Table 1.
Odds ratios & 95% confidence intervals for NHL in relation to quartile of lipid–adjusted PCB exposure.
| Exposure, in fourths | Median (ng/g lipid) |
Cases (n=145) |
Controls (n=290) |
Unadjusted OR* (95% CI) |
Adjusted** OR (95% CI) |
p for trend1 |
|---|---|---|---|---|---|---|
| ΣPCB | ||||||
| 1 | 406.9 | 33 | 72 | ref | ref | |
| 2 | 547.8 | 41 | 73 | 1.22 (0.69–2.18) | 1.25 (0.68–2.28) | |
| 3 | 678.0 | 41 | 73 | 1.22 (0.69–2.18) | 1.32 (0.71–2.43) | |
| 4 | 945.4 | 30 | 72 | 0.91 (0.50–1.67) | 1.02 (0.53–1.95) | 0.76 |
| Immunotoxic congeners† | ||||||
| 1 | 75.6 | 34 | 72 | ref | ref | |
| 2 | 111.5 | 56 | 73 | 1.62 (0.93–2.82) | 1.83 (1.01–3.31) | |
| 3 | 149.6 | 30 | 73 | 0.86 (0.48–1.57) | 0.94 (0.51–1.76) | |
| 4 | 228.7 | 25 | 72 | 0.75 (0.39–1.42) | 0.89 (0.45–1.77) | 0.48 |
| Σ (118, 138, 153, 180) | ||||||
| 1 | 185.7 | 33 | 72 | ref | ref | |
| 2 | 257.5 | 36 | 73 | 1.05 (0.58–1.89) | 1.04 (0.57–1.92) | |
| 3 | 334.4 | 48 | 73 | 1.45 (0.83–2.53) | 1.63 (0.90–2.95) | |
| 4 | 471.7 | 28 | 72 | 0.84 (0.45–1.54) | 0.91 (0.48–1.75) | 0.63 |
| PCB 118 | ||||||
| 1 | 27.4 | 38 | 72 | ref | ref | |
| 2 | 42.9 | 49 | 73 | 1.26 (0.73–2.19) | 1.39 (0.78–2.47) | |
| 3 | 61.0 | 31 | 73 | 0.80 (0.45–1.44) | 0.89 (0.48–1.64) | |
| 4 | 104.7 | 27 | 72 | 0.69 (0.37–1.29) | 0.81 (0.42–1.56) | 0.42 |
| PCB 138 | ||||||
| 1 | 34.3 | 31 | 72 | ref | ||
| 2 | 53.2 | 39 | 73 | 1.26 (0.71–2.22) | 1.33 (0.73–2.40) | |
| 3 | 75.7 | 48 | 73 | 1.53 (0.87–2.69) | 1.61 (0.89–2.92) | |
| 4 | 113.3 | 27 | 72 | 0.88 (0.47–1.63) | 0.95 (0.49–1.83) | 0.59 |
| PCB 153 | ||||||
| 1 | 64.9 | 37 | 72 | ref | ref | |
| 2 | 91.2 | 33 | 73 | 0.87 (0.49–1.54) | 0.85 (0.47–1.54) | |
| 3 | 120.3 | 45 | 73 | 1.20 (0.69–2.09) | 1.38 (0.76–2.51) | |
| 4 | 170.0 | 30 | 72 | 0.81 (0.45–1.47) | 0.82 (0.43–1.56) | 0.55 |
| PCB 180 | ||||||
| 1 | 47.8 | 36 | 72 | ref | ref | |
| 2 | 63.4 | 33 | 73 | 0.90 (0.50–1.64) | 1.02 (0.54–1.93) | |
| 3 | 80.5 | 44 | 73 | 1.23 (0.69–2.19) | 1.24 (0.66–2.31) | |
| 4 | 109.4 | 32 | 72 | 0.89 (0.49–1.63) | 1.03 (0.52–2.02) | 0.82 |
| p,p'-DDE | ||||||
| 1 | 343.6 | 30 | 72 | ref | ref | |
| 2 | 779.6 | 43 | 73 | 1.38 (0.79–2.43) | 1.41 (0.76–2.60) | |
| 3 | 1327.0 | 27 | 73 | 0.86 (0.45–1.63) | 0.77 (0.39–1.52) | |
| 4 | 2325.2 | 45 | 72 | 1.52 (0.84–2.73) | 1.56 (0.82–2.97) | 0.33 |
Conditional logistic regression adjusted for matching factors (race (white/non–white), age at blood draw, year & month of blood draw, fasting status at blood draw)
Multivariate conditional logistic regression further adjusted for region (Northeast, Midwest, West, South), BMI (<25, 25–29.9, 30+), current smoking status (never, past, current), parity/breastfeeding (nulliparous, parous & no breastfeeding, parous & some breastfeeding), & height (missing indicator method for BMI, parity).</p>
Test for trend modeled natural log of lipid–adjusted organochlorine as continuous variable.
The immunotoxic congeners include PCBs 66, 74, 105, 118, 156, and 167.