Table 2

ORs of diabetes associated with individual-level education and city-SEI by sex

Individual and city characteristics*Model 1Model 2†Model 3‡
OR (95% CI)OR (95% CI)OR (95% CI)
Women (n=51 903)
City-SEI1.00 (0.95 to 1.06)0.96 (0.91 to 1.02)
Education level 0.78 (0.73 to 0.83) 0.80 (0.75 to 0.85)
Random effects
Intercept variance (SE) 0.0798 (0.0197) 0.0364 (0.0135)
Education slope variance (SE) 0.0262 (0.0155) 0.0259 (0.0142)
Men (n=37 246)
City-SEI1.02 (0.96 to 1.09)0.97 (0.92 to 1.03)0.98 (0.92 to 1.05)
Education level0.93 (0.87 to 1.00)0.95 (0.89 to 1.02)
Argentina, Brazil, Chile, Colombia, Mexico 0.92 (0.86 to 0.99)
Peru, Panama, El Salvador 1.24 (1.04 to 1.49)
Random effects
Intercept variance (SE) 0.1004 (0.0240) 0.0201 (0.0117) 0.0210 (0.0117)
Education slope variance (SD) 0.0346 (0.0168) 0.0209 (0.0127) 0.0163 (0.0124)
  • Bold values have a p value<0.05.

  • *All models were adjusted for individual age. Education is modelled as a continuous variable, a one-unit increase reflects one higher level of education, for example, less than primary to primary complete. City Social Environment Index (SEI) was standardised to a mean of 0 and an SD of 1. The OR is estimated for a 1 SD (0.58703) difference in SEI.

  • †Model 2 is adjusted for country fixed effects.

  • ‡Model 3 includes an interaction between education and different country groups only for men (p for interaction = 0.002). Combinations of the main effect of education and the interaction coefficient were used to derive estimates for different countries.