Table 2. Incidence rate ratios (95% CI) for primary brain tumour risk associated with the residential radon concentrations.
| Radon | Cases, n | IRR (95% CI)b | ||
| Crudec | Model 2c,d | Model 3c,d.e | ||
| Time weighted average (Bq/m3)a | ||||
| <17.2 | 28 | 1.00 | 1.00 | 1.00 |
| 17.2–40.5 | 29 | 1.20 (0.69–2.08) | 1.44 (0.82–2.54) | 1.53 (0.87–2.71) |
| 40.5–67.6 | 31 | 1.27 (0.74–2.20) | 1.68 (0.94–3.01) | 1.86 (1.03–3.38) |
| >67.6 | 33 | 1.38 (0.82–2.36) | 1.90 (1.07–3.39) | 2.09 (1.16–3.79) |
| Linear trend per 100 Bq/m3 | 121 | 1.43 (0.81–2.55) | 1.96 (1.07–3.58) | 2.15 (1.16–4.01) |
| Cumulated exposure (Bq/m3-years)a | ||||
| <377 | 29 | 1.00 | 1.00 | 1.00 |
| 377–895 | 31 | 1.14 (0.69–1.97) | 1.40 (0.81–2.42) | 1.50 (0.86–2.60) |
| 895–1506 | 30 | 1.16 (0.67–1.94) | 1.49 (0.84–2.63) | 1.64 (0.92–2.94) |
| >1506 | 31 | 1.41 (0.83–2.39) | 1.92 (1.08–3.40) | 2.11 (1.17–3.81) |
| Linear trend per 103Bq/m3years | 121 | 1.19 (0.90–1.56) | 1.37 (1.03–1.82) | 1.44 (1.07–1.93) |
a
From 1 January 1971 until censoring, with inclusion of a 10 year latency period. The cut-off points between exposure groups were the 25th, 50th and 75th percentiles for all participants.
b
Analyses based on 51,674 cohort members and 121 brain tumours.
c
Adjusted for age by using it as the underlying time scale in the Cox model and sex.
d
Adjusted for consumption of fruit and vegetables, employment in the chemical industry for at least one year and traffic (time-weighted average NOx exposure between 1971 and the censoring date).
e
Adjusted for employment status, schooling and marital status. Due to exclusion of cohort members with missing value in any covariate, the number of persons is identical in the crude and the adjusted analyses.