This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.
Population surveillance data are essential for understanding population needs and evaluating health programs. Governmental and nongovernmental organizations in western Myanmar did not previous have means for conducting robust, electronic population health surveillance.
This study involved developing mobile health (mHealth)–based population health surveillance in a rural, low-resource setting with minimal cellular infrastructure in western Myanmar. This was an early formative study in which our goal was to establish the initial feasibility of conducting mHealth population health surveillance, optimizing procedures, and building capacity for future work.
We used an iterative design process to develop mHealth-based population health surveillance focused on general demographics (eg, total census, age category, sex, births, and deaths). Interviews were conducted with international consultants (nurse midwives) and local clinicians (nurses and physicians) in Myanmar. Our analytic approach was informed by the Systems Engineering Initiative for Patient Safety work systems model to capture the multilevel user needs for developing health interventions, which was used to create a prototype data collection tool. The prototype was then pilot-tested in 33 villages to establish an initial proof of concept.
We conducted 7 interviews with 5 participants who provided feedback regarding the domains of the work system, including environmental, organizational, sociocultural, technological, informational, and task- and people-based considerations, for adapting an mHealth tool. Environmental considerations included managing limited electricity and internet service. Organizational needs involved developing agreements to work within existing government infrastructure as well as leveraging the communal nature of societies to describe the importance of surveillance data collection and gain buy-in. Linguistic diversity and lack of experience with technology were both cited as people- and technology-based aspects to inform prototype design. The use of mobile tools was also viewed as a means to improve the quality of the data collected and as a feasible option for working in settings with limited internet access. Following the prototype design based on the findings of initial interviews, the mHealth tool was piloted in 33 villages, allowing our team to collect census data from 11,945 people for an initial proof of concept. We also detected areas of potentially missing data, which will need to be further investigated and mitigated in future studies.
Previous studies have not focused heavily on the early stages of developing population health surveillance capacity in low- and middle-income countries. Findings related to key design considerations using a work systems lens may be informative to others developing technology-based solutions in extremely low-resource settings. Future work will involve collecting additional health-related data and further evaluating the quality of the data collected. Our team established an initial proof of concept for using an mHealth tool to collect census-related information in a low-resource, extremely rural, and low-literacy environment.
Reliable and valid population surveillance data are essential for understanding population needs and evaluating health programs [
Mobile health (mHealth) has been presented by the World Health Organization as a means to mitigate some of the aforementioned challenges owing to its relatively low cost, expanding ubiquity throughout the world, and the ability to use cellular networks, which are much more widespread than broadband internet [
Myanmar presents a unique case related to development of population health surveillance tools, as its borders fully opened to foreign technological influence in the early 2010s; thus, the technological infrastructure lags behind many other low-income countries [
We partnered with a local nongovernmental organization (NGO) that served as the primary provider of health services in the region. The NGO provides health care through local clinics by training community health workers and traditional birth attendants. One of our collaborative goals was to develop a system for mHealth-based population health surveillance to facilitate ongoing needs assessments in villages and evaluation of health programs. Our NGO partner previously collected population surveillance information via paper forms, which presented multiple challenges owing to the time lag from data collection to data entry, as well as the harsh environmental conditions of the remote geography and rainy climate that impede the collection of the data forms in the rainy season. Population health surveillance information helped the NGO track general trends in population (eg, births and deaths), health needs that drive the programs they implement, and evaluate the impact of their programs on population health. Owing to the limitations related to paper-based surveillance, the NGO had never aggregated paper-based data collection in a systematic manner to understand the demographics and needs of their population. mHealth-based electronic data capture of population health surveillance information was proposed given the lower cost of mobile devices, their growing availability in the region, and their flexibility for use in settings with limited internet connectivity [
All data were collected via teleconference with stakeholders residing in the United Kingdom or Myanmar by a research team residing in the United States. Interviews with stakeholders in the United Kingdom were conducted via Zoom (Zoom Video Communications, Inc; audio and video), and interviews with stakeholders in Myanmar were conducted via WhatsApp (Meta) or Slack (Slack Technologies; audio only) owing to limitations on internet bandwidth in Myanmar.
This study aimed to translate population health surveillance forms previously collected via paper into an mHealth-based tool. Our NGO partners operated in the United Kingdom and Myanmar with clinical experts from the United Kingdom traveling to Myanmar to advise with the design and evaluation of the NGO’s health programs. Paper-based population health surveillance forms were developed by the clinical staff in Myanmar with guidance from United Kingdom–based midwife consultants.
