Vitamin D status and breast cancer prevalence among US women: a descriptive cross‑sectional analysis of NHANES 2001‑2018
Abstract
Studies have reported an inverse association between vitamin D and breast cancer risk. However, most prior research has used prospective designs with vitamin D measured before diagnosis, leaving limited descriptive data on vitamin D status among women with prevalent breast cancers in the general US population. This study describes the association between vitamin D status and prevalent breast cancer using a large, nationally representative sample. We conducted a descriptive cross-sectional analysis of data from 15,904 women aged ≥ 20 years in the National Health and Nutrition Examination Survey (NHANES) 2001–2018. Breast cancer status was derived from the Medical Conditions Questionnaire (self-reported). Serum 25-hydroxyvitamin D [25(OH)D] was categorized as deficient (< 50 nmol/L), insufficient (50–74 nmol/L), or sufficient (≥ 75 nmol/L). Survey-weighted logistic regression was used to estimate unadjusted and sequentially adjusted associations. The weighted prevalence of breast cancer was 2.9%. Overall, 24.1% of women were vitamin D deficient, with markedly higher prevalence among non-Hispanic Black women (58.0%). In unadjusted analyses, sufficient vitamin D was associated with higher odds of prevalent breast cancer compared to deficiency (OR 2.48, 95% CI 1.78–3.46). This association was attenuated and no longer statistically significant after adjustment for age, race, season, education, hypertension, and diabetes (OR 1.27, 95% CI 0.89–1.81). Similar patterns were observed in subgroup analyses. These findings suggest no independent association between vitamin D status and prevalent breast cancers after adjustment for available covariates. However, residual confounding and the cross-sectional design with post-diagnosis vitamin D measurement preclude causal inference. Improving vitamin D deficiency remains important for bone health, but routine screening for breast cancer prevention is not supported by these data. Not applicable.