Influence of doctor-patient conversations on behaviours of patients presenting to primary care with new or persistent symptoms: a video observation study.

Abstract

Background: Most cancers are diagnosed following contact with primary care. Patients diagnosed with cancer often see their doctor multiple times with potentially relevant symptoms before being referred to see a specialist, suggesting missed opportunities during doctor-patient conversations.

Objective: To understand doctor-patient communication around the significance of persistent or new presenting problems and its potential impact on timely cancer diagnosis.

Research design: Qualitative thematic analysis based on video recordings of doctor-patient consultations in primary care and follow-up interviews with patients and doctors. 80 video observations, 20 patient interviews and 7 doctor interviews across 7 general practices in England.

Results: We found that timeliness of diagnosis may be adversely affected if doctors and patients do not come to an agreement about the presenting problem's significance. 'Disagreements' may involve misaligned cognitive factors such as differences in medical knowledge between doctor and patient or misaligned emotional factors such as patients' unexpressed fear of diagnostic procedures. Interviews suggested that conversations where the difference in views is either not recognised or stays unresolved may lead to unhelpful patient behaviour after the consultation (eg, non-attendance at specialist appointments), creating potential for diagnostic delay and patient harm.

Conclusions: Our findings highlight how doctor-patient consultations can impact timely diagnosis when patients present with persistent or new problems. Misalignments were common and could go unnoticed, leaving gaps for potential to cause patient harm. These findings have implications for timely diagnosis of cancer and other serious disease because they highlight the complexity and fluidity of the consultation and the subsequent impact on the diagnostic process.