Association between sleep duration, depression and breast cancer in the United States: a national health and nutrition examination survey analysis 2009–2018

Objective: Breast cancer is the most common cancer in women, threatening both physical and mental health. Epidemiological evidence for associations between sleep duration, depression, and breast cancer is inconsistent. This study aimed to determine these associations and build machine-learning algorithms to predict breast cancer. Methods: 1,789 participants from NHANES 2009–2018 were included; 263 had breast cancer. Sleep duration was self-reported; depression was assessed by PHQ-9 (score ≥10 = depression). Multivariable logistic regression estimated adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for sleep duration and depression with four adjustment models. Six machine-learning algorithms (AdaBoost, Random Forest, Boost tree, Artificial Neural Network, Extreme Gradient Boosting, and Support Vector Machine) were trained to predict breast cancer. Results: BMI, race, and smoking differed significantly between breast cancer and non-breast cancer groups. Depression was associated with breast cancer (OR = 1.99; 95% CI: 1.55–3.51). Compared with 7–9 h sleep, ORs for <7 h and >9 h were 1.25 (95% CI: 0.85–1.37) and 1.05 (95% CI: 0.95–1.15), respectively. AdaBoost outperformed other algorithms (AUC = 0.84; 95% CI: 0.81–0.87). Conclusions: No significant association was observed between sleep duration and breast cancer, while depression was associated with higher breast cancer risk. Machine-learning (AdaBoost) showed good predictive performance, providing potential evidence for further mechanistic studies.

Yufan Cai, Yizhou Zhaoxiong, Wei Zhu, Haiyu Wang