Using a Mobile Phone-based Multimedia Technology to Support Maternal Health in Rural Southwestern Uganda

Conceptual model development: Between June 2018 and November 2018, 14 pregnant women and 6 healthcare professionals in the antenatal clinic in Mbarara Regional Referral Hospital (MRRH) will be recruited. Focus group discussions will be used to assess the optimal intervention in terms of the preferred content of multimedia messages, length, number, and strategies to make messages appropriate to illiterate population and protect confidentiality and patient privacy. The Investigators will also assess the predicted barriers to appropriate response to the multimedia messages, preferences and perceptions of using multimedia messages, understanding of messages, barriers to adoption of the multimedia prototype. Additionally, the investigators will explore the prevailing maternal health challenges experienced and how multimedia applications can best address the identified challenges. The output from the focus group discussion will be a conceptual model for content development, which will guide the development of the prototype. Dr. Musiimenta, in collaboration with the Research Assistant Mr.Wilson Tumuhimbise will take a lead in facilitating the focus group discussions in collaboration with all the team members. The purpose will be to inform the development the multimedia application, therefore this phase will be done concurrently with prototype design and development phase. Specifically, there will be prototype enhancement after the focus group discussion. A manuscript will be written documenting a conceptual framework for using a mobile phone-based multimedia technology to support maternal health in Uganda. This manuscript will be published in a peer reviewed health-informatics related journals.

Prototype design and development: After getting research ethics approval presumably in May 2018, the investigators will start on designing and developing the mobile phone-based prototype in June 2018, which will be completed by November 2018. In collaboration with other team members, Ms Jane Katusiime and two Masters students will carry out the development/enhancement of the prototype, supervised by Prof. Pinkwart and Dr. Musiimenta. The prototype will be developed using Java which is the default development language for android, while the database will be built using SQlite database-an open source stand-alone SQL database which will securely host the multimedia messages. Prototype refinements will include making it suitable as a standalone application (i.e., also functional in case of no network access) and making it more appropriate to illiterate women by incorporating multimedia video and audio messages. Due to the high costs of internet services and the limited financial capacity of the participants, the prototype will be made a stand-alone application that can also run without internet connection (while of course maintaining the communication functions that work only in case of network connection). The development of the prototype will follow an iterative and an interactive approach involving 20 women and healthcare providers in focus group discussions as described in the conceptual development work package above.This approach will facilitate the gathering of design/user requirements and testing for usability and functionality. The prototype will be re-designed until the investigators are sure that users are comfortable with it. Generally, the prototype design and development will follow a user centered approach which emphasizes involvement of users in all phases of the design process in order to come up with a system that is useful, usable and acceptable to the users. The research team will hold weekly meetings to review the developed prototype and suggest potential refinements if need be.

Formulation of multimedia messages: Between June 2018 and October 2018, multimedia messages will be developed by Quadral Products Uganda Limited (see letter of support), a multimedia development company with extensive experience in multimedia development in rural southwestern Uganda. The messages will be developed in collaboration with the prospective messages recipients, healthcare practitioners in the field of maternal health, and a nutritionist. Messages will cover relevant maternal health topics such as identifying danger signs, nutrition, birth preparedness, and attending antenatal care which will be based on the Ugandan Ministry of Health approved maternal health-related messages. The prototype will also have a functionality for setting antenatal visit reminders. Our MatHealth project team member-Dr. Mugyenyi Godfrey who is an experienced gynecologist and obstetrician (working with Mbarara Regional Referral Hospital (MRRH) will coordinate message development in consultation with Richard Ninsiima a qualified a nutritionist working with MRRH where the MaHealth project will recruit its participants. The development of the multimedia messages will also be guided by the conceptual model developed from the 20 participants as described in the conceptual development phase above. Messages will be developed in the local language (runyankole) of the prospective users, who will be heavily involved in message development e.g. as video actors in order to facilitate ownership, adoption, and usability. Each message will not exceed one minute in length and in total approximately 15 audio and 15 video messages will be developed. The multimedia video messages will be of quality mp4 ranging between 3 and 8mb per second. The messages will be compressed to reduce their size for easy transmission and storage on the memory cards of relatively low cost smart phones with limited processing power.

Implementing the prototype in practice: The study will be conducted at the antenatal clinic in Mbarara Regional Referral Hospital (MRRH) in Mbarara, Uganda, which is located approximately 268 kilometers southwest of the capital, Kampala. MRRH is the largest hospital in rural southwestern Uganda, which handles about 800 deliveries per month. The investigators will test the prototype with 80 pregnant mothers randomly assigned to two conditions: 1) the intervention condition, and 2) the control condition. 80 pregnant illiterate women receiving antenatal services at MRRH will be recruited in the study between December 2018 and February 2019. It should be noted that these women will be different from the 14 women involved in the conceptual model development phase above.

For 9 months, each of the 40 women in the intervention condition will use their phones to access the multimedia-based antenatal videos and audios tailored to their pregnancy stages. The 40 women in the control condition will receive the usual standard care of antenatal check-up and counselling but no multimedia intervention. Participants in the intervention condition will use the prototype until six weeks after giving birth. The intervention implementation is estimated to take 9 months including a 6 weeks follow-up after delivery, which implies that the implementation will be completed by November/ December 2019. The six weeks follow-up is important since maternal morbidity and mortality relate to illness or death occurring during pregnancy or childbirth, or within 1 to 2 months of the birth or termination of a pregnancy. Recruitment will be ongoing until all 80 participants are enrolled. All study subjects will be reimbursed up to of 2 Euros which is the standard remuneration rate for research participants in Uganda. This will be done to compensate for their time during data collection, and is often given in form of sugar or soap or rice. Participants will also be given up to 5€ to refund their transport if they come to the research offices for study-related activities. The investigators will install the final prototype on a phone and give each participant in the intervention group a phone with a prototype installed, and explain how it works. Women will get support in resolving technical issues that may arise with the application between March and May 2019.

Data collection for prototype evaluation: Between January 2019 and December 2019, the investigators will collect data for prototype evaluation which will follow a three-phased approach: baseline-assessment, midline assessment and endline assessment. During pre-assessment phase, the investigators will collect the following data at enrollment, which will be used to both track participants and to understand factors that may influence their maternal health: Age, education level, pregnancy stage, date of birth, phone numbers, home address, next of kin, cellphone ownership and use practices, stigma, socioeconomic status, and alcohol use

At endline assessment which is anticipated to take place six weeks after a participant has given birth, the investigators will again administer the questionnaire for capturing participants' maternal-health knowledge, attitudes and behaviors to both the intervention group and the control group to see if there are significant differences in the two groups. Additionally, the investigators will assess participants' experiences of using the multimedia application in order to document the acceptability and feasibility of the intervention and also their experiences in using the intervention through qualitative in depth interviews. Acceptability and feasibility data will be collected drawing from the UTAUT (Unified Theory of Acceptance and Use of Technology) model (Venkatesh et al. 2003). This will include data about the perceived usefulness of the intervention, perceived ease of use, social norms, and facilitating conditions.

Additionally, the investigators will consult participants' maternal health records at the MRRH for maternal-health related outcomes such as the number of maternal and newborn deaths, number of missed antenatal, number of maternal complications, and number of women delivering at MRRH.