Consultation frequency and urgent suspected cancer referral conversions: GP cohort study 2000-2023

Background GPs know that if previously infrequent attenders do consult with symptoms, they are more likely to be seriously unwell than frequent attenders with equivalent risk factors, but poor continuity of care in modern general practice means identification of infrequent attenders may not occur.

Aim Among patients with urgent suspected cancer referrals (USCRs), to explore the association between historic patterns of attendance and conversions to a cancer diagnosis.

Design & setting 2818 adults registered at a large primary care practice in Reading, England, with no prior cancer diagnosis, whose first USCR was sent 1.1.2000-31.12.2023, were used for this historical cohort study.

Method Logistic regression was used to quantify the association between quintiles of historic consultation frequency in the 36-19 months before USCR, and cancer diagnosis within six months. Age, sex, ethnicity, deprivation, family history of cancer, smoking, body mass index, hypertension, diabetes, cardiovascular, degenerative, metabolic and inflammatory diseases, USCR type and date were investigated as potential confounders.

Results Adjusted odds from logistic regression for USCR conversion to a cancer diagnosis were 2.7 (95% CI 1.8-4.1, P<0.001) times higher for the least compared to most frequently consulting quintile of patients, who consulted 0-1 compared to 12+times respectively in the 36-19 months before USCR, after adjustment for age, sex, smoking, cardiovascular disease and USCR type, with p_TREND <0.001 across quintiles.

Conclusion Historically infrequent consulters’ USCRs have considerably higher odds of cancer conversion compared to those from historically frequent consulters. Flagging notes of infrequent attenders might facilitate earlier cancer diagnosis for them.