The population health surveillance forms collected census data stratified by age, sex, births per year, and deaths per year. The census data included the total number of men and women in each of the following age categories: 0 to 5 years, 6 to 15 years, 16 to 59 years, and >60 years. Owing to the limitations of the collected paper data, this information had not been previously aggregated; therefore, there was no opportunity for high-level summary and comparison of information over time. The immediate target end users of the tool were the clinical staff from the NGO during outreach trips to villages. In the longer term, the clinical staff will train community health workers living in the villages to use the tool.
We first conducted iterative phases of semistructured interviews to select a technology platform and create a prototype of the mHealth-based census form. Our goal was not to achieve theoretical saturation of qualitative data but to iteratively complete formative data collection sufficient for developing a prototype that could be used in a pilot evaluation [
We completed 3 rounds of interviews with 5 key stakeholders as part of an iterative, formative design process, where smaller numbers of participants (ie, 4-6) were appropriate for needs assessment and design generation [
In round 2, participants viewed the prototype mHealth form and provided their feedback. Round 2 included 2 separate interviews: 1 with a nurse midwife from the United Kingdom and 1 with a nurse and physician in Myanmar. For the staff in Myanmar, forms for second-round interviews were shared via the Survey123 platform before the interview, as live screen sharing was not possible. A final design based on follow-up interviews was created before the field study. An interview was then conducted in the middle of the field studies with the same local physician and nurse from round 2 to understand how the prototype form was suited to the needs of the NGO.
All interviews focused on current practices, perceived barriers, and facilitators of mHealth data collection.
Timeline of interview and iterative design activities.
Can you tell us about how community health workers fill out the current data collection forms?
What are the positives of using the paper data collection forms?
What are some of the challenges for using the paper data collection forms?
Do you think the mobile phone app could make data collection better? If so, how?
Do you think the mobile phone could make data collection harder? If so, how?
Can you tell us more about the community health workers? For example, do most of them read or speak Burmese?
Do you think the community health workers have used a mobile phone before? A smartphone? Do you think it would be possible for them to learn if we helped train them?
Is there any more information you think is important for us to know?
We have made a census form. What do you think of it?
What can we improve?
What kinds of problems do you think your team may have using the data collection form?
What new information did you hope to learn from the population surveys?
How is the information you hope to learn going to help improve (nongovernmental organization name) programs?
All interviews were conducted by a single interviewer (NCB), with other research team members attending and participating as feasible. All interviews were audio recorded and transcribed, excluding an instance where internet connectivity issues prevented the interview from taking place synchronously, so written answers to the interview questions were provided by the interviewee asynchronously.
Our team conducted qualitative interviews using thematic analysis and a constant comparative process [
The interviews were followed by a field study to establish an initial proof of concept for the use of the mHealth population health surveillance tool in this low-resource setting. Real-time population health surveillance using the census forms developed allowed for the detection of large population changes that required further investigation, as well as allocation of resources to villages based on population. The goal of this pilot study was to establish a proof of concept for using mHealth to collect census-related information and determine potential data quality issues (eg, missingness) that could be optimized through the improvement of data collection tools or training.
The mHealth forms were piloted in villages in the Chin State during outreach visits. The region is characterized by remote, mountainous terrain, so the 33 villages included were a convenience sample located within 1 to 2 days motorbike ride from the NGO’s central headquarters.
In the selected villages visited, NGO staff members filled out the population health surveillance forms on mobile devices using a commercially available mapping and survey platform (Survey123). This platform was selected because of its ability to integrate survey-related information with detailed maps necessary to identify the villages served in this remote region. Survey123 is available via app (on Android and Apple) and via web browser. Our partners collected data using tablets and a web browser–based form. Staff members involved in data collection included trained nurses and “area coordinators” (former community health workers who had been promoted into an oversight role). Nurses and area coordinators completed the population health surveillance forms by reviewing paper-based census documentation in the villages (procured from the government and local churches). Paper-based documents often had different age categories, which required the team to add categories from the paper-based forms to suit the categories in the mHealth forms. Nurses and area coordinators then went door-to-door to confirm the census in each sex or age category, number of births that year, and number of deaths. All data were collected between August and September 2020.
