Variations in GP decision making in the diagnosis of lung cancer

• Main Author: Sequeira, R. E.
• Published: University College London (University of London) 2016
• Subjects: 610
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.746357

Background: The United Kingdom's lung cancer patients have lower survival than patients in comparable countries. Delays in diagnosis may contribute to this. There are significant socio-demographic variations in the interval between cancer patients first presenting to their general practitioner (GP) and referral, but it is unclear why these exist. Aim: To examine patient and GP characteristics associated with GPs' referral decisions, focusing on patients with symptoms indicative of lung cancer. Methods: Study 1: Systematic literature review considering non-clinical patient, GP and practice characteristics associated with variations in GPs' referral of patients for investigations or to secondary care. Study 2: GP decision making study: a factorial experiment using interactive multimedia vignettes to examine GPs' decisions to refer patients with symptoms indicative of lung cancer, and a survey to examine factors influencing decision making. Results: Study 1: 11,791 titles were screened; 47 were of sufficient quality and relevance for inclusion. There was strong evidence that patients over 75 were less likely to be investigated or referred, and of variations by patient gender. However few higher quality studies examined associations with patient ethnicity and GP or practice characteristics, or considered why socio-demographic variations occurred. Study 2: 227 GPs completed the study. GPs were less likely to investigate older than younger patients, and black patients than white. The survey identified several factors that GPs believe affect their referral decisions (such as patients' lifestyles), some of which may explain the observed differences in GPs' referral decisions. Conclusions: My thesis identified socio-demographic variations in GP decision making that are independent of clinical characteristics (for lung cancer and more widely) and factors that may underlie these. Further research addressing the extent to which these factors contribute to socio-demographic variations, and the development of primary care interventions which address these findings, could reduce delays in lung cancer diagnosis.