Table 4.
Risk estimates based on the linear ERR and log-linear models – impact of different analytic strategies - All studied hematological malignancies combined (results based on the main data set a)
| Cases/Controls | Linear ERR model |
Log-linear model |
||||||
|---|---|---|---|---|---|---|---|---|
| ERR/100 mGy | 90% CI | RR at 100 mGy | RR at 100 mGy | 90% CI | ||||
| Different populations | ||||||||
| Subjects who worked in 30 km zone | 54/180 | 0.66 | 0.00 | 2.65 | 1.66 | 1.29 | 0.95 | 1.77 |
| Cases with confirmed diagnosis and their matched controls | 34/140 | 0.28 | −0.24 | 2.34 | 1.28 | 1.26 | 0.69 | 2.35 |
| Cases with confirmed diagnosis and their matched controls (30 km zone) | 27/83 | 0.29 | −0.24 | 2.41 | 1.29 | 1.28 | 0.69 | 2.49 |
| Entire population b | 117/481 | 0.00 | nd c | 0.43 | 0.99 | 0.99 | 0.77 | 1.26 |
| Entire population excluding SEAD doses d | 105/429 | 0.25 | −0.10 | 1.13 | 1.25 | 1.15 | 0.87 | 1.50 |
| Dose without constraint | 70/287 | 0.14 | −0.06 | 0.89 | 1.14 | 1.07 | 0.91 | 1.23 |
| Unconditional logistic regression e | 70/429 | 0.39 | −0.06 | 1.57 | 1.39 | 1.24 | 0.92 | 1.65 |
| Dose uncertainties | 70/287 | 0.60 | −0.01 | 2.58 | 1.60 | 1.29 | 0.95 | 1.90 |