Are breast screening recall rates associated with body mass index? Analysis of data from a population-based screening program

Abstract

Background

Increased body mass index (BMI) is associated with increased risk of post-menopausal breast cancer, but little is known about its association with other key screening outcomes. This study investigates the association between BMI and the likelihood of being recalled after routine screening mammography.

Methods

To describe the associations between recall rates (crude and age-standardized) and BMI, descriptive statistics were estimated from screening events from clients attending the population-based BreastScreen Western Australia program. For those recalled, multivariable logistic regression models were fitted to investigate the association between BMI and invasive breast cancer and ductal carcinoma in situ (DCIS).

Results

The probability of being recalled following mammographic screening was non-linearly associated with BMI and differed by screening round (P_interaction=0.019). Women screening for the first time had lower odds of being recalled if they had very low BMI (<18.5 kg/m²) or very high BMI (>35 kg/m²), compared to those with normal BMI. For women who had screened previously, the odds of being recalled were also lower for those with very low BMI. Among those recalled, women with higher BMI had higher odds of being diagnosed with invasive breast cancer, but not DCIS, relative to those with normal BMI.

Conclusion

The decision to recall following mammographic screening may be influenced by prior screening information, which may help explain inconsistent non-linear associations between BMI and recall rates between first time and subsequent screeners. Once recalled, high odds of detection within women with obesity are indicative of increased specificity of mammography to detect cancers.