Table 3

Multivariable analysis of the factors associated with diabetes prevalence
Variables Fujian Shaanxi
OR 95% CI p value OR 95% CI p value
Age group (20–40 years as ref.)
  40-60 years 2.803 1.887-4.163 <0.001 4.389 2.811-6.851 <0.001
  ≥60 years 6.861 4.437-10.610 <0.001 6.977 4.171-11.673 <0.001
Sex (male as ref.) 0.589 0.429-0.808 0.001 0.621 0.465-0.823 0.001
Central obesity 1.778 1.348-2.345 <0.001 1.558 1.163-2.088 0.003
Cigarette smoking 1.582 1.053-2.378 0.027
Physical activity 0.706 0.526-0.947 0.020
Family history of diabetes 2.752 2.033-3.726 <0.001 3.074 2.150-4.394 <0.001
Metabolic risk factors (0 as ref.)
  1 2.860 2.076-3.939 <0.001 1.629 1.173-2.260 0.004
  2 3.010 1.928-4.699 <0.001 1.463 0.794-2.698 0.223
  3 4.327 2.415-7.751 <0.001 4.518 2.253-9.060 <0.001

All covariables listed were included in the model simultaneously. Ethnics, education level, yearly family income, and alcohol drinking in both provinces, plus cigarette smoking in Shaanxi and physical activity in Fujian, were not significantly associated with diabetes prevalence and were not included in the final model.

Central obesity was defined as waist circumference above 90 cm in men and 80 cm in women. Metabolic risk factors included hypertension, raised triglycerides and total cholesterol.

Xu et al.

Xu et al. BMC Public Health 2013 13:299   doi:10.1186/1471-2458-13-299

Open Data