Skip to main content

Table 3 Random-intercept logistic regression models for the presence of caries in The Chilean Dental Study

From: LEVEL (Logical Explanations & Visualizations of Estimates in Linear mixed models): recommendations for reporting multilevel data and analyses

 

Value of category

‘Null’ model (n = 2275)

‘Intermediate’ model (n = 2275)

‘Final’ model (n = 2134)b

OR (95% CI)

OR (95% CI)

OR (95% CI)

District-level variables

 Human Development Index

  

0.08 (0.01–0.82)

0.04 (0.01–0.39)

 Rural location

  

1.82 (1.17–2.81)

1.45 (0.98–2.16)

School-level variables

 Administrative dependency

Private (paid)

 

Reference

Reference

Private (subsidized)

 

2.74 (1.64–4.55)

1.12 (0.69–1.83)

Public

 

3.99 (2.28–6.98)

1.65 (0.98–2.77)

 School fluoride program

No program

 

1.33 (0.84–2.12)

1.17 (0.81–1.70)

Child-level variables

 Sex

Male

  

1.22 (1.00–1.48)

 Age

1 year

  

Reference

2 years

  

4.98 (2.18–11.38)

3 years

  

12.14 (5.45–27.04)

4 years

  

15.00 (6.69–33.64)

5 years

  

15.22 (6.68–34.68)

6 years

  

14.34 (6.20–33.23)

 Family income

   

0.93 (0.85–1.02)

 Presence of gingivitis

Absent

  

Reference

Present

  

2.14 (1.67–2.75)

 Educational level of main caretaker

University

  

Reference

No studies

  

0.90 (0.14–5.73)

Primary school

  

1.26 (0.81–1.97)

Secondary school

  

1.60 (1.14–2.24)

Technical school

  

1.15 (0.82–1.61)

 Access to health care

Private

  

Reference

Public

  

1.21 (0.89–1.65)

ICCa district

 

0.0495

0.0135

0.0134

ICCa school within district

 

0.1278

0.0378

0.0153

AICa

 

2863.3

2836.8

2521.6

AUCa

 

0.50 (0.50–0.50)

0.63 (0.60–0.65)

0.72 (0.70–0.74)

  1. aICC intra-level correlation coefficient, AIC Akaike Information Criterion, AUC area under the ROC curve
  2. bNote that the number of children in this model is lower due to non-response to various variables; ‘family income’ had the highest number (76) of non-responders (3.3%)