This study selected Banadir Maternal and Child Hospital, which is a public referral hospital, as the study location. This hospital is located in Mogadishu, the capital city of Somalia. Also known as Xamar, it is the most populous city in Somalia, with an estimated 2 million residents [21], and stretches along the southern coast of Somalia on the Horn of Africa. This particular hospital was chosen due to its strategic geographical location at the center of the city and its considerable size and high patient load, making it one of the largest hospitals in the area. The hospital comprises 2 main departments: maternity and pediatrics. Under the Maternity Department, the hospital operates an ANC clinic from Saturday to Thursday, providing support to expectant mothers throughout their pregnancy. Other essential services that are available at this hospital include the Diarrhea Treatment Center, Blood Transfusion Unit, Emergency Unit, TB Center, and Research and Training Center.

A single-blind randomized controlled trial will be conducted in this study, with only the participants blinded to their assigned group. The participants will be randomly categorized into the intervention group and control group. The study protocol will be reported in accordance with the Standard Protocol Items: Recommendations for Intervention Trial (SPIRIT) 2013 Guideline [22] (Figure S1, Multimedia Appendix 1).

Planning of the study protocol began in January 2021, and data collection was completed in June 2022.

The study population consists of pregnant women attending the Banadir Maternal and Child Hospital, Mogadishu, for ANC visits. The inclusion criteria include mothers with a gestational age range of 25 to 33 weeks, plans for postnatal follow-up at the hospital, and residence in Mogadishu, including internally displaced persons. The exclusion criteria include mothers who visited the hospital for other health conditions, those of non-Somali nationality, and those who had previous health education training related to immunization and TB prevention practices.

The 2-sample mean formula was used to determine the sample size for this study. This formula considers the mean and SD of TB preventive practices from a previous health education program conducted by John [19], where a SD of 4.40, an intervention group mean of 20.71, and a control group mean of 18.45 were used in the calculation. To ensure adequate statistical power, several adjustments were made for the estimated eligibility (90%) and attrition rate (30%), with a 95% significance level and 80% desired power. The anticipated high attrition rate is primarily due to the impact of the COVID-19 pandemic, which created significant barriers to participation in the study. Based on this formulation, the total sample size for this study was 370 participants, with an equal number of participants in both the intervention and control groups.

Potential participants were recruited over a 4-week period from the ANC clinic at Banadir Maternal and Child Hospital. To ensure an organized and systematic selection, their names were sorted into the ANC electronic register in chronological order. After obtaining informed consent from all participants who agreed to take part in the study, their eligibility was assessed by applying the inclusion and exclusion criteria checklist. The recruitment process is summarized in Figure S2 (Multimedia Appendix 1).

Randomization was performed at the individual level, with pregnant women as the sampling unit. Ensuring an equal allocation ratio (1:1), participants were randomly assigned into the intervention group, which received the IMB-based health education intervention, or the control group, which was placed on a waiting list. The randomization process was conducted using computer-generated random numbers using a web-based calculator (GraphPad; GraphPad Software) by hospital staff who were independent of all other research processes. The generated sequences were placed in sealed opaque envelopes. Subsequently, another hospital staff member, who was also independent of the research team, was responsible for opening the envelopes sequentially and assigning participants to either the intervention or control group.

Data were collected using paper questionnaires at 3 distinct time intervals: baseline, 2-months postintervention, and 4-months postintervention. Participants were asked to complete a consent form before participating and were informed that their information would remain confidential and used solely for research purposes. The recorded data were securely stored in protected files, and individual responses were not reported publicly; instead, findings were presented in aggregate form. The random assignment took place on the same day that participants were selected for the study, after the collection of baseline data. To facilitate the data collection process, each participant was given a study follow-up card and was requested to carry them during each follow-up visit to the hospital.

Several steps were taken to ensure the validity and reliability of the questionnaire. First, content and face validity were evaluated to ensure that the wording, ordering, and language were appropriate and well understood. The analysis categorized items as good, average, or poor; language clarity was assessed as clear, average, or confusing; and the appropriateness of construct measurement was classified as good, average, or poor. Feedback from experts and participants led to important modifications, including rewording several items for clarity and removing 3 items deemed irrelevant to the study objectives.

Additionally, factor analysis was conducted using SPSS software (version 27.0; IBM Corp) on a separate sample set of 134 participants who were not part of the main study but were recruited from the same hospital to assess sampling adequacy. Cohen’s κ and intraclass correlation coefficient values were also utilized to evaluate the reliability of the questionnaire. All values exceeded 0.60, which is considered acceptable [25].

Table 2 presents the primary and secondary outcome assessments, including the timing of each measurement.

