Skip to main content
. 2021 May 20;106(9):e3483–e3494. doi: 10.1210/clinem/dgab355

Table 4.

Covariate attenuation of association of log pro-neurotensin/neuromedin N with incident metabolic syndromea

Modelb OR per SD (95% CI) log pro-NT/NMN % Attenuation (95% CI)c
Model 1, with age, race, and sex 1.22 (1.11 to 1.35)
 Waist circumference 1.19 (1.08 to 1.32) 13.3% (–3.9% to 30.6%)
 Smoking, pack-years 1.23 (1.12 to 1.35) –3.5% (–14.9% to 7.9%)
 Weekly exercise frequency 1.22 (1.11 to 1.35) -0.8% (–9.6% to 8.0%)
 Alcohol use 1.21 (1.09 to 1.33) 6.0% (–3.3% to 15.4%)
 Systolic blood pressure 1.22 (1.11 to 1.35) 0.8% (–5.9% to 7.5%)
 Body mass index 1.17 (1.06 to 1.30) 21.1% (–0.4% to 42.6%)
 Low-density lipoprotein 1.20 (1.09 to 1.32) 10.9% (0.3% to 21.5%)
 High-density lipoprotein 1.24 (1.13 to 1.37) –10.4% (–27.3% to 6.6%)
 Triglycerides 1.24 (1.12 to 1.37) –7.3% (–15.9% to 1.3%)
 HOMA-IR 1.12 (1.00 to 1.25) 47.9% (1.7% to 94.1%)
 Southern diet score 1.20 (1.07 to 1.34) 9.2% (–23.1% to 41.5%)
 Sweets/Fats diet score 1.20 (1.08 to 1.35) 7.9% (–24.1% to 39.8%)
 Convenience diet score 1.20 (1.07 to 1.34) 10.1% (–22.4% to 42.7%)

n = 2770

Abbreviations: HOMA-IR, homeostatic model assessment of insulin resistance; OR, odds ratio; pro-NT/NMN, pro-neurotensin/neuromedin N.

a Weighted to analytic cohort.

b Each covariate was individually added to model 1.

c Estimates obtained using bootstrapping with 1000 replicate samples.