Table 2.

Baseline characteristics and completeness of "dual" BMI ascertainment (i.e. BMI measurement on both screening episodes). Significance levels from multiple regression models.

	Women (n = 16,932)			Men (n = 13,812)
	Without "dual" BMI ascertainment	Dual BMI ascertainment	p	Without "dual" BMI ascertainment	Dual BMI ascertainment	p
Mean age (years)	45.3	44.6	0.717	45.3	44.2	0.001
% current smokers	14.3	36.1	0.980	18.6	44.4	0.024
% with SBP>140 mmHg	23.0	15.7	<0.001	30.9	22.3	<0.001
% with DBP>90 mmHg	12.0	8.1	*	21.7	15.1	*
% cholesterol > 6.5 mmol/l	7.6	15.2	0.064	11.5	21.2	0.253
% BMI >30 kg/m2	4.8	11.4	0.150	6.0	11.2	<0.001
% in the two most deprived groups (4 and 5)	29.6	41.2	<0.001**	31.7	38.8	0.006**

SBP: Systolic Blood Pressure; DBP: Diastolic Blood Pressure

*Significant in multiple logistic regression model excluding SBP but omitted from presented model to avoid over-adjustment, as highly correlated with SBP.

** Deprivation group (quintile) entered as a continuous variable.