Skip to main content
. 2016 Feb 16;11(2):e0147926. doi: 10.1371/journal.pone.0147926

Table 8. Multivariate logistic regression to determine predictors of biomedical risk factors.

Models
Variables Hypertension High fasting blood glucose High cholesterol levels TC/HDL- Cholesterol ratio Raised triglyceride
Hosmer-Lemeshow = 0.04 Sensitivity = 62.54% Specificity = 65.85% Hosmer-Lemeshow = 0.07 Sensitivity = 0.0% Specificity = 100% Hosmer-Lemeshow = —Sensitivity = 0.0% Specificity = 100% Hosmer-Lemeshow = 0.36 Sensitivity = 0.0% Specificity = 100% Hosmer-Lemeshow = 0.63 Sensitivity = 73.26% Specificity = 53.22%
Age
    15–39 years Reference (1) Reference (1) Reference (1) Reference (1) Reference (1)
    ≥40 years 4.7 (3.2–6.9)*** 1.8 (1.1–3.0)* 1.7 (1.2–2.4)** 2.1 (1.2–3.6)* 2.2 (1.5–3.2)***
Overweight/Obesity
    No Reference (1) Reference (1) Reference (1) Reference (1) Reference (1)
    Yes 1.7 (1.2–2.3)** a a a a
Smoking
    No Reference (1) Reference (1) Reference (1) Reference (1) Reference (1)
    Yes 1.4 (0.8–2.4) a 0.5 (0.3–0.8)* 0.3 (0.1–0.5)* a
Alcohol consumption
    No Reference (1) Reference (1) Reference (1) Reference (1) Reference (1)
    Yes 1.6 (1.1–2.5)** a 0.8 (0.5–1.2) 0.5 (0.2–1.2) 0.6 (0.4–0.9)*
Low fruit and vegetable intake
    No Reference (1) Reference (1) Reference (1) Reference (1) Reference (1)
    Yes a 1.8 (1.0–3.2)* a a a

Values are reported as odds ratios (95%CI)

* significant at p<0.05

** significant at p<0.005

*** significant at p<0.001

Hypertension = BP >140/90 mmHg, High fasting blood glucose = ≥7.0 mmol/l, High cholesterol levels = ≥5.0 mmol/l, TC/HDL- Cholesterol ratio = >5.0, Raised triglyceride = ≥1.7 mmol/l, High waist circumference = ≥ 102 men and ≥ 88 women, Overweight/Obesity = BMI kg.m2 ≥25 to ≤29.9/ BMI kg.m2 ≥30

a = Not significant in univariate model then dropped