Table IV.

AARP study: estimated regression coefficient for log per cent calories from fat in a logistic regression of mortality on log per cent calories from fat and age, using six methods of estimation: naïve logistic regression of mortality on FFQ, regression calibration for FFQ in the main study (RC1), regression calibration for 24HR in the calibration sub-study (RC2), efficient regression calibration (ERC), moment reconstruction (MR) and multiple imputation (IM); standard errors^* are in parentheses.

Gender	N (No. of deaths) in main study	n (No. of deaths) in calibration sub-study	Naïve	RC1	RC2	ERC	MR	IM
Male	322 321 (43810)	987 (114)	0.55 (0.02)	1.06 (0.07)	1.27 (0.72)	1.06 (0.07)	1.37 (0.75)	1.17 (0.74)
Female	223 278 (20382)	966 (94)	0.71 (0.03)	1.54 (0.14)	1.17 (0.75)	1.53 (0.13)	1.14 (0.82)	0.74 (0.69)

^*Standard errors are estimated using a non-parametric bootstrap method with 100 replications.