The association between socioeconomic status and colorectal cancer incidence
ForfatterHøverstad, Kjersti Næs
Background: Social inequalities in health persist even in egalitarian countries such as Norway. There is a social gradient found for many cancers, with higher incidence and mortality for lower socioeconomic groups. The social gradient can be positive, with higher incidence for higher socioeconomic groups. Colorectal cancer is one of the most frequently diagnosed cancers worldwide, with varying results in reference to a social gradient in incidence and mortality. Objective: To investigate colorectal cancer incidence associated with socioeconomic status in a Norwegian population. Method: Data from NOWAC (The Norwegian Women and Cancer Study) is used in a prospective cohort study, with data collected from 1991 until 2012. 83 524 women are included. The Cox Proportional Hazards model is applied to calculate hazard ratios (HR) for risk of colorectal cancers by level of education. The analyses are performed for colon and rectal cancer separately, with level of education as a categorical variable, adjusted for age. The final model for colon cancer is also adjusted for smoking, alcohol consumption and income, and stratified for subcohorts. Incidence rates by level of education are calculated. Results: A negative social gradient is found for incidence of colon cancer. HR for those with 10-12 years of education compared to 7- 9 years of education is 0,98 (CI 0,81-1,17), HR for 13-16 years is 0,73 (CI 0,59-0,91) and HR for 17 years or more is 0,61 (CI 0,44-0,83). Smoking and alcohol are confounding factors and added into the model. Smoking reduces the negative social gradient, whilst adding alcohol increases it. Furthermore income is added and stratification for subcohorts done, resulting in a HR for 10-12 years of education at 1,07 (CI 0,88-1,3) compared to 7- 9 years of education, HR for 13-16 years is 0,78 (CI 0,61-0,99) and HR for 17 years or more is 0,66 (CI 0,47-0,92). Incidence rate for colon cancer is 65,28 per 100 000 person years. For each level of education the incidence rates per 100 000 person years are: 7-9 years: 93,36, 10-12 years: 68,86, 13-16 years: 46,95 and 17 years or more: 37,67. No significant social gradient is found for incidence of rectal cancer. HR for 10-12 years of education is 0,99 (CI 0,76-1,23), HR for 13-16 years is 1,01 (CI 0,74-1,37) and HR for 17 years or more of education is 0,95 (CI 0,64-1,41) compared to 7-9 years of education. Incidence rate for rectal cancer is 30,35 per 100 000 person years. For each level of education the incidence rates per 100 000 person years are: 7-9 years: 36,03, 10-12 years: 29,3, 13-16 years: 28,31 and 17 years or more: 26,14. Conclusion: There is a negative social gradient associated with incidence of colon cancer, which remains after adding behavioural risk factors such as smoking and alcohol consumption. No social gradient is found for incidence of rectal cancer. The negative social gradient for colon cancer incidence cannot be fully explained by known behavioural risk factors.
ForlagUiT Norges arktiske universitet
UiT The Arctic University of Norway
Følgende lisensfil er knyttet til denne innførselen: