A protocol for a pilot cluster randomized control trial of e-vouchers and mobile phone application to enhance access to maternal health services in Cameroon

Background: Cameroon still has relatively high maternal mortality rate (MMR) of 596/100,000 live births. Approximately 40% of births are unattended by skilled healthcare personnel with high out-of-pocket expenditures. Poor resource allocation, poorly functioning referral systems, long trekking distances to health facilities, all of which lead to low rates of use of maternal health services. Objectives: The aim of this pilot study is to explore perception and acceptability of mobile health (mhealth) and e-voucher and to determine the feasibility of conducting a large cluster randomized trial to determine the effects of combining e-vouchers and a mobile application compared with usual care in improving access to and use of maternal health services. Methods: This is a multimethod study that comprises two phases. The first phase is the development of the mobile phone app, which includes a qualitative formative study through in-depth key informant interviews and focus group discussions. The second phase is a cluster randomized control trial assessing the combination of e-vouchers and a mobile application compared with usual care in improving access to and use of maternal health services. Feasibility will be determined based on evaluating randomization, contamination, enrollment rate, complete follow up, compliance rate, success in matching data from different sources, and data completeness. Ethics and discussion: Ethics approval has been granted, and the trial has been registered in the Pan-African Clinical Trials Registry. We will disseminate our findings through peer-reviewed manuscripts and conference presentations. Findings from this study will inform the design and conduct of a larger randomized trial. Trial registration: PACTR201808703097367. The trial on the Pan African Clinical Trials Registry. Three conceptual frameworks are considered in this study: The first framework is the three delays model by Thaddeus and Maine (1998), which maps out key factors (delays) at various stages that may affect the intervention [29]. This model was considered in the design of the platform for the mobile phone applications and to explore the experiences of the community to map out contextual barriers to accessing reproductive maternal and child health services and to inform or modify aspects of interventions and features of the mobile application where applicable. The second framework is the Anderson behavioral theory on healthcare utilization; this will assess health-seeking behavior and utilization of RMNCH [30], which is a function of three characteristics [30].The third framework is the transtheoretical model [31, 32], which we will use to understand and map out the five stages of health behavior change within the community in relation to family planning. This framework and these stages will be used to develop social marketing messages for the family planning intervention, which will be based on the community’s needs [31–34]. The rationale for using the three frameworks above is based on the hypothesis that access to services does not necessarily translate to utilization of services. Thus, the three delays model is used to map out the contextual barriers in accessing RMNCH services at different stages, and to guide the development of the mobile application and help inform specific or relevant areas of intervention and modification of the mobile app. The transtheoretical model is specific to understanding the behavioral change process and stages of the family planning aspect to help develop the messages that will be incorporated within the family planning icon, while the Anderson behavioral model will be used to understand the process of the intervention and utilization of services at different levels (see Fig. ​Fig.11). Project framework The study is a multimethod study that follows an exploratory sequential design [35–37]. The study is comprised of two phases: a formative study through key informant interviews and focus group discussions, followed by a cluster randomized control trial including a follow-up focus group discussion for post-intervention groups. The rationale for using this exploratory design is that the formative phase will inform some aspects of the intervention from the outset and the mobile application. This is also an important criterion, as recommended in the guidelines for mhealth evidence reporting and assessment [38]. In addition, the data from the formative study will provide supportive information and findings that can be easily generalized within the context. The intervention will be carried out in the Bali and Ndop Health Districts. These health districts are in the northwest region of Cameroon and are rural areas. The population sizes of these health districts are as follows: Bali: 73,614, Ndop: 218,505. The study districts will be purposefully selected because the district medical officers approved of the implementation of the intervention trial in their respective districts. We planned to stratify PBF and non PBF districts into intervention and control districts; however, with the adoption of PBF as a national strategy, the project may take a pragmatic approach if the non-PBF district eventually gets enrolled into PBF. The objective of this phase is to explore the community’s experiences in accessing RMNCH and their perception and acceptability of using mhealth in the delivery of RMNCH care. This phase will explore the perspectives of pregnant women, relevant stakeholders, and health care providers. Purposive sampling will be used to initially recruit participants for the key informant interviews and focus group discussions. The snowballing technique will be used to reach out to pregnant women who do not have access to the health center. Sampling in each group will continue until emergent themes begin to appear [39–41]. The study aims to explore the following outcomes—community perception