Background
Older adults receive most of their end-of-life care in the community, but there are few published data to guide researchers on recruitment to studies in primary care. The aim of this study was to compare recruitment of patients and bereaved carers from general practices in areas with different research network support, and identify challenges in obtaining samples representative of those in need of end-of-life care.

Methods
Comparative analysis of recruitment from general practices to two face-to-face interview studies concerned with 1) carers’ perceptions of transitions between settings for decedents aged over 75 years and 2) the experiences of older patients living with cancer at the end-of-life.

Results
33 (15% of invitees) patients and 118 (25%) carers were interviewed. Carers from disadvantaged areas were under-represented. Recruitment was higher when researchers, rather than research network staff, were in direct contact with general practices. Most practices recruited no more than one carer, despite a seven fold difference in the number of registered patients. The proportion identified as eligible for patient interviews varied by a factor of 38 between practices. Forty-four Primary Care Trusts granted approval to interview carers; two refused. One gave no reason; a second did not believe that general practitioners would be able to identify carers.

Conclusion
Obtaining a representative sample of patients or carers in end-of-life research is a resource intensive challenge. Review of the regulatory and organisational barriers to end-of-life researchers in primary care is required. Research support networks provide invaluable assistance, but researchers should ensure that they are alert to the ways in which they may influence study recruitment.