visits (45 min duration) by a clinical dietician at 1, 2 and 4 weeks post hospital discharge. Intervention group 2: participants received dietary advice plus ONS if required in the form of standard care during hospital stay followed by nutritional counselling via telephone consultations (15 min duration) by a clinical dietician at 1, 2 and 4 weeks post hospital discharge. Control group: participants received no dietary advice and no ONS in the form of standard care during hospital stay and no follow‐up care from the hospital after discharge. *based on nutritional needs identified during the hospital stay, and tailored to the individual’s preferences and circumstances; since reduced appetite and low food intake had become normal, the intervention focused on nutritional and meal behaviour that improve appetite and increase nutritional intake. The counselling sessions were attended by the participant’s home carer, who holds a key position in supporting the participant on a daily basis.
Outcomes	Primary outcome: change in ADL (Barthel‐100 score) at discharge and 8 weeks later. Secondary outcomes: change in physical performance (handgrip strength*, 30‐sec. chair stand test, CAS), QoL and depression measurements (SF‐36, Depression List, Geriatric Depression Score), and Avlund mobility‐tiredness score (Mob‐T).
Publication details	Language: English. Funding: not declared. Publication status: peer‐reviewed journal.
Notes	This study is a duplicate of Pederson 2016a. The study has 2 intervention arms and a single control group. Therefore the number of participants in the control group is divided by 2. This study ID describes the data of intervention group 'telephone counselling' versus control, whereas Pederson 2016a describes 'home visits' versus control. The primary aim of the study was to compare the effects of 2 nutritional follow‐up intervention strategies (home visit and telephone consultation) with no follow‐up, with regard to preventing short‐term deterioration in ADL, and the second, to compare the effect of the interventions on physical function, health‐related QoL, and emotional health. QoL data could not be used because only subscores were reported.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation executed electronically via the web‐based, clinical‐trial‐ support‐system, ‘TrialPartner’ (Public Health and Quality Improvement, Central Denmark Region). This central computer program used permuted block‐sizes and stratified the randomisation according to nutritional status.
Allocation concealment (selection bias)	Low risk	Randomisation took place upon discharge and was executed electronically via the web‐based, clinical‐trial‐ support‐system, ‘TrialPartner’ (Public Health and Quality Improvement, Central Denmark Region). This central computer program used permuted block‐sizes and stratified the randomisation according to nutritional status.
Blinding (performance bias and detection bias) Clinical outcomes	Unclear risk	Not measured.
Blinding (performance bias and detection bias) Functional outcomes	Low risk	Owing to the nature of the interventions, it was not possible to blind the participants to the intervention. The principal investigator obtained baseline characteristics before randomisation, and was not in contact with the participants after that. The research assistant who conducted the baseline and follow‐up measurements (week 0 and week 8) was not informed of the results of the randomisation, but it cannot be ruled out that the participants may have mentioned their assignment.
Blinding (performance bias and detection bias) Nutritional outcomes	Low risk	Owing to the nature of the interventions, it was not possible to blind the participants to the intervention. The principal investigator obtained baseline characteristics before randomisation, and was not in contact with the participants after that. The research assistant who conducted the baseline and follow‐up measurements (week 0 and week 8) was not informed of the results of the randomisation, but it cannot be ruled out that the participants may have mentioned their assignment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not blinded and likely that researchers and participants were aware of group allocation as this was a nutritional intervention. It is possible that assessment of some outcomes was influenced by lack of blinding.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The research assistant who conducted the baseline and follow‐up measurements (week 0 and week 8) was not informed of the results of the randomisation, but it cannot be ruled out that the participants may have mentioned their assignment.
Incomplete outcome data (attrition bias) All outcomes	Low risk	At 8 weeks after discharge, 157/208 (75%) completed the follow‐up (home visit n = 52, telephone consultation n = 51, and control group n = 54). Drop‐outs (25%) were equally divided across groups but the reasons for drop‐out not mentioned.
Selective reporting (reporting bias)	Low risk	The study protocol was published in 2015 (Journal of Ageing Reasearch & Clinical Practice) with primary outcome: ADL and secondary outcomes: physical performance (handgrip strength, chair stand), QoL, depression measurements, and tiredness score.
Other bias	Low risk	Baseline characteristics given and similar between groups