and acceptability of using mobile phones and e-vouchers in the delivery of maternal health services and their experiences. The outcome will help to refine features of the app and to understand and inform how changes in behavior are initiated and how they can be sustained. Women who have given birth, pregnant women, and/or breastfeeding mothers will be eligible for the KI interviews, in addition to key informants as identified within the community during stakeholder meetings. A minimum of six and maximum of eight participants (both men and women) will be enrolled in each focus group discussion. Participants for focus group will be recruited using a theoretical sampling approach. The formative study will explore the following sociocultural and structural factors associated with the access to and utilization of maternal health services: the barriers and facilitators during pregnancy; the participants’ perception of antenatal care, skilled birth delivery, and family planning; their experiences with using maternal health services; and their perception of using mobile phones in delivery care. Components as well as the sociocultural and structural factors that facilitate or hinder the providers’ quality of delivery care will also be assessed. Focus group discussions including 6–8 members will be conducted separately for both men and women. The post-intervention focus group will explore their experiences in using the mobile phones and their difficulties, and feedback and recommendations will be solicited to improve and refine the mobile application. The data analysis will include a directed content analysis, which starts with a theory (for example, the three delays model) or relevant research findings as guidance for the initial codes [39–41]. Transcribed text that describes any of the three delays will be highlighted, and a deductive approach will be used in all highlighted text, which will be compared and sorted using all predetermined categories of all the delays. Specifically, for family planning, we will upload transcripts to MaxQDA (https://www.maxqda.com/) and develop codes using grounded theory approach. Next, we will visualize all the data on MaxQDA using maps, emerging themes and modeling, and we will retrieve and export results as code books, maps, or themes. We will use results from the formative study to develop a social marketing strategy for behavior change using the transtheoretical model [31–34] to map out the stages of change for the uptake of family planning and antenatal care services. Finally, we will identify the five levels of change using barriers and enablers identified from key informants and focus group discussion narratives and use these findings to design binary pictorial decision aids for the uptake of family planning and antenatal care services. The development phase of the application involved continuous consultations with relevant stakeholders, including health providers, district medical officers, and community members within the specific context and literature reviews. These processes involved key informant interviews and focus group discussions to inform the intervention and refine the features, as well as to ensure project buy-in and to identify relevant contextual barriers and conditions to facilitate the intervention and ensure availability of the necessary resources to implement the intervention to achieve the desirable outcomes. The theory of change is attached as supplemental material (suppl figure 1) with examples of assumptions, the rationale, and some aspects of the intervention highlighting the various pathways of the intervention (suppl table 1). The final theory of change (ToC) will be updated at the end of the project after consultation with relevant stakeholders and user feedback evaluation including lessons learnt during the implementation processes. To determine the feasibility of conducting a cluster randomized trial to increase access and utilization of antenatal care, skilled birth delivery, and family planning awareness by assessing the processes involved, management of resources, the mobile phone application software, and resources used. We will stratify health areas according to their equity scores (a higher equity score defines how remote or rural the area is, the distance from the district health service, and the enclaved nature of the topography with poor accessibility). Therefore, health areas with equity scores of ≥ 40 in the two participating health districts (Bali and Ndop) will be included. The second stage of stratification involves health areas with equity scores of ≥ 40 with a medicalized health center. All health areas with equity scores of ≥ 40 with no medicalized health center will be excluded at this stage. A medicalized health center implies the presence of a physician. Three health areas in Ndop meet the inclusion criteria with a ≥ 40 equity score and a medicalized center, namely Bambalang, Babesi, and Balikumbat. However, considering that the study requires two health areas, one for intervention and one for control, Balikumbat (equity score 42) will be excluded because it has a lower equity score than Bambalang (equity score 47) and Babesi (equity score 47). Bambalang and Babesi will be further randomized into a control and intervention health area via computer number generation. In the Bali district, Bosa and Nakka had the highest equity scores and had a medicalized center. After stratification, a health facility assessment will be conducted to ensure the quality of health facilities for the intervention using the WHO Service Availability and Readiness Assessment (SARA) tool [42]. Health facilities will be assessed using this tool based on the context of Cameroon to ensure the quality of the services to be delivered and to ensure that all facilities are of equal standard and have the minimum required services to deliver the intervention, followed by the randomization of health areas into intervention and control groups using computer number generation (Fig. ​(Fig.22). Graphical presentation of CRCT