Table 3.
Lutein and Age Related Macular Degeneration (ARMD), With Dietary Intake Assessed by FFQ: Standardized Relative Risks (RR*), Statistical Power and Relative Sample Size** (rss) Required for Various Analysis Strategies Under Different Disease Risk Models
| Analysis Strategy | Disease Risk Model | |||
|---|---|---|---|---|
| (a) Not mediated through marker |
(b) Mediated entirely through marker |
(c) Partially mediated through marker |
||
|
RDIa (univariate) |
RR Power (rss) |
0.54 0.655 (1.00) |
0.64 0.413 (1.00) |
0.58 0.533 (1.00) |
|
MBLb (univariate) |
RR Power (rss) |
0.47 0.814 (0.68) |
0.32 0.993 (0.16) |
0.38 0.952 (0.32) |
| Bivariatec | RR RDI RR MBL Power (rss) |
0.65 0.54 0.839 (0.64) |
0.90 0.33 0.986 (0.18) |
0.76 0.41 0.941 (0.34) |
| Howed (ranks) | RR Power (rss) |
0.42 0.891 (0.55) |
0.36 0.958 (0.22) |
0.38 0.948 (0.32) |
| Principal Componentse | RR Power (rss) |
0.43 0.904 (0.52) |
0.37 0.966 (0.21) |
0.39 0.956 (0.31) |
* Standardized relative risk is defined as the RR between the 90th and 10th percentiles of the distribution.
Bold type indicates the most powerful method among the available choices investigated.
** Sample size requirement relative to the univariate RDI analysis.
a Regression of ARMD on RDI
b Regression of ARMD on MBL
c Regression of ARMD on RDI and MBL
d Regression of ARMD on combination of RDI and MBL using Howe's method
e Regression of ARMD on combination of RDI and MBL using Principal Components