Traeger 2015.

Study characteristics
Methods	Setting: Massachusetts, USA Recruitment: eligible participants were 18 years or older; had a diagnosis of stage I‐III breast, lung, or colorectal cancer and were scheduled to start chemotherapy; and were able to respond to questionnaires in English. Participants were recruited by a study co‐ordinator during their first chemotherapy visit. They were recruited consecutively until target enrolment was reached Randomisation: by a clinical trials office (1:1 allocation sequence) with stratification by tumour type
Participants	120 participants initiating chemotherapy for non‐metastatic disease (60 breast cancer; 30 colorectal cancer; 30 lung cancer)
Interventions	Intervention: provocative telephone‐based nursing guidance and support during first 2 chemotherapy cycles. Interventionists had a brief outline to structure calls but were encouraged to complete them according to clinical judgement to reflect participant‐centred practice. Two calls were provided during the first week after first chemotherapy administration, and 2 in the first week after the second cycle. Calls lasted around 15 minutes
Outcomes	Number of symptoms Symptom distress Depression Anxiety Methods for assessing outcomes: Memorial Symptom Assessment Scale‐Short Form (MSAS‐SF) Patient Health Questionnaire‐4 (PHQ‐4)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Dana Faber/Harvard Cancer Center Quality Assurance Officer for clinical trials randomly assigned participants
Allocation concealment (selection bias)	Unclear risk	For control participants, the nurse practitioner was not notified of study enrolment or group assignment
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details
Incomplete outcome data (attrition bias) All outcomes	Low risk	Study retention rate 97.5%
Selective reporting (reporting bias)	Low risk	Balanced reporting of all measures
Other bias	High risk	If the electronic health record showed a documented in‐person oncology visit during the intervention delivery window, the in‐person visit was counted as a completed call