Cancer is common in stroke (>10%). Cancer and stroke both relate to social position in society. The role of social position in stroke patients' risk of cancer has, however, not been studied. We investigated social position expressed by income and length of education and stroke patients' risk of being diagnosed with cancer.
All patients >40 years of age with first-ever stroke in Denmark 2003-2015 were identified in the Danish Stroke Register (n=85,893) and matched 1:10 on age and sex with the Danish background population without stroke (n=858,740). Information on Charlson co-morbidity, income, education, marital status, ethnicity and cancer diagnosed in the first year after stroke/index date in both cohorts was obtained from nationwide registries. Cox regression models were used to estimate stroke patients' risk of cancer compared to the background population.
In the stroke/background population 2018/15152 patients were diagnosed with cancer. Stroke patients' risk of cancer was confined only to smoking-related cancers (e.g. lung, colon, kidney) (HR 1.64;CI 1.53-1.75). Risk was not increased for non-smoking related cancers (e.g. prostate, breast, malignant melanoma) (HR 1.09;CI 0.99-1.20). For smoking-related cancers risk increased significantly with lower education and income. For non-smoking related cancers the association with social position was reverse.
Stroke patients' cancer risk was socially biased, however, in a bidirectional way. Low social position associated with higher risk of smoking-related cancers while at the same time also being associated with lower risk of non-smoking related cancers The opposite was the case for patients with high social position.