Skip to main content
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Dig Dis Sci. 2021 Feb 25;67(2):689–696. doi: 10.1007/s10620-021-06869-8

Table 2.

Multivariable-adjusted ORs and 95% CIs of nonalcoholic fatty liver disease (NAFLD) by quartiles of toenail Cd concentrations: the CARDIA study

Quartiles of toenail Cd concentrations P for trend
Q1 (lowest) Q2 Q3 Q4 (highest)
Range (ppb) < 4.40 4.40–8.30 8.30–18.70 ≥ 18.7
No. of cases/participants 130/613 141/613 138/609 171/611
Model 1 1 (Ref.) 1.19 (0.90, 1.58) 1.07 (0.80, 1.42) 1.25 (0.94, 1.66) 0.23
Model 2 1 (Ref.) 1.18 (0.88, 1.58) 1.03 (0.77, 1.39) 1.26 (0.93, 1.71) 0.19
Model 3 1 (Ref.) 1.21 (0.87, 1.67) 1.05 (0.76, 1.47) 1.43 (1.02, 1.99) 0.04

All models were constructed by using logistic regression models. P for trend was examined by using the medians of toenail Cd quartiles

Model 1 was adjusted for age (continuous), sex (female or male), race (black or white), and study center

Model 2 was additionally adjusted for education through Y25 (< 12.0, 12.0–15.9, or ≥ 16.0, year), baseline smoking status (never, former, or current), cumulative average alcohol consumption through Y25 (0, 0.1–11.9, 12.0–23.9, or ≥ 24.0, mL/day), cumulative average physical activity through Y25 (quartiles), and baseline BMI (< 25.0, 25.0–29.9, or ≥ 30.0, kg/m2)

Model 3 was additionally adjusted for baseline levels of HDL-cholesterol and insulin (continuous) based on stepwise model selection

BMI Body mass index, CARDIA coronary artery risk development in young adults, CI confidence interva, OR odds ratio, Q quartile, Ref. reference, Y CARDIA exam year