Table IV.
Best estimates with 95% CIs (in parentheses) and –loglikelihood values for different model fits with ε1 = m1
| 1 | 2 |
3 |
4 |
|---|---|---|---|
| EPIC males, fit-RIVM1 | EPIC females, fit-RIVM2 | Males and females, fit-RIVM3 | |
| Initiation | |||
| μ0 (per year) | 2.03·10−7 (1.36·10−7–2.60·10−7) | 1.85·10−7 (1.36·10−7–2.47·10−7) | 1.87·10−7 (1.79·10−7–1.95·10−7) |
| y1 (day per cigarette) | – | – | – |
| y 2 | – | – | – |
| Promotion | |||
| ε0 (per year) | 6.6·10−2 (5.1·10−2–8.6·10−2) | 6.3·10−2 (4.5·10−2–7.9·10−2) | 6.5·10−2 (5.3·10−2–7.7·10−2) |
| ε1 (day per cigarette) | m 1 | m 1 | m 1 |
| ε 2 | 1.08 (0.65–1.63) | 1.28 (0.82–2.14) | 1.19 (0.88–1.64) |
| Malignant transformation | |||
| m1 (day per cigarette) | 0.16 (0.11–0.26) | 0.15 (0.11–0.26) | 0.15 (0.12–0.21) |
| m 2 | 1.78 (0.91–6.12) | 1.75 (0.59–10.99) | 1.83 (0.98–3.96) |
| –loglikelihood | 2213.16 | 2204.79 | 4420.56 |
| Sum: | 4417.96 | (Δcorr = −2.93) |
In the fourth column the EPIC data for males and females were fitted simultaneously.