Skip to main content
. Author manuscript; available in PMC: 2017 Dec 5.
Published in final edited form as: Caries Res. 2017 May 25;51(4):271–282. doi: 10.1159/000471811

Table 3.

Identification of Caries Risk Determinants in Toddlers using Seemingly Unrelated Bivariate Probit Regression (Final-Fitted Model)a,b,c

Factors Dental Caries Dental Plaque Accumulation
P-value Adjusted OR (95 % CI) P-value Adjusted OR (95 % CI)
Biological characteristics

Ethnicity Chinese 0.006 2.13 (1.24-3.64) 0.668 1.18 (0.55-2.52)
Malay 0.441 1.28 (0.68-2.42) 0.338 0.65 (0.27-1.58)
Indian 1 1

Mother’s Age ≥34 years 0.007 1.90 (1.20-3.03) 0.808 0.93 (0.52-1.66)
29-33 years 0.087 1.50 (0.94-2.39) 0.617 0.88 (0.53-1.45)
<28 years 1 1

Mother’s Education Level University 0.132 0.75 (0.52-1.09) 0.401 0.83 (0.54-1.28)
≤GCE 1 1

Maternal Health Status Existing Condition 0.011 0.61 (0.42-0.89) 0.075 1.50 (0.96-2.34)
No Existing Condition 1 1

Prenatal Plasma Folate Level ≥6 ng/mL 0.002 0.39 (0.21-0.71) 0.733 1.14 (0.54-2.42)
<6 ng/mL 1 1

Night-time Breast-feeding (3 Weeks) Yes 0.003 1.87 (1.24-2.83) 0.307 1.27 (0.80-2.00)
No 1 1

Child’s BMI (24M) kg/m2 0.002 1.18 (1.06-1.32) 0.141 1.10 (0.97-1.24)

Eruption Timing of First Tooth Age in Months - - 0.052 0.88 (0.78-1.00)

Behavioural characteristics

Weaning onto Solids (9M) Age in Months - - 0.014 1.32 (1.06-1.64)

Night-time Bottle-feeding (24M) Yes 0.136 1.35 (0.91-1.99) 0.324 0.79 (0.50-1.26)

No 1 1

Child’s Dental Visits (9M-24M) ≥Once 0.049 2.31 (1.00-5.34) <0.001 0.01 (0.01-0.02)
None 1 1

Parental Belief (Reason for Tooth Decay) (24M) Sugar or Bacteria 0.128 0.65 (0.38-1.13) 0.728 1.13 (0.58-2.20)
Sugar and Bacteria 0.067 0.65 (0.41-1.03) 0.650 1.16 (0.61-2.22)
Beliefs/Ineffective Tooth Brushing 1 1

Parental Belief (Importance of Baby Teeth) Yes 0.078 0.58 (0.31-1.06) <0.001 0.29 (0.16-0.54)
No 1 1

Child’s Daily Brushing Frequency (24M) ≥Once 0.012 1.88 (1.15-3.07) 0.040 0.62 (0.39-0.98)
None 1 1
a

Sample Size=327

b

Variables in final model with p value >0.05 were added in the model since they are well-established risk factors/confounders for the outcome.

c

Modified Hosmer-Lemeshow goodness-of-fit test for biprobit (p=0.67) and Murphy’s score test for biprobit (p=0.64)