Recruitment of study participants was challenging. The majority of women approached declined to participate. Anecdotally, this was predominantly because of a concern about taking medication during pregnancy. However, it was also evident that there was a lack of awareness of the problems associated with obesity during pregnancy, particularly potential longer-term harms for the offspring. We held education sessions for study midwives to enable them to pass on this message to their patients and the wider midwifery community. We distributed posters advertising the study to all clinical areas where potentially eligible women might see them. Following feedback from the public and potential patients that they found the word ‘obese’ offensive, we revised all of our materials and used the more acceptable terminology ‘high BMI’. We employed the use of social media and a website to disseminate information to the public and participants about the progress of the trial, for example updates when recruitment targets had been reached. We included a qualitative substudy to explore the reasons behind patients’ reluctance to participate in the trial and why participants withdrew from the trial once they had been recruited.

Patients and the public were not formally involved in the trial at the design and concept stage. However, the clinical investigators (including the lead investigator) were strongly influenced by the pregnant women (patients) they look after and these influences were key to the trial.

Patient/public involvement was formally instituted during the conduct of the trial. A patient/public representative was appointed to the Trial Steering Committee. She made helpful comments on the patient information leaflet and attended several meetings of the Trial Steering Committee. She resigned in the last year of the study and, given that the trial was ending, she was not formally replaced.