Table 2—

Adjusted ORs and 95% CI of prevalent diabetes by category of six POPs

AnalyteNot detectableDetectable*
Ptrend
<25th*25th to <50th50th to <75th75th to <90th≥90th
PCB153
    Conc. (ng/g of lipid)14.336.760.293.6164
    Cases/n10/41320/39748/40467/40142/23730/164
    Prevalence (%)2.45.011.916.717.718.3
    Adjusted OR (95% CI)Referent2.5 (1.1–6.0)4.3 (2.2–8.6)5.9 (3.0–11.9)5.9 (3.1–11.3)6.8 (3.0–15.5)<0.001
HpCDD
    Conc. (pg/g of lipid)20.737.860.897.5170
    Cases/n12/26319/43644/43956/43940/26246/177
    Prevalence (%)4.64.410.012.815.326.0
    Adjusted OR (95% CI)Referent1.0 (0.5–2.1)1.7 (0.8–3.8)1.8 (0.9–3.3)1.6 (0.6–3.9)2.7 (1.3–5.5)0.007
OCDD
    Conc. (pg/g of lipid)1943235148051,485
    Cases/n13/39030/40147/41046/40850/24131/166
    Prevalence (%)3.37.511.511.320.818.7
    Adjusted OR (95% CI)Referent1.7 (0.8–3.6)2.2 (1.0–5.0)1.6 (0.7–3.7)2.7 (1.2–6.2)2.1 (0.9–5.2)0.094
Oxychlordane
    Conc. (ng/g of lipid)8.515.425.139.165.5
    Cases/n11/35911/40432/42263/41651/24949/166
    Prevalence (%)3.12.77.615.120.529.5
    Adjusted OR (95% CI)Referent0.8 (0.3–2.2)1.9 (0.7–5.5)3.1 (1.0–10.1)4.5 (1.4–14.3)6.5 (2.0–21.4)<0.001
DDE
    Conc. (ng/g of lipid)1122927171,5603,700
    Cases/n16/50235/50656/50257/30453/202
    Prevalence (%)3.26.911.218.826.2
    Adjusted OR (95% CI)Referent1.5 (0.7–3.1)1.6 (0.8–3.4)2.3 (1.0–5.5)4.3 (1.8–10.2)<0.001
trans-Nonachlor
    Conc. (ng/g of lipid)11.021.735.760.6114
    Cases/n4/20311/45128/45560/45160/27454/182
    Prevalence (%)2.02.76.614.023.032.4
    Adjusted OR (95% CI)Referent1.2 (0.4–3.2)2.5 (1.0–6.1)4.9 (2.0–11.8)7.6 (2.9–19.8)11.8 (4.4–31.3)<0.001
Sum of six POPs
    Cases/n2/46334/50553/52765/27563/246
    Prevalence (5)0.46.710.123.625.6
    Adjusted OR (95% CI)Referent14.0 (3.0–65.0)14.7 (3.4–63.9)38.3 (8.0–183.1)37.7 (7.8–182.0)<0.001
    Adjusted OR (95% CI)0.07 (0.02–0.33)Referent1.1 (0.6–1.7)2.7 (1.5–4.9)2.7 (1.5–4.8)<0.001
  • *

    * Percentiles were computed among detectables, retaining all nondetectables as the lowest category.

  • Adjusted for age, sex, race and ethnicity, poverty income ratio, BMI, and waist circumference.

  • ‡ORs with an alternative reference category are presented to more precisely estimate the confidence limits in the higher categories, given theoretical statistical instability in the lowest category, which has only two cases.