Lung cancer - the truth behind the headlines
A new study suggests that doctors are missing opportunities to diagnose lung cancer. Perhaps not surprisingly, the media reporting focused on blaming GPs for not spotting the signs and making the diagnosis early enough. But, as is so often the case, there may be more to the study than meets the eye.
Everyone, doctor or otherwise, knows that smoking is the biggest risk factor for lung cancer - although in about one in five people diagnosed with this form of cancer there's another cause, cancer is to blame in the other four out of five. Heavy smoking is an even bigger risk - people smoking one to 14 cigarettes a day are eight times more likely than non-smokers to get lung cancer, while in smokers of more than 25 a day, the figure is a 25 times higher risk. So if I see a smoker (or an ex-smoker) with symptoms that might suggest lung cancer, I send them for an urgent chest X-ray, as recommended by NICE. Thousands of other doctors do the same. But this is where the criticism of GPs starts to fall down - because according to the study, the doctors who referred the most patients for chest X-rays were most likely to have patients dying early after diagnosis.
The NICE guideline goes further - it says that even if the chest X-ray is normal, patients should be referred urgently to a lung cancer specialist if there is a 'high suspicion of lung cancer'. What does the guidance give as symptoms that might suggest lung cancer? A persistent (more than three weeks) or unexplained history of cough, shortness of breath, abnormal signs on examining the chest, chest or shoulder pain, weight loss, hoarseness or changes in the shape of the end of the fingers, called 'clubbing'. But most lifelong smokers have long-standing cough and shortness of breath, because most of them have COPD because of their smoking. One of the problems with lung cancer is that in the early stages there are usually no symptoms - or at least, none due to cancer.
The media picked up on the idea that doctors were missing opportunities because the patients who died early (within three months of diagnosis) had visited their GPs five times in the four months before they were diagnosed, compared with the patients who survived for longer, who had been an average of four times. What they didn't report was that the study didn't look for the reasons the patients had seen their doctor. Many lifelong smokers have a host of medical conditions, and may have been seeing their GP for completely different reasons - heart problems, osteoarthritis, or tiredness, to name but a few. There's nothing to suggest these patients actually went to see their doctor with symptoms of possible lung cancer.
Where this research might be helpful is in developing tools which could be used to work out whose symptoms are particularly worrying. For instance, doctors know that a cough or shortness of breath in an otherwise healthy non-smoker is less likely to be due to lung cancer than in an older smoker. But there were other findings which could help doctors tease out whose symptoms are due to their COPD and whose need investigating urgently.
In this study, certain factors were linked with dying within three months of diagnosis - they included continuing to smoke (almost 50% higher risk than people who survived longer than three months); being male (17% higher risk), being older (80% higher risk for over-80s compared to those in their late 60s); living in the countryside (20% higher) and living in an area of social deprivation (16% higher).
There's an awful lot of technology out there. Some (like everything on the patient.info website) is reliable and helpful. Some (like much of the bureaucratic red tape every GP lives with) is not. A software tool that really helped GPs pick up cases more quickly without swamping their local chest clinic? That I'd like to see.