Recognition of cancer warning signs and anticipated time to help-seeking in a population sample of adults in the UK


Session type:

Samantha L Quaife1, Lindsay JL Forbes2, Amanda J Ramirez2, Kate E Brain3, Conan Donnelly4, Alice E Simon5, Jane Wardle1
1Health Behaviour Research Centre, University College London, London, UK, 2Promoting Early Presentation Group, King's College London, London, UK, 3Cochrane Institute of Primary Care and Public Health, Cardiff University, Cardiff, UK, 4Centre for Public Health, Queen's University Belfast, Belfast, UK, 5City University, London, UK


Lack of recognition of cancer warning signs is implicated in delay in help-seeking in retrospective reports from cancer patients. In one population-based study, an aggregated recognition score was associated with anticipating a longer average help-seeking delay for potential cancer symptoms. The present analysis tested the hypothesis that anticipated delay for individual symptoms would be longer in those who did not recognise the relevant warning sign.


A population-based sample of 6965 UK adults (age >50 years) completed telephone interviews using the validated Awareness and Beliefs about Cancer scale,1 which assesses anticipated help-seeking for potential cancer symptoms, recognition of warning signs, demographics, and healthcare access. Symptoms used in this analysis were breast changes, rectal bleeding and persistent cough; asking how long the respondent would wait before making a doctor's appointment. We used >2 weeks as the cut-off to match the previous study,2 but also did sensitivity analyses with alternative durations. The related warning signs were unexplained lump, unexplained bleeding and persistent cough; asking whether each one could be a warning sign for cancer (yes/no).


For each symptom, more respondents who did not recognise the related warning sign anticipated >2 weeks delay: 17% vs 8% for breast changes, 11% vs 7% for rectal bleeding, 53% vs 49% for persistent cough. Multivariable logistic regression showed significant effects of recognition for delay for breast changes (OR=2.45, 95% CI 1.47-4.08), rectal bleeding (OR=1.77, 1.36-2.30), and persistent cough (OR=1.30, 1.17-1.46), independent of age, sex, education, ethnicity, and healthcare access. Effects were the same using a 4 week cut-off.


Lack of recognition that a symptom could be a warning sign for cancer was associated with anticipating greater delay in help-seeking. Future studies should use behavioural rather than anticipated outcomes to estimate the contribution of awareness of warning signs to patient delay in diagnosis.