Table 2.

Power of unconditional and conditional logistic regression models.

	Age distribution of unexposed and exposed subjects (in years)
	N(65, 102) vs. N(70, 102)		N(60, 102) vs. N(70, 102)		N(50, 102) vs. N(70, 102)
d	Unconditional	Conditional	Unconditional	Conditional	Unconditional	Conditional
Odds ratio associated with a 10-year increase in age = 1
0	0.73	0.73	0.78	0.77	0.78	0.80
1	0.77	0.76	0.76	0.77	0.78	0.81
2	0.73	0.72	0.76	0.75	0.78	0.81
3	0.73	0.73	0.77	0.78	0.78	0.80
Odds ratio associated with a 10-year increase in age = 1.5
0	0.76	0.76	0.80	0.80	0.82	0.84
1	0.75	0.74	0.81	0.82	0.79	0.83
2	0.80	0.80	0.81	0.80	0.81	0.83
3	0.78	0.78	0.82	0.82	0.81	0.84
Odds ratio associated with a 10-year increase in age = 2
0	0.80	0.79	0.80	0.80	0.76	0.78
1	0.79	0.79	0.83	0.82	0.81	0.83
2	0.79	0.79	0.82	0.80	0.78	0.82
3	0.78	0.77	0.80	0.80	0.75	0.76
Odds ratio associated with a 10-year increase in age = 3
0	0.76	0.76	0.83	0.83	0.77	0.80
1	0.80	0.80	0.85	0.85	0.76	0.78
2	0.79	0.78	0.82	0.82	0.75	0.79
3	0.81	0.78	0.85	0.83	0.71	0.74

Cases and controls were matched by age ± d. The odds ratio associated with the exposure was 1.5 under the alternative hypothesis, H₀: β_e ≠ 0. Numbers of matching sets were 400, 500, and 900 in the three scenarios of age distributions. Power simulation results that gave a difference between models 5% or greater were highlighted in bold.