Collected data were entered into SPSS software (version 27.0; IBM Corp) before proceeding with data cleaning procedures. Both statistical and graphical normality tests were applied to continuous variables, such as age, knowledge, attitude, and self-efficacy. At baseline, the comparison of variables between groups used appropriate statistical measures based on the distribution of data. For normally distributed data, results are reported as mean (SD), and comparisons were conducted using an independent t test. For non-normally distributed data, results were reported as median (IQR), and comparisons were conducted using the chi-square test and the Mann-Whitney U test.

To evaluate the main effect of the IMB-based intervention modules on the response variables over time, as well as the group-time interactions, repeated-measures ANOVA was used for normally distributed data. For non-normally distributed data, generalized linear mixed models were applied to account for individual-level variability among participants, or generalized estimating equations were used when focusing on overall population-level trends. Intention-to-treat analyses were conducted separately for the intervention and control groups, and missing data were addressed using multiple imputations. Results were reported as differences between groups with 95% CIs. Statistical significance in this study was set at α=.05.

The trial protocol was approved by the Ethics Committee for Research Involving Human Subjects at the Universiti Putra Malaysia (UPM) (UPM/TNCPI/RMC/JKEUPM/1.4.18.2), the Ethical Review Committee of the Ministry of Health of Somalia (MOHHS/DGO/1166/Feb2024), and Banadir Hospital (BH/30594/4/22). The study was registered on May 24, 2022, with the Pan African Clinical Trials Registry and assigned a unique identification number PACTR202105554130782.

A participant information sheet in the Somali language was provided in written form by trained enumerators to each participant, from whom informed consent was obtained. Participant information and identities remained confidential and used solely for the purpose of this research. Participation in the study was voluntary, and participants had the right to withdraw at any point during the intervention. Participants received compensation of USD 2.00 per follow-up visit to cover transportation costs.

Any significant protocol modifications, such as alterations to eligibility criteria, outcomes, or analysis methods, will be promptly communicated to both the Pan African Clinical Trials Registry and the Jawatankuasa Etika Universiti Putra Malaysia.

Health theories play a crucial role in public health by providing frameworks for understanding health behaviors, designing effective interventions, and evaluating improvements of outcomes. This study will use IMB-based theory constructs to improve knowledge, attitudes, and self-efficacy among antenatal mothers, ultimately aiming to improve BCG uptake and childhood TB prevention practices. However, previous studies in Somalia have primarily been observational in design and have lacked a theoretical framework, which limits an inclusive understanding of the factors influencing behavior change. Therefore, investigating this gap is essential for improving outcomes for mothers and children.

This study is the first to examine the effectiveness of an IMB-based health education intervention in improving BCG uptake and childhood TB prevention practices by comparing intervention and control groups. In particular, the findings of this study may highlight the potential for improving maternal knowledge, attitudes, and self-efficacy regarding BCG vaccination and childhood TB prevention practices in Mogadishu, Somalia. Similarly, it is expected that pregnant women in the intervention group will demonstrate higher TB prevention practices and increased BCG uptake for their children compared with those in the control group. Previous intervention studies supported the effectiveness of health education interventions in improving BCG uptake and TB prevention practices [26-28]. Finally, the findings are anticipated to contribute to improving TB prevention efforts in Somalia, where low BCG coverage, poor maternal knowledge, and fragile health systems continue to pose major public health challenges.

This is the first IMB-based health education intervention designed to improve BCG uptake and childhood TB prevention practices. Additionally, its randomized controlled design, single-blind allocation, theoretical framework, and use of validated tools enhance internal validity and reduce the risk of bias. Furthermore, conducting the trial at Banadir Hospital, a referral facility for maternal and child health, allows for the recruitment of a diverse population of pregnant women, including internally displaced persons, thereby enhancing the generalizability of outcomes.

Nevertheless, this study has several limitations. Despite blinding, cross-group information sharing among participants cannot be entirely ruled out, which may influence the findings. Additionally, outcomes are largely based on self-reported measures, introducing the potential for reporting bias. The inability to blind outcome assessors may also introduce bias in the evaluation of findings. Furthermore, because the trial is facility-based, the findings may not be fully generalizable to women who lack access to health care. Finally, external factors, such as sociopolitical instability or disruptions in the health system, may independently affect BCG uptake regardless of the intervention.

This study is a single-blinded randomized controlled trial involving 370 mothers at Banadir Hospital in Mogadishu, Somalia. Participants were randomized into either an intervention group receiving the IMB-based health education intervention or a waitlist control group. It is anticipated that pregnant women in the intervention program will show significant improvements in the primary outcomes of BCG vaccine uptake and TB prevention practices. Additionally, secondary outcomes, including knowledge, attitudes, and self-efficacy related to childhood TB prevention and the BCG vaccine, are expected to improve compared with those in the control group.

The effectiveness of the IMB-based health education modules developed in this intervention may have the potential to be adopted in routine health education programs for pregnant women during their ANC visits in Mogadishu, Somalia.