Table 3.
Hazard ratios (HR) of bladder cancer and 95% confidence interval (95% CI) according to lagged benzene exposures in the Norwegian Offshore Petroleum Workers cohort followed 1999–2017.
| Complete-case analysesa | Multiple imputation analysesa | |||||
|---|---|---|---|---|---|---|
| (n = 2156) | (n = 2254) | |||||
| Benzene metric | Cases | Person-years | HR (95% CI)b | HR (95% CI)b | ||
| Duration (years)c | ||||||
| 10-year lagd,e | ||||||
| 0 | 47 | 11,101 | 1.00 | (reference) | 1.00 | (reference) |
| >0–<5.5 | 46 | 8777 | 1.25 | (0.81–1.92) | 1.18 | (0.77–1.81) |
| 5.5–<13.3 | 38 | 8080 | 1.12 | (0.72–1.75) | 1.03 | (0.66–1.60) |
| 13.3–<18.8 | 24 | 4191 | 1.40 | (0.82–2.40) | 1.43 | (0.85–2.41) |
| 18.8–33.5 | 22 | 2671 | 1.61 | (0.91–2.85) | 1.86 | (1.10–3.15) |
| p-trendf | 0.139 | 0.046 | ||||
| Continuous | 177 | 34,820 | 1.01 | (0.99–1.04) | 1.02 | (1.00–1.04) |
| 15-year lagd,e | ||||||
| 0 | 49 | 12,389 | 1.00 | (reference) | 1.00 | (reference) |
| >0–<5.5 | 50 | 9395 | 1.32 | (0.87–1.99) | 1.25 | (0.83–1.86) |
| 5.5–<13.3 | 44 | 8239 | 1.32 | (0.86–2.02) | 1.24 | (0.81–1.89) |
| 13.3–<18.8 | 19 | 3191 | 1.21 | (0.68–2.14) | 1.30 | (0.76–2.24) |
| 18.8–33.5 | 15 | 1605 | 1.45 | (0.77–2.76) | 1.64 | (0.91–2.97) |
| p-trendf | 0.311 | 0.139 | ||||
| Continuous | 177 | 34,820 | 1.01 | (0.99–1.03) | 1.02 | (1.00–1.04) |
| 20-year lagd,e | ||||||
| 0 | 57 | 15,076 | 1.00 | (reference) | 1.00 | (reference) |
| >0–<5.5 | 52 | 10,351 | 1.28 | (0.86–1.90) | 1.21 | (0.82–1.79) |
| 5.5–<13.3 | 43 | 6748 | 1.40 | (0.90–2.18) | 1.34 | (0.87–2.06) |
| 13.3–<18.8 | 19 | 2050 | 1.57 | (0.89–2.78) | 1.62 | (0.96–2.83) |
| 18.8–33.5 | 6 | 594 | 1.26 | (0.50–3.20) | 1.66 | (0.75–3.70) |
| p-trendf | 0.152 | 0.053 | ||||
| Continuous | 177 | 34,820 | 1.01 | (0.99–1.04) | 1.02 | (0.99–1.04) |
| Cumulativec, g | ||||||
| 10-year lagd,e | ||||||
| 0 | 47 | 11,101 | 1.00 | (reference) | 1.00 | (reference) |
| >0–<2.0 | 45 | 8380 | 1.24 | (0.80–1.91) | 1.16 | (0.76–1.78) |
| 2.0–<7.6 | 39 | 8029 | 1.16 | (0.74–1.81) | 1.15 | (0.75–1.77) |
| 7.6–<15.3 | 22 | 3873 | 1.31 | (0.75–2.27) | 1.35 | (0.80–2.28) |
| 15.3–51.4 | 24 | 3437 | 1.56 | (0.92–2.66) | 1.59 | (0.96–2.66) |
| p-trendf | 0.141 | 0.077 | ||||
| Continuous | 177 | 34,820 | 1.01 | (1.00–1.03) | 1.02 | (1.00–1.03) |
| 15-year lagd,e | ||||||
| 0 | 49 | 12,389 | 1.00 | (reference) | 1.00 | (reference) |
| >0–<2.0 | 48 | 8692 | 1.31 | (0.86–1.99) | 1.23 | (0.81–1.85) |
| 2.0–<7.6 | 39 | 7742 | 1.20 | (0.77–1.87) | 1.18 | (0.77–1.82) |
| 7.6–<15.3 | 18 | 3414 | 1.17 | (0.65–2.10) | 1.24 | (0.71–2.15) |
| 15.3–51.4 | 23 | 2583 | 1.79 | (1.04–3.08) | 1.82 | (1.08–3.07) |
| p-trendf | 0.097 | 0.048 | ||||
| Continuous | 177 | 34,820 | 1.02 | (1.00–1.04) | 1.02 | (1.00–1.04) |
| 20-year lagd,e | ||||||
| 0 | 57 | 15,076 | 1.00 | (reference) | 1.00 | (reference) |
| >0–<2.0 | 48 | 8896 | 1.29 | (0.86–1.93) | 1.26 | (0.84–1.87) |
| 2.0–<7.6 | 37 | 6806 | 1.24 | (0.79–1.95) | 1.18 | (0.76–1.83) |
| 7.6–<15.3 | 20 | 2533 | 1.59 | (0.89–2.83) | 1.65 | (0.95–2.84) |
| 15.3–51.4 | 15 | 1507 | 1.76 | (0.95–3.27) | 1.74 | (0.96–3.17) |
| p-trendf | 0.070 | 0.050 | ||||
| Continuous | 177 | 34,820 | 1.02 | (1.00–1.04) | 1.02 | (1.00–1.04) |
aAdjusted for age as timescale, year of first employment, tobacco smoking and education (Model 1). Missing: tobacco smoking (n = 79), education (n = 21). Missing values were imputed in the multiple imputation analysis.
bCox regression adapted to a case-cohort design.
cTime-dependent exposure.
dDisregarding exposure during the most recent 10-, 15-, or 20-year period before any time-point (t) during follow-up.
eCategorised into unexposed (0) and tertiles among exposed. To capture the tail of the distribution, the upper tertile was divided into two by its median.
fModelled by using the medians of each exposure category to test for linear trend.
gThe cumulative benzene exposure metric was based on an intensity score derived from a task-based semi-quantitative JEM.