Does emergency presentation of cancer represent poor performance in primary care? Insights from a novel analysis of linked primary and secondary care data.

Background: People diagnosed with cancer following emergency presentation have poorer short-term survival. To what extent this signifies a missed opportunity for earlier diagnosis in primary care remains unclear as little detailed data exist on the patient/general practitioner interaction beforehand.

Methods: Analysis of primary care and regional data for 1802 cancer patients from Northeast Scotland. Adjusted odds ratios (OR) and 95% confidence intervals (CIs) for patient and GP practice predictors of emergency presentation. Qualitative context coding of primary care interaction before emergency presentation.

Results: Emergency presentations equalled 20% (n=365). Twenty-eight per cent had no relevant prior GP contact. Of those with prior GP contact 30% were admitted while waiting to be seen in secondary care, and 19% were missed opportunities for earlier diagnosis. Associated predictors: no prior GP contact (OR=3.89; CI 95% 2.14-7.09); having lung (OR=23.24; 95% CI 7.92-68.21), colorectal (OR=18.49; CI 95% 6.60-51.82) and upper GI cancer (OR=18.97; CI 95% 6.08-59.23); ethnicity (OR=2.78; CI 95% 1.27-6.06).

Conclusions: Our novel approach has revealed that emergency cancer presentation is more complex than previously thought. Patient delay, prolonged referral pathways and missed opportunities by GPs all contribute, but emergency presentation can also represent effective care. Resources should be used proportionately to raise public and GP awareness and improve post-referral pathways.