This study describes the pathways lung cancer patients took from first symptom recorded in primary care to diagnosis. There were three main routes. The majority (61%) were referred by their GP for specialist investigation as outpatients, though approximately one third of these were to non-respiratory departments. However, there were improvements in the proportion referred to respiratory clinics in later years. Overall this means that only 45% of the cohort (110 of 246) took the standard pathway. Almost all of these had had a chest X-ray taken in primary care. The next commonest route was by emergency admission, accounting for nearly a quarter of the cohort. The interval from the first recorded symptom and diagnosis was longest for misdirected referrals, and shortest for those admitted as an emergency. The difference in speed of diagnosis was in the interval after referral, not before referral. The final route was for those without a respiratory symptom, many of whom were also admitted as an emergency, but without predominant respiratory symptoms.
Strengths and weaknesses
This is a single cohort from one area, and may not be typical. It is relatively small, though it did identify all the cancers occurring over a five year period. It overlaps the introduction of the first referral guidance for suspected cancer sent to GPs in 2000, and does not take account of the minor changes when these were revised in 2005 [8]. We do not know if the higher proportion referred to a respiratory clinic was due to the NICE guidance. However, it seems unlikely, given the unchanged ease of access to primary care chest X-rays, and that respiratory departments had been offering urgent appointments for suspected lung cancer (largely based on abnormal chest X-rays) for many years before the referral guidance was issued. We did not have mortality data, so could not assess if there was any association between the routes of diagnosis and clinical outcomes. Furthermore, the data originate from the GP records, and any omissions in medical recording of symptom, investigations or referrals will have weakened this study.
Comparison with previous literature
No primary care study has identified what proportions of patients take the different routes to diagnosis. Much recent work has concentrated on outpatient referrals, and whether patients take the urgent or non-urgent route [9–11]. In a large UK study of cancers in 1999 and 2000, Allgar reported that 80% of lung cancer patients had seen their GP before diagnosis, a lower percentage than reported here [11]. These were self-reports rather than our GP reports, perhaps explaining part of the differences. That study also reported that patients who had seen their GP before diagnosis experienced longer delays in diagnosis than those who had not: this was again not seen in the study reported here. A Swedish cohort of 364 patients collected in secondary care found 7% to be asymptomatic, though how these patients had their cancer actually identified was not described [12]. Only one patient in the study reported here was symptomless; this is not surprising as few screening chest X-rays are now performed. Abnormal X-rays are the main trigger for referral to a respiratory physician, both in this study and another comparing urgent with non-urgent referrals [13].
Implications of the findings
The main finding is that only a minority of patients take the standard route of GP referral to a respiratory physician. This is a smaller percentage of patients taking the standard route than for colorectal, breast or prostate cancers [14–16]. In part this reflects the group who were admitted as an emergency (though roughly the same proportion of patients with colorectal cancer are first diagnosed as part of an emergency admission) [17]. The other likely explanation is that lung cancer patients may be more systemically unwell, with respiratory symptoms either minor, or even absent. This may also explain those patients who were referred to other departments. These were also less likely to have been X-rayed. These patients all had a symptom associated with lung cancer, but other features may have predominated, making the GP omit an X-ray. Whatever the explanation, this finding is important, as most diagnostic initiatives have been concentrated upon the standard pathway.
Nearly a quarter of patients were admitted to hospital as emergencies. Emergency admissions have not previously been reported as a common route to diagnosis in lung cancer. Many of these admissions will have been for a respiratory infection, and the underlying cancer will only have become apparent because of delayed recovery, or an abnormality on the chest X-ray.
The third group was those without a respiratory symptom. It is arguable that this group of unexpected diagnoses is even smaller than reported, as many of these patients were ill enough to be admitted to hospital, even if lung cancer was not being considered. Every condition has atypical presentations; that these are relatively rare in lung cancer is helpful.
Lessons for expediting diagnosis of lung cancer
Several messages arise from this study. The first is for GPs, in that some patients with lung cancer are referred to the wrong speciality, and their diagnosis is delayed as a result. This may be improved if there was a greater willingness to take a chest X-ray. The NICE guidelines encourage this practice, though it could be argued that the threshold for ordering an X-ray is set too high in those recommendations.
The median interval between first recording of the cancer symptom in the notes and eventual diagnosis is nearly four months; around half of this is delay is in the GP making a referral. This finding suggests that there is some potential for expediting diagnosis in symptomatic lung cancer patients. The median time from referral to diagnosis is short. This may reflect the smaller numbers of urgent referrals made to respiratory clinics when compared to the other common cancers. This is a major advantage of the chest X-ray. Those who have an abnormal X-ray – and thus qualify for urgent referral – have a much higher likelihood of cancer than their counterparts with possible breast or colorectal cancer. Thus, respiratory clinics should be able to offer an efficient service – this study suggests they do.
Thirdly, researchers who wish to expedite lung cancer diagnosis will have to consider the three different main routes to diagnosis. Concentrating on the standard pathway will miss around half of patients, including those with the greatest delays in diagnosis.
Finally, policymakers need to address how patients can be encouraged to attend when they have a lung cancer symptom [3, 4]. Early presentation may obviate some of the emergency admissions, and allow more accurate direction of referrals. This is not an easy task, as many of the symptoms of cancer carry a low risk [2], and there is a fine dividing line between promoting early presentation, and a need to avoid overwhelming GPs and radiology departments.
