Risk prediction models with incomplete data with application to prediction of estrogen receptor positive breast cancer: prospective data from the Nurses’ Health Study
admin IntroductionA sort of boob cancer venture prevision models hit been matured to wage brainwave into a woman’s individualist boob cancer risk. Although circulating levels of oestrogen in postmenopausal women prognosticate ensuant boob cancer risk, whether the constituent of oestrogen levels adds significantly to a model’s prophetic noesis has not previously been evaluated.
Methods:
Using linelike regression, the authors matured an imputed oestrogen reason using rhythmic oestrogen levels (the outcome) and both housing position and venture bourgeois accumulation (e.g., embody accumulation index) from a nested case-control think conducted within a super likely people think and utilised binary imputation methods to amend an coverall venture help including both venture bourgeois accumulation from the important people and oestrogen levels from the nested case-control study.
Results:
The authors evaluated the constituent of imputed oestrogen take to the previously publicised Rosner and Colditz index frequency help for boob cancer venture prevision within the large Nurses’ Health Study cohort. Follow-up was from 1980-2000; during this instance 1559 intrusive oestrogen organ constructive boob cancer cases were confirmed. The constituent of imputed oestrogen levels significantly reinforced venture prediction; the age-specific concordance datum accumulated from 0.635+/-0.007 to 0.645+/-0.007 (p<0.001) after constituent of imputed estradiol.
Conclusions:
Circulating oestrogen levels in postmenopausal women materialize to add to another style factors in predicting a woman’s individualist venture of boob cancer.
Posted in Cancer |