Skip to main content
. Author manuscript; available in PMC: 2009 Aug 17.
Published in final edited form as: Biometrics. 2008 Jul 18;65(2):599–608. doi: 10.1111/j.1541-0420.2008.01096.x

Table 3. Rate ratios for dose categories and estimated ERR (βˆ) for CBC in the < 45, 5+ subgroup using location and covariate effects estimated from all subjects (Global) and from the < 45, 5+ subgroup (Subgroup). CCML: Case-control-matched-location, CL: case-location, CCAL: case-control-all-location.

Analysis method Dose (Gy) Number < 45, 5+ casesa Global δˆ, γˆb RR (95% CI) Subgroup δˆ, γˆc RR (95% CI)
CCML 0 50 1 1
.01-1 34 1.36 (0.79-2.34) 1.34 (0.77-2.32)
1+ 26 2.03 (1.08-3.79) 2.03 (1.07-3.85)
CLd 0-.5 57 1 1
.5-1 27 1.58 (0.50-5.01) 1.74 (0.53-5.70)
1+ 26 1.25 (0.37-4.22) 2.10 (0.52-8.49)
CCAL 0 50 1 1
.01-1 34 1.67 (1.01-2.78) 1.46 (0.86-2.47)
1+ 26 1.50 (0.86-2.61) 1.78 (0.98-3.20)
a

Dose at location of CBC.

b

Location (δ) and subject-level (γ) parameters were estimated from the full data set, 609 case-control sets.

c

All parameters estimated from the < 45, 5+ subgroup, 110 case-control sets.

d

Cannot use zero as baseline (see text).