Education-Based Oral Health Promotion Intervention During Pregnancy: Implementation Protocol in Bangladesh

Study Design and Ethical Oversight This cluster-randomized, hybrid effectiveness-implementation trial (Protocol Version 1.1, March 2026) utilizes the PRECEDE-PROCEED framework to address maternal oral health in Bangladesh. The study is conducted in eight Upazila Health Complexes (UHCs) across eight administrative divisions. This protocol and all recruitment materials have been approved by the IRB of Bangladesh Medical University. Any future amendments to the protocol or study sites will be submitted for board approval before implementation. To prevent treatment contamination, staff at intervention sites will not rotate to control sites, and clinical examiners remain blinded to the 1:1 cluster allocation throughout the 12-week follow-up.

Participant Enrollment and Power A total of 688 pregnant women (86 per cluster) with a gestational age ≤24 weeks will be recruited. The sample size provides 80% power at a 5% significance level to detect a 20% improvement in the primary outcome, assuming a 1.5 design effect (ICC = 0.05) and a 20% attrition rate.

Statistical Analysis Plan (SAP) and Assumption Verification Analysis follows the Intention-to-Treat (ITT) principle using Mixed-Effects Linear Regression for clinical scores (OHI-S) and Mixed-Effects Logistic Regression for binary behavioral outcomes. The UHC is included as a random effect to account for clustering.

• Verification of Assumptions: To meet PRS requirements for statistical transparency, the normality of continuous outcomes will be verified using Shapiro-Wilk tests and Q-Q plots. If assumptions are violated, non-parametric methods or log transformations will be employed.
• Multiple Testing: To account for testing multiple variables (OHI-S, Knowledge Score, and Behavior), a Bonferroni correction will be applied to the p-values.
• Missing Data: If missingness exceeds 5%, Multiple Imputation (m=20) will be used to handle incomplete data.

Outcome Measures The primary clinical endpoint is the Oral Hygiene Index-Simplified (OHI-S), measuring debris and calculus on six tooth surfaces (score 0-6). Secondary endpoints include a 10-item validated Knowledge Score and self-reported behavioral changes in twice-daily brushing frequency from baseline to 12 weeks.

Inclusion criteria

• Pregnant women attending ANC services at participating UHCs
• Gestational age ≤24 weeks at enrolment to allow follow-up
• Age 18 years or older
• Willing and able to provide informed consent

Exclusion criteria

• Presence of severe medical or obstetric complications requiring immediate referral
• Current participation in another behavioural or oral health intervention study
• Conditions preventing participation in oral examination procedures