Risk prediction models with incomplete data with application to prediction of estrogen receptor-positive breast cancer: prospective data from the Nurses' Health Study
2008

Improving Breast Cancer Risk Prediction with Estradiol Levels

Sample size: 59812 publication 10 minutes Evidence: high

Author Information

Author(s): Bernard Rosner, Graham A Colditz, J Dirk Iglehart, Susan E Hankinson

Primary Institution: Brigham and Women's Hospital and Harvard Medical School

Hypothesis

Does the addition of estradiol levels improve breast cancer risk prediction models?

Conclusion

Estradiol levels in postmenopausal women significantly enhance the prediction of breast cancer risk.

Supporting Evidence

  • The addition of estradiol levels improved the C statistic from 0.635 to 0.645.
  • 1,559 invasive estrogen receptor-positive breast cancer cases were confirmed during the study.
  • Plasma estradiol levels were significantly associated with breast cancer risk.
  • The study used a large cohort of female nurses for data collection.
  • Imputed estradiol levels were derived from a nested case-control study.

Takeaway

This study found that measuring estradiol levels can help predict breast cancer risk better, especially for women after menopause.

Methodology

The study used linear regression and multiple imputation methods to develop a risk model incorporating estradiol levels and other risk factors.

Potential Biases

Potential bias due to the small size of the nested case-control study affecting the precision of estimates.

Limitations

Not all participants had measured estradiol levels, and the nested case-control study had a small sample size.

Participant Demographics

The study involved 59,812 female nurses aged 30 to 55 years at baseline, followed from 1980 to 2000.

Statistical Information

P-Value

<0.001

Confidence Interval

1.8 to 6.0

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.1186/bcr2110

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication