Skip to main content

Table 4 Individual predictors of log sum of dioxin-like compoundsa, adjusted for age in years

From: Predictors of serum dioxin levels among adolescent boys in Chapaevsk, Russia: A cross-sectional pilot study

Predictor

Estimate

p-value

Multiplicative factor on dioxin (95% CI)

R-Square Value

Age of index son (years)

0.26

0.057

1.30 (1.00–1.72)

0.13

BMI of index son (kg/m2)

0.03

0.40

1.03 (0.95–1.13)

0.15

Consumption of local non-chicken meat (y/n)

0.56

0.042

1.75 (1.05–2.92)

0.26

Consumption of local chicken (y/n)

0.13

0.54

1.14 (0.75–1.72)

0.14

Consumption of local fish (y/n)

0.48

0.079

1.62 (0.97–2.71)

0.23

Consumption of local eggs (y/n)

0.20

0.36

1.23 (0.80–1.88)

0.16

Consumption of local dairy (y/n)

-0.19

0.65

0.82 (0.36–1.89)

0.13

Consumption of local fruits/vegetables (y/n)

-0.12

0.84

0.89 (0.28–2.77)

0.13

Current residential distance from Khimprom (km)

-0.06

0.37

0.94 (0.82–1.07)

0.15

Years resided in Chapaevsk

0.009

0.75

1.01 (0.95–1.07)

0.13

Monthly family income levelb

-0.09

0.50

0.92 (0.71–1.18)

0.14

Maximum parental educationb

0.21

0.26

1.23 (0.86–1.76)

0.17

Log (PCB 118) (ng/g lipid)

0.64

<0.001

1.90 (1.37–2.63)

0.46

Congenital abnormality of index son

-0.015

0.95

0.99 (0.64–1.51)

0.13

Parityb,c

-0.011

0.94

0.99 (0.74–1.32)

0.13

Weeks of gestation

-0.08

0.072

0.92 (0.84–1.00)

0.23

Weeks of breast-feeding index son (per 12 weeks)

-0.004

0.94

1.00 (0.90–1.10)

0.13

Prior breastfeeding durationd (years)

0.09

0.53

1.10 (0.82–1.47)

0.14

  1. Abbreviations: Log, logarithm base 10; BMI, body mass index; PCB 118, polychlorinated biphenyl 118
  2. aSum of dioxin-like compounds includes PCDDs, PCDFs, and coplanar PCBs.
  3. bEach variable was modeled using three ordinal levels.
  4. cParity refers to the number of live births prior to the index son.
  5. dPrior maternal breastfeeding refers to the cumulative number of weeks of breastfeeding children born prior to the index son.
  6. N = 30 for age, BMI, and distance from Khimprom at time of blood draw; N = 29 for all food consumption and medical history predictors; N = 27 for model with PCB118.