ReferenceBritish Medical Journal 2003; 326: 1310
Authors and institutionsCarolyn Tarrant, research associate; Kate Windridge, research fellow; Richard Baker, professor of quality in health care, Department of General Practice and Primary Health Care, University of Leicester, UK. Mary Boulton, professor of sociology, School of Social Sciences and Law, Oxford Brookes University, UK. George Freeman, professor of general practice, Centre for Primary Care and Social Medicine, Imperial College London, UK
SettingPrimary care, Leicestershire, UK. Six (out of 12 approached) general practices
AimsTo explore patients' perceptions of the features of personal care and how far these are shared by healthcare providers; whether a continuing relationship between a health professional and a patient is essential for personal care; and the circumstances in which a continuing relationship is important
Research designSemi-structured interviews and focus groups to test the validity of initial interpretations
SamplingPractice drew a quota sample of patients (excluding patients deemed ‘inappropriate’) and sent an invitation letter. Recruitment continued ‘until sampling frame requirements were met for diversity in age, sex, ethnicity, frequency of attendance, and health status’. Practices varied in size (1 single-handed, 2 with 2–4 partners, 3 with ≥5 partners) and location (two inner city, three suburban or urban, and one rural). Semi-structured interviews: 40 patients aged ≥18 years. Mixture of socioeconomic and ethnic group characteristics. 13 GPs, 10 practice and community nurses, and six practice administrative staff (1–4 GPs, nurses, and receptionists per practice). Focus groups: three focus groups of 28 patients and four of health professionals (18 GPs, eight practice or community nurses, and eight administrative staff)
Data collection‘Narrative based approach in interviews, with a topic guide specifying open ended exploration of the meaning, value, and priority given to personal care, and of factors that facilitated or inhibited it, in the context of each responder's experience.’ Interviews lasted 30–90 minutes; all but two interviews were audiotaped and transcribed verbatim. (One GP and one patient requested note taking only). The same investigators conducted interviews and did analysis
ReflexivityMain researchers kept reflective diaries, ‘providing an audit trail relating the content and context of each interview to themes emerging during concurrent analysis’
Ethical issuesParticipants recruited via practices and gave consent. Ethical approval granted by local research ethics committee
Data analysis‘Framework’ analysis. Descriptive codes developed from independent repeated readings of transcripts, then identified emerging themes on the basis of initial indexing, hierarchical grouping of codes, and discussion of individual transcripts. Themes were validated by: discussion among all authors after independent reading of a sample of transcripts; focus groups (participants discussed statements relating to identified themes and were asked to give examples of any opposing beliefs); and inviting all the original interviewees to provide postal feedback on an interim report of the findings. Consequently, preliminary themes were revised and developed into thematic frameworks. Charts were drawn-up for each interviewee, summarising the meanings of personal care and the contexts within which it featured