Skip to main content
. 2020 Sep 4;20:246. doi: 10.1186/s12903-020-01238-9

Table 4.

Count ratios (95% confidence intervals in brackets) of tooth loss in multilevel models with random intercept and fixed effect between individual and municipalities level variables among 35–44-year old in Brazil, 2010

Parameters Empty model (Model 1) Model 2b Model 3c
Fixed part Crude Count Ratio (95% CI) Adjusted Count Ratio (95% CI) Adjusted Count Ratio (95% CI)
Individual level variables
 Constant 8.62 (8.02, 9.26) 8.83 (8.22, 9.48) 10.29 (9.09,11.65)
Education (in years of study)
 0 to 4 1 1
 5 to 8 0.92 (0.90,0.94) 0.92 (0.90,0.94)
 9 to 11 0.72 (0.70,0.74) 0.72 (0.70,0.74)
  > 12 0.50 (0.49,0.52) 0.50 (0.49,0.52)
Income (in minimum wages)
 Up to 1 1 1
 1 to 2.9 0.96 (0.94,0.98) 0.96 (0.94,0.98)
 3 to 4.9 0.84 (0.82,0.86) 0.85 (0.82,0.86)
  > 5 0.65 (0.63,0.67) 0.65 (0.63,0.68)
Municipalities level variables
Municipal Human Development Index
  Low + Medium 1
  High 0.86 (0.76,0.98)
Random part
 Area level variance (Random intercept) 20.09 (15.88,25.41) 14.17 (11.15,18.02) 12.37 (9.71,15.76)
 PCVa −29.46% −12.7%
 Median Rate Ratio MRR = 1.53 MRR = 1.43 MRR = 1.39

Results of Multilevel Poisson Regression Model assuming random intercept and fixed effect. Exponentiated coefficients; aPCV: Proportional change in variance. bModel 2: Adjusted for individual-level variables: sex, age group, skin color, and time since the last dental visit. cModel 3: Adjusted for individual and municipalities level variables: sex, age group, skin color, time since the last dental visit, presence of fluoridated water supply, and coverage of public oral health service above the Brazilian goal