Exploring geographical differences in the incidence of colorectal cancer in the Norwegian Women and Cancer Study: a population-based prospective study

Abstract

Purpose

Norway has experienced an unexplained, steep increase in colorectal cancer (CRC) incidence in the last half-century, with large differences across its counties. We aimed to determine whether geographical distribution of lifestyle-related CRC risk factors can explain these geographical differences in CRC incidence in Norwegian women.

Methods

We followed a nationally representative cohort of 96,898 women with self-reported information on lifestyle-related CRC risk factors at baseline and at follow-up 6–8 years later in the Norwegian Women and Cancer Study. We categorized Norwegian counties into four county groups according to CRC incidence and used Cox proportional hazard models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for risk factors. We used the Karlson, Holm, and Breen (KHB) method of mediation analysis to investigate the extent to which the risk factors accounted for the observed differences in CRC incidence between counties.

Results

During an average of 15.5 years of follow-up, 1875 CRC cases were diagnosed. Height (HR=1.12; 95% CI 1.08, 1.17 per 5 cm increase); being a former smoker who smoked ≥10 years (HR=1.34; 95% CI 1.15, 1.57); or being a current smoker who has smoked for ≥10 years (HR=1.28; 95% CI 1.12, 1.46) relative to never smokers was associated with increased CRC risk. Duration of education >12 years (HR=0.78; 95% CI 0.69, 0.87) vs ≤12 years, and intake of vegetables and fruits >300 g (HR=0.90; 95% CI 0.80, 0.99) vs ≤300 g per day were associated with reduced CRC risk. However, these risk factors did not account for the differences in CRC risk between geographical areas of low and high CRC incidence. This was further confirmed by the KHB method using baseline and follow-up measurements (b=0.02, 95% CI −0.02, 0.06, p=0.26).

Conclusion

Lifestyle-related CRC risk factors did not explain the geographical variations in CRC incidence among Norwegian women. Possible residual explanations may lie in heritable factors.

Keywords:

• lifestyle
• diet
• risk factors
• colorectal cancer
• women
• NOWAC study

Supplementary materials

Table S1 Comparison of the complete-case and imputed dataset, the Norwegian Women and Cancer study

Table S2 Hazard ratios (HRs) and 95% confidence intervals (CIs) before and after multivariable risk adjustment at baseline (complete-case analysis) in the Norwegian Women and Cancer study

Acknowledgments

We thank the staff and participants in the Norwegian Women and Cancer Study for the time and effort they put forth to advancing medical science. Many thanks to Rolf Wynn, Dolley Charles, Runa Borgund Barnung, and Jan Håkon Rudolfsen for their constructive critiques during the work on this study. The publication of this manuscript was funded by the publication fund of UiT-The Arctic University of Norway. The funder had no role in the study design, data analysis, preparation of the manuscript, and decision to publish.

Abbreviations

CRC, colorectal cancer; NOWAC, Norwegian Women and Cancer Study; FFQ, Food frequency questionnaire; BMI, body mass index; NOK, Norwegian kroner; KHB method, Karlson, Holm, and Breen method.

Ethical approval and informed consent

The Norwegian Women and Cancer Study was approved by the Regional Committee for Medical Research Ethics and the Norwegian Data Inspectorate (P REK NORD 141/2008). All participants gave written informed consent.

Data availability

To access the data supporting the findings presented, kindly contact the person in charge of the NOWAC Study - https://site.uit.no/nowac/contact-information/.

Author contributions

SOO and KBB conceived the study idea. All authors contributed to the data analysis. SOO organised the writing and wrote the initial draft. All authors contributed toward drafting and critically revising the article, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.