Table 2.

Multivariable-adjusted mixed linear models, demonstrating associations between educational attainment and longitudinal trajectories of mean systolic blood pressure, Framingham Offspring Study, 1971-2001.

		Model Adjustment
Sex (n)	Education (Years)	Age	Age, Conventional Risk Factors	Age, Baseline Blood Pressure	Age, Baseline Blood Pressure, Conventional Risk Factors
Female (n = 2024)	≤12	3.26 (1.46, 5.05)	2.86 (1.13, 4.59)	2.69 (1.09, 4.30)	2.53 (0.93, 4.14)
	13-16	2.00 (0.20, 3.79)	2.14 (0.42, 3.87)	1.30 (-0.31, 2.91)	1.47 (-0.12, 3.07)
	≥17	0	0	0	0
Male (n = 1866)	≤12	2.26 (0.87, 3.66)	1.25 (-0.16, 2.66)	1.20 (-0.07, 2.46)	0.34 (-0.97, 1.64)
	13-16	1.55 (0.16, 2.94)	0.88 (-0.51, 2.27)	0.98 (-0.28, 2.24)	0.39 (-0.90, 1.68)
	≥17	0	0	0	0

Point estimates (and 95% confidence intervals shown in parentheses) represent mean differences in systolic blood pressure (mmHg) between comparison and referent groups. Age adjustment refers to adjustment for baseline age and time from baseline age. Modeling for baseline age and time from baseline was as follows: age+age²+time+ time²+age*time+age*time²+age²*time.

Conventional risk factors include antihypertensive medication, smoking, body mass index and alcohol consumption.