Table 2 Overview of the findings from the feasibility assessment of the trial procedures

Retention

This study had a good retention rate (84%).

_

Recruitment

Staff at the clinic were co-operative and allowed researchers to screen notes in their busy working atmosphere and to use their system for printing labels (names and addresses of the potential participants) which helped to speed up the process.

The process could not meet the target of recruiting 30 women per arm in 4 month time scale within the resources available.

Solution: More time would be required to achieve target number of participants if the same procedure were to be used in a future trial. Reminders could be sent to the non-responders.

The actual response rate of eligible women to the invitation letter is unknown as information on BMI was not available in the clinic notes.

Solution: Weight measurements made at out-patient clinics should be entered in clinic notes.

Randomisation

The overall group allocation process worked well.

Median years since diagnosis reported was 2.0 years but the allocation process used < 1 vs > 1 year since diagnosis.

Solution: A future trial should use median time since diagnosis (e.g. 2 years) for group allocation.

Unintended consequences

No serious adverse events were recorded.

Two participants were distressed and one withdrew due to the setting and topics discussed.

Solution: Venue needs to be carefully chosen and topics related to sensitive issues should be avoided.

Setting

Maggie’s centre was feasible for conducting baseline and end-point meetings and convenient for health professionals to deliver the dietetic led sessions.

There were problems with using slides in the dietitian-led group sessions as the room had no blinds or projector screen. There was a lack of parking space for the participants.

Solution: Issues related to room setting could be addressed with portable equipment. Another non-hospital venue such as the CLAN cancer support centre or a community centre could be used in the future to avoid parking issues.

Data collection tools

The height measure and weighing scales were easy to use.

The FACT-B QoL questionnaire did not take long to complete by participants.

The weighing scale could not measure weight above 150 kg.

Solution: Scales with higher weighing capacity could be used.

Some questions of the QoL questionnaire were felt by participants not to be applicable after a few years of finishing breast cancer treatments.

Solution: FACT-B could be used with more clear instructions to avoid missing data or a different QoL questionnaire could be used.

Fidelity/delivery of the dietetic-led sessions

All facilitators adhered to the study protocol.

Workload and time were recorded as issues for delivering dietitian led group sessions.

Solutions: Non-health professionals or volunteers could be trained to deliver the group sessions.

It was not possible to observe WW sessions to report their fidelity.

Solution: Random sessions could be observed or recorded to report fidelity.

Also see issues discussed above under setting.

Meeting attendance

Good attendance rates at dietetic led sessions (85%) and WW programme (WW Plus = 78% and WW group = 68%).

Attendance data at WW programme was provided by participants.

Solution: Participants’ booklet provided by WW could be assessed for accurate recording of attendance.

Cost

WW group was not expensive to run.

WW Plus group was expensive to run compared to WW group.

Solution: Non-health professionals or volunteers can be trained to deliver the group sessions, which can be tested before inclusion. But then there might not be as good adherence and fidelity of delivery as observed when ran by a dietitian.