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

Breast Cancer Research

Table 2 Imputation of log estradiol using data from the Nurses' Health Study breast cancer nested case-control study^a

Variable	Beta	Standard error	P value
Intercept	2.241	0.085
Case status	0.130	0.048	0.008
Body mass index (postmenopausal)^b	0.0033	0.0003	<0.001
Birth index^c	-0.0021	0.0006	<0.001
Duration of natural menopause	-0.015	0.004	<0.001
Duration of menopause after bilateral oophorectomy	-0.013	0.004	0.004
R ² = 0.219

^aBased on 164 cases and 346 controls. ^bDuration postmenopause × average body mass index postmenopause. ^cBirth index = $∑ i = 1 s t (t m − t i) b i t MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuqqRPxAKvMB6bYrY9gDLn3AGiuraeXatLxBI9gBaebbnrfifHhDYfgasaacPi6xNi=xH8viVGI8Gi=hEeeu0xXdbba9frFj0xb9qqpG0dXdb9aspeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGaciGaaiaabeqaaeqabiWaaaGcbaWaaabCaeaacqGGOaakcqWG0baDdaWgaaWcbaGaemyBa0gabeaakiabgkHiTiabdsha0naaBaaaleaacqWGPbqAaeqaaaqaaiabdMgaPjabg2da9iabigdaXaqaaiabdohaZXGaemiDaqhaniabggHiLdGccqGGPaqkcqWGIbGydaWgaaWcbaGaemyAaKMaemiDaqhabeaaaaa@4334@$ if parous and = 0 if nulliparous where s = parity; t _i= age at ith birth, i = 1, ..., s; t _m= age at menopause; and b _it= 1 if parity is greater than or equal to i at age t and = 0 otherwise.