Background During the pilot of an RCT of provider, contract and patient referral, for a 66 practice RCT of partner notification in UK primary care, it emerged that there is uncertainty about real life clinical practice. Our objectives are to describe how health advisers negotiate provider, contract and patient partner notification in clinical practice. To determine the feasibility of comparing both provider and contract referral separately against patient referral.

Methods We recruited 10 health advisers from contrasting settings to attend a 1-day workshop led by partner notification specialists and researchers. They participated in focus group discussions, observations of practice, role play with each other and with actors. They and the actors then contributed to further focus groups reflecting on their observations and experiences, and advising on feasible strategies for delivering standard partner notification interventions in the RCT context. All discussions were recorded for qualitative analysis.

Results All health advisers practised provider referral, but the extent was variable. Contract referral as defined in the UK (a time period is explicitly agreed for the patient to notify, after which provider notification is initiated) is practised for HIV, other bloodborne viruses and sometimes syphilis. However for common bacterial STIs, a patient's choice of provider referral often emerges over multiple consultations, and provider referral is not a distinct intervention from contract referral. Health advisers saw their role as helping patients find solutions to partner notification, rather than applying specific interventions.

Conclusions Provider and contract referral are not sufficiently distinct to be compared with each other in a trial setting for chlamydia and gonorrhoea. However contract referral does have an important role in management of partner notification for bloodborne viruses.