Table 4.

Response rates and reasons for non-completion of ePRO questionnaire

	Baselinea	2–3 days	5–7 days	Week 2	Week 3	Week 4	Week 5	Week 6	Week 7	Week 8
Active participantsb, n	29	29	29	27	25	24	24	23	22	22
Active participants completing ePRO questionnaire, n (%)c	27 (93)	16 (55)	23 (79)	20 (74)	21 (84)	21 (88)	22 (92)	19 (83)	16 (72)	17 (77)
Active participants not completing ePRO questionnaire, n	2	13	6	7	4	3	2	4	6	5
Reasons for non-completion:											Totals
Withdrawn from the study	0	0	2 (33)	2 (29)	1 (4)	0	1 (50)	1 (25)	0	0	7
Unknown - participant could not be reached for weekly phone interview	0	7 (54)	4 (67)	2 (29)	0	2 (67)	1 (50)	2 (50)	2 (33)	1 (20)	21
Did not want to	1 (50)	1 (8)	0	1 (14)	1 (4)	0	0	0	0	1 (20)	5
Started chemotherapyd	0	0	0	0	0	0	0	1 (25)	2 (33)	3 (60)	6
Admin failure (e.g. overlap of dates/ University closure)	1 (50)	1 (8)	0	0	1 (4)	1 (33)	0	0	1 (17)	0	5
Re-admitted to hospital	0	1 (8)	0	2 (29)	0	0	0	0	0	0	3
Too busy	0	1 (8)	0	0	1 (4)	0	0	0	1 (17)	0	3
Too unwell	0	2 (15)	0	0	0	0	0	0	0	0	2
Total number of ePRO questionnaire completions	19e	18	23	21f	21	21	22	19	16	17	197

^aBaseline completion takes place at the point of hospital discharge. All subsequent timepoints are length of time since hospital discharge

^b Participants who had not withdrawn from the study

^cThe design of the ePRO system ensures all items are completed, except for completions that were abandoned prior to submission

^d Patients halted completion of ePRO if they commenced chemotherapy during follow-up

^e 8 baseline completions triggering a Level 3 action were removed from the dataset post-hoc because they were later determined to be clinically irrelevant by participants and clinicians (as they were retrospectively reporting symptoms experienced immediately post-surgery that had resolved). This will inform the next iteration of algorithms to be using in a full RCT

^f The ePRO system allows multiple completions at each timepoint