Sir, we read with interest the article by McLeod et al.1 on delays in the diagnosis of oral cancer.

An assumption of the paper is that reducing diagnostic delays will reduce tumour stage at diagnosis. The current literature indicates there is no significant association between duration of diagnostic delay and stage of oral cancer at diagnosis, suggesting advanced stage disease is not always a consequence of delayed diagnosis (delay being defined as the period from the onset of symptoms to receipt of a definitive diagnosis).

We have investigated this relationship in a consecutive cohort of patients (n = 250) presenting to Guy's Hospital Head and Neck Service, with squamous cell carcinoma of the oral cavity.2 Although just over half the cohort followed the logical delay-stage relationship (29% had no delay and early stage disease, while 24% had a long delay and had advanced stage disease), 27% paradoxically had no delay yet had advanced stage disease and 20% had a prolonged diagnostic delay yet had early stage disease. It is suggested the paradox is due to different rates of tumour growth.3 However the time durations involved are simply not long enough for a squamous cell carcinoma to develop from an early lesion into late stage disease.

A more plausible explanation (but one which has received relatively little attention) is that some oral cancers may be silent (asymptomatic) until late in the disease period.3 Conversely, alert individuals may identify the visual signs or symptoms of oral cancer when it is ore-neoplastic or early stage disease. As the pre-neoplastic phase can extend over a long period of time,4 prolonged diagnostic delay can accompany early stage disease at diagnosis.

Given that for 27% of patients, advanced stage disease is not a consequence of delayed diagnosis, it is important that factors predisposing to advanced stage are identified. Our data indicate that being female, white and married is predictive of early stage disease, while being non-white, male and single is predictive of advanced stage disease. Thus, public education might best be targeted towards both 'at risk' individuals (ie heavy smokers and those who have a high intake of alcohol) and those more likely to present with advanced stage disease — those who are male, without partners and non-white. Such interventions should be aimed at self-examination and regular dental attendance as a form of screening for early oral lesions.

Patients were considered to have experienced prolonged diagnostic delay when diagnostic delay exceeded three months (median value).