Evaluating Interventions for Breast Cancer Screening in Ghana

Research Design This study will employ a randomised controlled trial (RCT) design. Specifically, a multi-arm parallel RCT design will be used. Participants will be assigned to receive either a CHN-led intervention, a text messaging intervention, or be in the control group. This parallel RCT will be a superiority trial in nature (Spieth et al., 2016). It will seek to compare the two interventions and establish the intervention that is most effective in improving screening uptake, enhancing knowledge, and promoting positive health beliefs. Essentially, participants in the CHN-led and text messaging groups will receive the same information but delivered through distinct media. The baseline for this study will be collected as part of a larger prospective longitudinal study in Ghana, India, and Nepal (ID: SREIC/2025/067).

Description of interventions CHN-led intervention: Participants in this group will engage in structured CHN-led education sessions on breast cancer screening for five months. Each month will focus on a specific theme. The sessions will be facilitated by trained community health nurses who are experienced in health education and culturally competent in working with the study communities. The CHN will receive a brief orientation and training on the study protocol, educational content, and facilitation techniques to ensure consistency in facilitation. This intervention will take the form of bi-weekly sessions that last 60 minutes over five months. These sessions will cover understanding and demystifying breast cancer and its screening (month one), risk factors of breast cancer (month two), suggestive signs and symptoms to prompt early healthcare seeking (month three), costs and benefits of undergoing breast cancer screening (month four), and building confidence and intention to get screened (month five). The intervention will be interactive, incorporating group discussions and culturally tailored educational materials (a take-home pamphlet) to maximise engagement. To ensure engagement, the CHN will be required to monitor attendance, conduct post-session reflections, and do follow-up calls for those who may miss the session. In all five, CHNs will be recruited. Each CHN will serve 5 groups, each consisting of 11 participants.

Text messaging intervention: For this intervention, participants will receive text messages on breast cancer screening for five months. Similar to the CHN-led intervention, the text messaging intervention will cover the following topics: cover understanding and demystifying breast cancer and its screening (month one), risk factors of breast cancer (month two), suggestive signs and symptoms to prompt early healthcare seeking (month three), costs and benefits of undergoing breast cancer screening (month four), and building confidence and intention to get screened (Asare et al., 2024). Informed by the findings of Bonful et al. (2022), text messages will be sent to participants between 6:00am and 8:00pm two times a week - Wednesdays and Saturdays.

Control group: The control group will receive no interventions, but the standard care provided at the community level. This standard of care is the usual community healthcare information that borders on environmental hygiene, healthy eating, exercising, family planning and reproductive health, and motivations to seek prompt healthcare.

Sampling and randomisation procedures Previous studies have found a significantly lower risk of breast cancer among young women (<25 years) (Okyere et al., 2024; Xu \& Xu, 2023). Hence, only women aged ≥25 years will be eligible to participate in the study. Prospective participants must be people who will be residing in the Danfa community for the next nine months. Any woman who is below the eligible age or who knows for a fact that they will be unavailable throughout the duration of the study will be excluded. A list of all eligible participants will be drawn from the baseline study that will precede the RCT. With this baseline information, a simple random sampling technique will be used to select participants into the RCT.

To ensure balance across groups and minimise confounding, a priori stratification will be employed before randomisation. Participants will be stratified by age (≤50 years vs. >50 years) and screening history at baseline (i.e., ever undergone breast examination or no breast examination). After stratification, block randomisation will be used to assign participants to one of the three study arms in a 1:1:1 ratio. This technique will be adopted to guarantee that the number of participants in each stratum remains balanced throughout the study. This will be achieved by listing all participants and applying Microsoft Excel's random number generator to assign individuals to either arm of the study.

A study by Okyere et al. (2024) estimates a breast cancer screening prevalence of 18.39%. This study posits a 20% increase in screening uptake through the interventions. Therefore, with a power of 80% and design effect of 1.2, the sample size is estimated as:

Where:

• Zα is the Z-value for the significance level (for α=0.05, one-tailed: Zα=1.645),
• Zβ is the Z-value for the power (for 80% power, Zβ=0.84),
• p0 is the baseline proportion (control arm),
• p1 is the intervention arm proportion.

Adjusted sample size for design effect (effect=1.2):

Final sample size adjusted for non-response (rate=5%):

Total sample size for three arms:

Substituting the figures, the total sample size will be 828 (i.e., 276 participants per arm).

Data collection procedures Before the RCT, a baseline survey will be conducted. Three research assistants will be recruited and trained to assist with the collection of data. The research assistants will be people with at least a bachelor's degree in public health or nursing and have at least 12 months of experience conducting public health surveys. They will undertake a one-week intensive training that will cover issues of ethical responsibility, the research methodology, interrogation of the questionnaire, and some role plays.

The baseline survey will collect information about the socio-demographic characteristics of the participants (e.g., age in completed years, highest level of education attained, marital status, employment status, religion, health insurance coverage, and exposure to media), risk factors (i.e., family history, smoking status, alcohol consumption, etc.), and reproductive health characteristics (e.g., age at menarche, menopausal status, age at first sex, use of oral contraceptives, etc.). And so, the baseline study will assess the state of depression and anxiety among participants using the Patient Health Questionnaire (PHQ-9) and generalised anxiety disorder (GAD-7) scale, respectively (Shevlin et al., 2022). Additionally, baseline assessment of the participants' attitudes to screening will be done using the attitude to cancer screening (ASCS-15) scale (Sarkhani et al., 2023). Each participant recruited into the baseline (pre-intervention survey) will be assigned a unique code that will be used to follow up on them after the implementation of the intervention.

Five months after the implementation of the intervention, a follow-up data collection will be conducted to measure key outcomes of interest. These include the proportion of women who would have the intention to screen post-intervention, the proportion of women who would have been screened post-intervention, the proportion of women with good attitudes to breast cancer screening post-intervention, and the breast cancer literacy rate post-intervention. Participants will be contacted in-person to minimise non-response and loss to follow-up. The questionnaire will be designed and administered using REDCAP (Harris et al., 2019). The data collection will be stored on my student OneDrive, which will be accessible only to my supervisors.

Data analyses All statistical analyses will be performed in STATA version 18 (StataCorp, College Station, TX) and Rv4.3.2. The study will adhere to the Consolidated Standards of Reporting Trials (CONSORT) guidelines. Descriptive analyses will be performed to compare screening uptake, attitude to screening, and screening intentions at baseline and post-intervention. Pearson's chi-square test will be used to assess whether the observed prevalence differs significantly. To mitigate attrition bias, the study will implement a rigorous intention-to-treat (ITT) approach, wherein all randomised participants will be analysed within their originally assigned groups, irrespective of adherence to the intervention or dropout status. This analytical strategy preserves the benefits of randomisation, provides an unbiased estimate of the intervention's effectiveness, and enhances the generalisability of the findings (Spieth et al., 2016)