Adult Attained Height and Breast Cancer Incidence (NOWAC)

Abstract

<p>Introduction: Adult attained height has been associated with breast cancer incidence consistently in the literature. The inclusion of covariates to adjust the regression models is widely used as a tool for controlling for error in models and reducing bias. However, several epidemiological studies regarding height have not accounted for confounding variables in their analysis. <p>Material and Methods: The data was collected from the Norwegian Women and Cancer (NOWAC) cohort study (1991-2003). To investigate the association between adult attained heigh and breast cancer risk we used Cox Proportional Hazard regression to estimate the Hazard Ratio (HR) for height and overall breast cancer, stratified by menopausal status. To handle missing data multiple imputation was performed. Cubic splines were fitted to the main exposure to investigate potential non-linearity. The regression-model was constructed using a causal framework (DAG), including variables such as age at baseline, age at menarche, body shape in childhood, breast cancer in mother and years of education. Sensitivity testing was performed to determine the overall influence each covariate had on the effect estimates. <p>Results: The imputed multivariable model indicated a positive association between increased height (per 10 cm increase) and overall breast cancer incidence (HR: 1.21, 95% CI: 1.17-1.25). The association was also observed both in premenopausal women (HR = 1.20, 95% CI 1.14-1.26) and postmenopausal women (HR = 1.22, 95% CI 1.16-1.28). Plots with restricted cubic splines indicated a linear association between height and breast cancer risk. Sensitivity analysis suggested that none of the included variables affected the model to any great extent except year of education. <p>Conclusion: The analysis found a positive association between increased height in Norwegian women and incidence of breast cancer.