• Clinical epidemiology
• lung cancer
• thoracic surgery

We read with interest the article by Hubbard et al who raise late diagnosis as a key determinant of poor lung cancer survival in the UK.1 We would like to highlight discrepancies in rates of radical treatment use as another contributory factor. Figures from the Eurocare 4 study2 suggest 5-year survival figures in Northern Ireland have been higher than in England and Wales or Scotland (table 1).2–4 This could reflect differences in patient demographics, disease, method of recording or treatment modality. There are no published data that directly compare the health or the stage at presentation of patients with lung cancer in relation to geographical location in the UK. If significant regional differences exist, one might expect to find evidence for differential survival from other cancers as well. Current data do not suggest a survival advantage for cancer patients in general from Northern Ireland over the rest of the UK.

Data on cancer survival in Northern Ireland are collected by the Northern Ireland Cancer Registry (NICR), a population-based registry collecting data from pathological records, hospital discharges and death registrations. Over the last decade the introduction of LUCADA in England and Wales, and CaPPS in Northern Ireland have made it possible to look for regional differences in cancer treatment. In Northern Ireland the number of patients included in CaPPS exceeds that recorded in recent years by the cancer registry. This would suggest most patients have been captured. One-year survival is likely to be influenced by palliative treatments, but 5-year survival is largely influenced by radical treatments, of which surgery is the mainstay. The use of radical radiotherapy may also have some influence, but comparative data are not available. We looked at the surgical resection rate and 5-year relative survival rate for lung cancer in Northern Ireland and compared them with those published for England, Wales and Scotland.2–4

Reported rates of surgical resection in all areas of Northern Ireland in 2004–08 exceeded the average for England and Wales for 2008. We believe this may offer a plausible reason for better 5-year survival differences and may be worthy of further study. Regional differences in lung cancer treatment are not new and are as yet unexplained.5 Addressing the remediable differences in appropriate use of radical treatment in the UK may offer a more immediate potential improvement in lung cancer survival than early diagnosis.