Forget 2014.

Study characteristics
Methods	Study design: randomised controlled trial, parallel, single centre Study dates: not reported Country: not reported
Participants	Total participants: 54 Inclusion criteria: people with performance status 0, 1 or 2 on chemotherapy for various cancer; have cachexia (weight loss ≥ 5% bodyweight within prior 6 months) or at high risk for cachexia (radiochemotherapy planned on upper part of digestive tract) Exclusion criteria: not reported Participants characteristics: not reported
Interventions	Experimental group: participants received mirtazapine 30 mg/day, weekly advice from a dietician and physical exercise. Psychological support available if needed Type of exercise: not reported Duration of exercise programme: 3 months Intensity of exercise: not reported Frequency of exercise (day/week): daily (no other information) Volume of exercise (minutes/day): not reported Supervision: daily alone and weekly with a physiotherapist Setting: not reported Comparator group: best supportive care (no other information) Concomitant interventions: unclear Follow‐up: 3 months
Outcomes	Primary outcomes Body mass index How measured: not reported Time points measured: baseline, 3 months Secondary outcomes Hand grip strength How measured: not reported Time points measured: baseline, 3 months Mid‐upper muscle circumference How measured: not reported Time points measured: baseline, 3 months Quality of life How measured: QLQc‐30 Time points measured: baseline, 3 months PG‐SGA How measured: not reported Time points measured: baseline, 3 months
Notes	Conference abstract It was not possible to extract any data since authors presented only mean values with no measure of variability. No funding or conflict of interest were stated in the conference abstract.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not reported.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	High risk	23/54 participants were included in final analysis; 9 participants died.
Selective reporting (reporting bias)	Unclear risk	The data provided in the table regarded only mean values, authors did not provide confidence interval, standard deviation or standard errors.
Other bias	Unclear risk	Lack of information to make a judgement.