Age and cancer type:  Associations with increased odds of receiving a Late Diagnosis in people with advanced cancer.

Abstract Purpose: COVID has increased the rates of late cancer diagnoses. Clinicians need to be aware of which patients are at higher risk of receiving a late cancer diagnosis, in order to deliver appropriate and timely care planning and minimise avoidable late diagnoses. We aimed to determine which demographic and clinical factors are associated with receiving a ‘late’ cancer diagnosis (within the last 12 weeks of life). Method: Retrospective cohort study of 2,443 people who died from cancer (‘cancer decedents’) in 2013-2015. Demographic and cancer registry datasets linked using patient-identifying Community Health Index numbers. Analysis used binary logistic regression, with univariate and adjusted odds ratios (SPSS v25). Results : One third (n=831,34.0%) received a late diagnosis. Age and cancer type were significantly associated with late cancer diagnosis (p<0.001). Other demographic factors were not associated with receiving a late diagnosis. Cancer decedents with lung cancer[i]were more likely to have late diagnosis than those with bowel (95%CI 1.52 (1.12 to 2.04)), breast or ovarian (95%CI 3.33 (2.27 to 5.0) or prostate (95%CI (9.09 (4.0 to 20.0)) cancers. Cancer decedents aged >85 years had higher odds of late diagnosis (95%CI 3.45 (2.63 to 4.55)), compared to those aged <65 years. Conclusions: Cancer decedents who were older and those with lung cancer were significantly more likely to receive late cancer diagnoses than those who were younger or who had other cancer types. [i] Odds Ratios presented in abstract are the inverse of those presented in the main text, where lung cancer is the reference category. Presented as 1/(OR multivariate).