Table 3.
Multivariable linear and logistic repeated measures models to identify dwelling-specific values that are predictive of level of indoor radon (n = 62).
| Variable | Linear repeated measure mixed model |
Generalized estimating equations model with repeated measures |
|||
|---|---|---|---|---|---|
| Estimate (std. error) | F (p-value) | Estimate (std. error) | |Z| (p-value) | Odds ratio (OR) and (95% CIa for OR) | |
|
| |||||
| Time | −0.138 (0.167) | 0.7 (.41) | −0.575 (0.631) | 0.9 (.36) | — |
| Tobacco users in the home | −0.149 (0.286) | 0.3 (.60) | 0.239 (0.944) | 0.3 (.80) | 1.27 (0.20–8.07) |
| Home built <40 years ago | 0.629 (0.263) | 5.7 (.020) | 3.553 (1.680) | 2.1 (.034) | 34.90 (1.30–938.51) |
| 14 d precipitation | 0.0540 (0.0649) | 0.7 (.41) | 0.130 (0.210) | 0.6 (.54) | 1.14 (0.75–1.72) |
| 14 d barometric pressure | −.0324 (.0158) | 4.2 (.043) | −0.195 (0.0822) | 2.4 (.018) | 0.82 (0.70–0.97) |
| Carbonate bedrock | 1.019 (0.286) | 12.7 (<.001) | 3.161 (1.170) | 2.7 (.0069) | 23.59 (2.38–233.43) |
| Near soil radon | 0.103 (0.0818) | 1.6 (.21) | 0.852 (0.376) | 2.3 (.024) | 2.34 (1.12–4.90) |
| Far soil radon | 0.115 (0.0862) | 1.8 (.18) | 0.515 (0.328) | 1.6 (.12) | 1.67 (0.88–3.19) |
| Slope | 8.29x10−3 (0.0450) | <0.1 (.85) | 0.0341 (0.115) | 0.3 (.77) | 1.03 (0.83–1.30) |
| Roughness | 0.0140 (0.0655) | 0.1 (.83) | 0.0696 (0.214) | 0.3 (.75) | 1.07 (0.70–1.63) |
| TPIb | −0.403 (0.541) | 0.6 (.46) | −1.682 (1.998) | 0.8 (.40) | 0.19 (0.0037–9.35) |
a
CI: confidence intervals.
b
TPI: topographic position index.
Note: Bold reflects significant associations with level of indoor radon.