Table 5.

Hazard ratios (95% confidence interval) for CHD according to quartiles of dietary lipophilic index/load among participants without missing covariates in the cohort study (1994–2014, N = 60,079) of the Women’s Health Initiative

	Lipophilic index/load				P for trendc
	Q1	Q2	Q3	Q4
Dietary LI
No. of CHD events/patients	652/14,869	669/15,030	649/15,108	759/15,072
Median LI (IQR)	24.4 (2.1)	26.8 (0.9)	28.4 (0.8)	30.4 (1.6)
Unadjusted model, HRs (95% CI)	Ref.	1.18 (1.05, 1.33)	1.16 (1.04, 1.31)	1.30 (1.16, 1.47)	<0.01
Adjusted model 1a, HRs (95% CI)	Ref.	1.13 (1.00, 1.27)	1.12 (0.99, 1.26)	1.23 (1.08, 1.40)	0.02
Adjusted model 2b, HRs (95% CI)	Ref.	1.13 (0.97, 1.32)	1.12 (0.92, 1.36)	1.23 (0.96, 1.58)	0.33
Dietary LL
No. of CHD events/patients	617/14,450	658/15,229	685/15,342	769/15,058
Median LL (IQR)	656.4 (236.2)	1055.4 (200.0)	1511.2 (281.2)	2373.15 (850.9)
Unadjusted model, HRs (95% CI)	Ref.	1.04 (0.93, 1.18)	1.03 (0.91, 1.16)	1.24 (1.10, 1.40)	<0.01
Adjusted model 1a, HRs (95% CI)	Ref.	1.02 (0.89, 1.17)	0.98 (0.83, 1.16)	1.14 (0.89, 1.46)	0.23
Adjusted model 2b, HRs (95% CI)	Ref.	0.98 (0.85, 1.12)	0.91 (0.77, 1.09)	1.02 (0.78, 1.32)	0.25

BMI body mass index, CHD coronary heart disease, CI confidence intervals, IQR interquartile range, LI lipophilic index, LL lipophilic load, HR hazard ratio, PUFA polyunsaturated fatty acid, SFA saturated fatty acid

^aAdjusted for age, region, race/ethnicity, education, income, BMI, physical activity, smoking, family history of myocardial infarction/diabetes/stroke, medication use, postmenopausal hormone use, self-reported baseline hypertension/diabetes/cancer/hypercholesterolemia/hysterectomy status, dietary alcohol, percent calories from protein and carbohydrates, and total energy

^bModel 1 plus dietary PUFA:SFA ratio

^cTest for trends was conducted by treating the median value for each quartile of LI/LL as a continuous variable