Skip to main content
. 2016 Aug 16;17(8):1338. doi: 10.3390/ijms17081338

Table 5.

Sensitivity analysis of the association between the pre-miR-27a-rs895819 A>G polymorphism and risk of having a high alcohol intake 1.

Variable % Drinker High 2 Risk 3
Sex AA + AG GG p 4 OR 95% CI p 5
Men (n = 368) 13.0% 30.0% 0.011 2.84 (1.12–7.17) 0.028
Women (n = 639) 2.9% 10.7% 0.003 3.79 (1.36–11.64) 0.012
p 6 for interaction: 0.774
Variable % Drinker High 2 Risk 3
Obesity AA + AG GG p 4 OR 95% CI p 5
Non-obese (n = 495) 8.2% 21.0% 0.005 3.31 (1.34–8.18) 0.010
Obese (n = 512) 5.1% 13.6% 0.022 3.87 (1.21–12.35) 0.022
p 6 for interaction: 0.934
Adherence to MedDiet AA + AG GG p 4 OR 95% CI p 5
Low < 9 (n = 511) 5.1% 17.1% 0.002 4.56 (1.71–14.34) 0.003
High ≥ 9 (n = 496) 8.2% 17.8% 0.033 2.49 (0.09–6.60) 0.069
p 6 for interaction: 0.546
Variable % Drinker High 2 Risk 3
Diabetes AA + AG GG p 4 OR 95% CI p 5
No (n = 539) 9.0% 24.5% 0.001 3.56 (1.54–8.23) 0.003
Yes (n = 468) 3.9% 8.1% 0.221 2.06 (0.52–8.18) 0.304
p 6 for interaction: 0.547
Variable % Drinker High 2 Risk 3
Hypertension AA + AG GG p 4 OR 95% CI p 5
No (n = 163) 8.0% 23.1% 0.103 4.59 (0.77–27.59) 0.096
Yes (n = 844) 6.4% 16.4% 0.004 3.22 (1.50–6.90) 0.003
p 6 for interaction: 0.818

1 OR and 95% CI were estimated by multivariable logistic regression models adjusted for the covariates indicated below; 2 % of subjects having a high alcohol intake (>26.4 g/day in men and >13.2 g/day in women) depending on the pre-miR-27a-rs895819 polymorphism; 3 OR of being a high alcohol drinker in comparison with non-drinker + moderate, depending on the variable considered for GG individuals versus AA + AG (recessive model); 4 unadjusted p-value for comparison of percentages; 5 model adjusted for sex, age, type 2 diabetes, hypertension, dyslipidemia, obesity, smoking, physical activity and total energy intake; 6 p-value for the interaction term between the corresponding variable (sex, obesity, adherence to MedDiet, diabetes or hypertension) and the pre-miR-27a-rs895819 polymorphism (recessive) in the multivariable adjusted model.