Table 2.

Adjusted ORs and 95% CIs for incident any, early and late AMD at visit 5, by tertiles of Western and prudent dietary pattern scores using average visit 1 and visit 3 diet data (n=1278)

Any AMD	Tertile 1	Tertile 2	Tertile 3	P trend*
Western pattern	n=426	n=426	n=426
Events n†	49	50	45
Model 1‡	Reference	1.19 (0.76 to 1.87)	1.43 (0.83 to 2.46)	0.287
Model 2§	Reference	1.31 (0.78 to 2.20)	1.69 (0.88 to 3.25)	0.144
Prudent pattern	n=426	n=426	n=426
Events n	48	54	42
Model 1	Reference	1.11 (0.72 to 1.72)	0.90 (0.56 to 1.46)	0.623
Model 2	Reference	1.24 (0.73 to 2.11)	0.84 (0.49 to 1.46)	0.555
Incident Early AMD¶	Tertile 1	Tertile 2	Tertile 3	P trend
Western pattern	n=417	n=412	n=407
Events n	44	40	33
Model 1	Reference	1.03 (0.64 to 1.66)	1.10 (0.61 to 2.00)	0.882
Model 2	Reference	1.24 (0.71 to 2.14)	1.41 (0.69 to 2.88)	0.493
Prudent pattern	n=411	n=410	n=415
Events n	37	44	36
Model 1	Reference	1.17 (0.73 to 1.89)	1.00 (0.59 to 1.68)	0.858
Model 2	Reference	1.38 (0.76 to 2.52)	0.97 (0.53 to 1.78)	0.996
Incident late AMD	<Median	≥Median	-	P trend
Western pattern	n=639	n=639	-
Events n	9	18	-
Model 1**	Reference	3.44 (1.33 to 8.87)	-	0.014
Model 2	Reference	3.00 (1.06 to 8.47)	-	0.005
Prudent pattern	n=639	n=639	-
Events n	17	10	-
Model 1**	Reference	0.51 (0.22 to 1.18)	-	0.054
Model 2	Reference	0.43 (0.18 to 1.05)	-	0.012

^*P-trend was computed using the continuous diet patterns score.

^†Events=Total cases in the tertile group.

^‡Model 1: Adjusted for age, race, education, daily energy intake (kcal) and smoking status.

^§Model 2: Model 1 further adjusted for inverse probability weights.

^¶The n=1236 because participants with early AMD at visit 3 and/or those who developed incident late AMD by visit 5 were excluded from this analysis.

^**There were no cases of African–American participants with late AMD; therefore, the model for risk of late AMD could not be adjusted for race.

AMD, age-related macular degeneration.