Results of power analysis for a one‐sided, two‐sample proportions test (α = 0.05)a
Ncontrol	Nexposed	Proportion controlb	Proportion exposed	Relative risk	Power
17,710	1,710	0.00562	0.00674	1.2	0.1634
17,710	1,710	0.00562	0.00843	1.5	0.4353
17,710	1,710	0.00562	0.01124	2.0	0.8182
17,710	1,710	0.00562	0.01685	3.0	0.9935
17,710	1,710	0.01124	0.01348	1.2	0.2259
17,710	1,710	0.01124	0.01685	1.5	0.6379
17,710	1,710	0.01124	0.02247	2.0	0.9652
17,710	1,710	0.01124	0.03371	3.0	1
17,710	1,710	0.02247	0.02697	1.2	0.3353
17,710	1,710	0.02247	0.03371	1.5	0.8632
17,710	1,710	0.02247	0.04495	2.0	0.9991
17,710	1,710	0.02247	0.06742	3.0	1

Stata code used to generate the above power calculation results: power two proportions ('= 0.5 * 199/17710'= 199/17710'= 2 * 199/17710’), test(chi2) RR (1.2 1.5 2.0 3.0) n1(17710) n2(1710) one‐sided table(N1:''N control'' N2:''N exposed'' p1:''proportion control'' p2:''proportion exposed'' RR:''relative risk'' power:''power'').

^aOne‐sided test α = 0.05 Ho: p2 = p1 vs Ha: p2 > p1; N_controls = 17,710, N_exposed = 1,710; Number of Iterations = 1,000 (data sets).

^bRepresenting 1/2x‐, 1x‐ and 2x‐ the observed background rate of lung cancer of 199/17710 in Jones et al. (2015). Highlighted/bolded region in table above represents power associated with this 1x observed background rate of lung cancer in cited study.