Table 2.

RRs and ORs and corresponding CIs of associations between a rare event (incidence = 5%) and three independent variables, estimated by Log-binomial regression, ordinary logistic regression, Cox regression with robust variance and logistic regression with the proposed modification

Independent variable	Log-binomial regression: RR (CI)	Logistic regression: OR (CI)	Cox regression - robust: RR (CI)	Modified Logistic regression: RR (CI)
Predictor A
Unadjusted	6 (2.4 - 14.98)	6.41 (2.52 - 16.28)	6 (2.4 - 14.99)	6 (2.36 - 15.25)
Adjusted *	4.96 (1.89 - 12.98)	5.26 (1.97 - 14.06)	4.97 (1.91 - 12.92)	4.99 (1.86 - 13.34)
Predictor B
Unadjusted	2.57 (1.4 - 4.71)	2.69 (1.43 - 5.06)	2.57 (1.4 - 4.71)	2.57 (1.37 - 4.83)
Adjusted *	1.59 (0.85 - 2.97)	1.64 (0.85 - 3.18)	1.59 (0.84 - 3.01)	1.59 (0.82 - 3.09)
Predictor C
Unadjusted	1.28 (0.74 - 2.2)	1.29 (0.73 - 2.29)	1.28 (0.74 - 2.2)	1.28 (0.72 - 2.26)
Adjusted *	0.98 (0.57 - 1.69)	0.97 (0.54 - 1.74)	0.97 (0.57 - 1.65)	0.96 (0.54 - 1.72)

* Adjusted by the other independent variables.