Diagnostic delays in head and neck cancers.

Abstract

Head and neck cancers are a significant and worsening health problem in the UK. In the absence of screening, minimising diagnostic delay after the onset of symptoms improves prognosis. Delay, from the patient's initial experience of symptoms to the ultimate diagnosis, consists of two elements--the delay prior to presenting to a clinician plus that due to the health professional consulted. This study aimed to establish the period of delay between recognition of the initial tumour symptoms and the formal diagnosis among a sample of patients recently diagnosed with head and neck cancer. Using a semistructured questionnaire, 133 men and 55 women were interviewed by a research nurse, and the results were related to the clinical findings. Tumour size at diagnosis was classified according to T1 (22%), T2 (29%), T3 (27%) and T4 (22%). Of the 186 patients with complete hospital records, 48 (26%) were diagnosed with cancer of the lip and oral cavity (CLOC). From the onset of symptoms to the patients' initial decision to seek professional advice, the median period was 4 weeks among those with CLOC and 3 weeks for those with other head and neck cancers (OHNC). The distribution was highly skewed with delays beyond 6 months occurring among 9% of the OHNC group, compared with 3% of CLOC. From the onset of symptoms to a consultant appointment, the median delay was 8 weeks for OHNC, but 12 weeks for CLOC, with delays beyond 6 months of 13% in each group, respectively. First symptoms included 'change in voice' (26%), 'pain' (27%), 'lump' or 'growth' (12%) as well as dysphagia, 'infection', 'sore throat', 'ulcers' or 'abscess'. No significant association was found between the nature of the first symptoms and the urgency with which patients interpreted their symptoms, nor was this related to diagnostic delay, sex, age or social class. It is concluded that there is substantial variation in time to clinical presentation, particularly for OHNC, although professional delay for the majority of these cases was minimal. For patients with CLOC there was less variation in patient delay, but clinician delay was relatively longer.