Skip to main content
. Author manuscript; available in PMC: 2008 Jan 10.
Published in final edited form as: Am J Clin Nutr. 2007 Dec;86(6):1626–1632. doi: 10.1093/ajcn/86.5.1626

TABLE 4.

Adjusted differences in retinal arteriolar caliber (μm) between the highest and lowest quintiles of energy-adjusted intakes of total fiber, by sex, smoking status, diabetes status, hypertension, and physical activity1

Adjusted differences2 P for interaction3
Sex
 Female (n = 5947) 0.68 (−0.68, 2.03) 0.744
 Male (n = 4712) 1.37 (−0.03, 2.78)
Smoking status
 Never or past (n = 8790) 1.22 (0.17, 2.28) 0.242
 Current (n = 1864) 0.80 (−1.63, 3.22)
Diabetes status
 No (n = 9096) 1.08 (0.05, 2.11) 0.678
 Yes (n = 1515) 1.01 (−1.71, 3.72)
Hypertension
 No (n = 6409) 1.59 (0.36, 2.83) 0.656
 Yes (n = 4197) 0.48 (−1.04, 2.00)
Sports index
 <2.5 (n = 4893) 1.69 (0.28, 3.10) 0.305
 ≥2.5 (n = 5720) 0.26 (−1.07, 1.59)
1

n = 10 659. 95% CIs in parentheses.

2

Comparison of the highest with the lowest quintiles of energy-adjusted intakes after adjustment for central retinal venular equivalent, age, sex, race, center, BMI, smoking (smoking status, smoking years, age at which smoking started, and cigarettes smoked per day), alcohol intake, occupation, education, physical activity, diabetes status, hypertension, long-term systolic and diastolic blood pressure, lipids (HDL, LDL, and triacylglycerol), and dietary factors from both food and supplements (total energy intake, glycemic index, carotenoids, folate, n–3 fatty acids, and vitamins B-6, B-12, C, and E).

3

Likelihood ratio test for interaction (effect modification) by sex, smoking, diabetes, hypertension, or physical activity.