Imputation method for lifetime exposure assessment in air pollution epidemiologic studies

Table 1 Pearson correlation coefficients between doses computed with different functional forms for the biologic effectiveness factor, averaged over 15 imputations ^a

Samples of women limited by allowed percentage of imputed dose (PDI)
Dose surrogate	CCA (PDI = 0%)		PDI < 20%	PDI < 40%	PDI < 60%	PDI < 80%	(PDI < = 100%)
	N:	547	1164	1477	1764	2051	2986
Cumulative dose 1960-1995		(comparison dose variable)
Dose for 1995 only (Promoter model)		0.59 (0.11)	0.39 (0.15)	0.39 (0.16)	0.39 (0.16)	0.41 (0.16)	0.33 (0.15)
Peak annual dose in 1960–1995 (Threshold model)		0.99 (0.00)	0.93 (0.00)	0.93 (0.00)	0.92 (0.01)	0.91 (0.01)	0.75 (0.017)
Cumulative dose X (onset age)^-2 (Age sensitive model)^b		0.89 (0.024)	0.87 (0.028)	0.85 (0.033)	0.84 (0.039)	0.82 (0.044)	0.79 (0.06)
Pre-1960 surrogate^c		−0.020	−0.031	−0.031	−0.030	−0.016	+0.00044

^aNumbers in parenthesis are the square root of the variance across imputations divided by the square root of 15, which provides an estimate of the standard deviation of the average over imputations.
^bWeighting based on the mathematical fit to radiation risks of excess breast cancer in atomic bomb survivors.
^cCalculated for one imputation only.

ISSN: 1476-069X