Table 6 :
Association between CLP and anomalies in number of teeth (outcome variable) adjusted for race, gender and age by using Multiple Logistic Regressiona
| Variable | Adjusted OR (95% CI) | LR statistic (df) | p-value |
|---|---|---|---|
| CLP | 15.26 (6.18, 37.65) | 49.13 (1) | < 0.001 |
| Non-cleft | 1.00 | ||
| Raceb | |||
| Malay | 3.35 (0.35, 32.30) | 1.30 (1) | 0.296 |
| Others | 1.00 | ||
| Genderb | |||
| Female | 1.11 (0.54, 2.29) | 0.08 (1) | 0.777 |
| Male | 1.00 | ||
| Ageb | 1.20 (1.04, 1.37) | 7.04 (1) | 0.010 |
a
the multiple logistic regression is reasonably fit (Hosmer-Lemeshow goodness-of-fit: Chi square = 3.34, df = 8, p-value = 0.851, correctly classified= 78.7%, sensitivity= 30.8%, specificity= 94.8%; area under ROC curve= 0.706).
b
controlled variables: these variables included in the model to control their confounding effect.
b
there is no significant interaction between CLP and controlled variable.