Table 3. Power for interaction effects on total cancer in VITAL, a 2×2 factorial trial of 10,000 men aged ≥50 and 10,000 women aged ≥55, with 5 years of follow-up
RR | RR | Power | ||
---|---|---|---|---|
Single agenta | Interactionb | Both agents | Main Effect | Interaction |
0.90 | 1.0 | 0.81 | 49.9 | |
0.9 | 0.73 | 80.6 | ||
0.8 | 0.65 | 95.9 | 51.5 | |
0.7 | 0.57 | 99.6 | 87.6 | |
0.85 | 1.0 | 0.72 | 83.4 | |
0.9 | 0.65 | 96.2 | ||
0.8 | 0.58 | 99.5 | 48.8 | |
0.7 | 0.51 | >99.9 | 85.2 | |
0.80 | 1.0 | 0.64 | 97.4 | |
0.9 | 0.58 | 99.6 | ||
0.8 | 0.51 | >99.9 | 46.0 | |
0.7 | 0.45 | >99.9 | 82.5 |
RR = intent-to-treat RR, including noncompliant participants (assuming 80% compliance). Represents the effect among those not assigned to the other intervention and assumes the same effect for both agents.
The interaction is the RR for the combined group divided by the product of risks for the two separate groups—i.e., RRint=RRboth/(RR vitamin D alone*RR omega-3 fatty acids alone). An interaction=1 implies additive effects (no interaction).
RR, rate ratio.