Table 1: Effect of Cross-Over Between the Intervention and the Control Group (Zelen’s Double Consent Design) and Loss to Follow-Up on the Effectiveness of an Early Childhood Caries Prevention Programme

Type of Analysis S-ECC Observed (n/N) * Outcome Parameters (95% Confidence Interval)
Intervention Control Relative Risk Risk Difference NNT **
Analysis for all participants recruited irrespective of their pregnancy outcome and presence or absence of a mother-infant pair
Primary intention to treat analysis 4/322 20/327 0.20 (0.07 - 0.59) 0.05 (0.02 - 0.08) 20 (13 - 50)
Analysis for all mother-infant pairs in the study
Intention to treat analysis 4/307 20/318 0.21 (0.07 - 0.60) 0.05 (0.02 - 0.08) 20 (13 - 50)
Analysis per treatment received 4/312 20/313 0.20 (0.07 - 0.58) 0.05 (0.02 - 0.08) 20 (13 - 50)
Analysis for mother-infant pairs with outcome assessment
Analyses per group allocation 4/227 20/214 0.19 (0.36 - 0.52) 0.09 (0.03 - 0.12) 11 (8 - 33)
Analysis per treatment received 4/232 20/209 0.18 (0.06 - 0.52) 0.08 (0.04 - 0.12) 13 (8 - 25)

* S-ECC = severe early childhood caries.
** Number Needed to Treat to prevent one case of S-ECC. It is calculated as (1/Risk Difference)