Background NICE lung cancer guidelines recommend choosing investigations that give the most information about diagnosis and staging with the least risk to the patient. In particular, performing CT scan before bronchoscopy, PET scan before surgery and avoiding tests that give only diagnosis when information on staging is also needed to guide treatment. We report current UK practice for lung cancer patients diagnosed following a GP referral and assess the impact of the timing of diagnostic and staging tests on the number of tests, diagnostic times and survival.

Methods Data submitted to the National Lung Cancer Audit 2006 to 2011 were analysed. Completeness of the “date CT performed” field was used as a marker of diagnostic data quality. The audit periods 2006–8 were excluded due to CT completeness less than 80%. This study focuses on outpatient pathways and includes GP referrals only.

Results 43,747 patients were identified. The proportion of patients recorded as having each test is shown in the table. The proportion of patients with two or more tests recorded dropped from 30% in 2009 to 26.5% in 2011 (p < 0.001). Mean diagnostic time was significantly shorter in patients undergoing CT before first appointment (30 days versus 36 days, p < 0.001). The mean number of tests was marginally lower in patients undergoing CT before bronchoscopy (1.5 tests versus 1.6 tests, p < 0.01). This equates to 4375 fewer tests in this study population. Of the 7,340 patients who underwent surgery, mean survival in those with a date of death recorded was higher in those with PET recorded (356 versus 300 days, p < 0.001) but no different in those with pre-operative histology recorded (334 versus 322 days, p = ns).

Conclusion The data suggest an improvement in practice consistent with current guidance and that diagnostic pathways with early CT are associated with shorter diagnostic times and fewer diagnostic tests. However, a significant proportion of patients undergoing surgery do not have a PET scan recorded which is associated with worse survival. Further study is required to understand this association but it may relate to unrecognised metastatic disease in patients not undergoing PET before surgery.