Data were downloaded from the Survey123 and summarized descriptively using Microsoft Excel. The proof of concept was assessed by investigating the completeness of the fields included in the form. Summarization included the total number of villages from which data were captured, total persons accounted for, total births, total deaths, and the number of potential missing data fields (Phase Two: Pilot Field Study in the Results section).
This study was approved (20-02021442) by the Weill Cornell Medicine Institutional Review Board and conducted in 2 phases. The first phase involved a series of semistructured interviews and iterative design of an mHealth-based population health surveillance tool with key stakeholders. The second phase involved piloting the tool into a field study.
A total of 7 interviews were conducted with 5 unique participants: 2 midwife consultants and 3 staff members from the NGO. Further description of the interviewees is provided in
Description of participants.
Identifier | Participant description |
LocalNurse1 | Nurse working for the NGOa in Myanmar, in charge of CHWb training programs |
LocalPhysician1 | Physician working in Myanmar also serving as the NGO chief operating officer at the time of the interviews |
LocalPhysician2 | Physician working in Myanmar overseeing the local clinic and medical outreach programs |
MidwifeConsultant1 | United Kingdom–based midwife that had developed training and data collection for traditional birth attendants affiliated with the NGO |
MidwifeConsultant2 | United Kingdom–based midwife that had developed training and data collection for traditional birth attendants affiliated with the NGO |
aNGO: nongovernmental organization.
bCHW: community health worker.
Key themes pertained to various components of the work system, including the physical environment, organizational concerns, tasks, technology or physical tools, people (user skills and motivation), and information. In line with the traditional SEIPS model, many themes intersected across work system components [
Work system depiction for the current app. CHW: community health worker; NGO: nongovernmental organization.
The themes described herein focus on important design considerations for population health surveillance. We have presented the quotes to bolster the description of themes, followed by an identifier indicating the source of the quote. Information included in brackets within the quotes was not explicitly spoken by the participant but provides a clarifying context.
Discussion of environmental themes highlighted the need to consider access to electricity, availability of devices, and internet in low-resource settings when surveillance involved the use of mHealth:
90% [of the villages] they don’t get electricity and they depend on [a] solar system...so whenever they want to charge the phone they would plug in the solar. And not even every household gets solar...only 10 in 30 households get solar.
It was also important to consider weather and topography, and how they may impact whether resources can be transported to certain areas during these times of the year:
But in monsoon season, you know we have almost five months rainy season in our country...They rarely...travel. But when they have a better road, a better access to villages to go from here to there by walking or by motorbike, then we ask them to do that. But again, it’s quite difficult.
The organizational themes were multilevel, involving the government, cultural norms, and training support from the local organization (ie, our NGO partners). From a governmental perspective, planning was required to effectively manage collaborations with government-funded health entities. Our participants explained that successful integration would involve being aware of the formal and informal structures that exist and using both positively, for example, interfacing with government-funded midwives for maternal and childcare:
But the only problem there is that because they’re [non-NGO affiliated midwives] government-paid...so you'd have to get permission from the government...to use them somehow.
Consideration of cultural norms was viewed as a necessity. The region contains predominantly agrarian societies where community health workers provide care to other villagers during their free time when they are not farming. Therefore, it was important to make data collection as simple and efficient as possible to avoid interfering with the competing demands of their daily lives:
They have CHWs, but they are farmers, and they are just volunteer people...so they have to work for their living as well. At the same time, they have to serve the community, so it is really challenging for them.
The cohesive nature of the villages was seen as a potential motivator for buy-in with data collection if the NGO trainers could illustrate how collecting population health surveillance information impacted the health of the larger community (People-based considerations). Finally, providing in-person training and additional ongoing support, possibly remotely, was also a necessary consideration.
Participants described the need to support key tasks related to population health surveillance, including detection of problematic public health patterns, measurement of program progress or effectiveness, and promotion of safe health practices. The detection of public health problems would allow for better allocation of resources (eg, medicines and first aid) and additional training to address emerging problems, such as the COVID-19 pandemic:
Data were very helpful especially during Covid-19 outreach. It was easy for us to choose the villages based on the data form and was easy for preparation of materials based on the population they provided. As for me data is most useful in preparation of medicines and materials to distribute to the active villages. I can prepare medicines and materials based on the number of active CHWs and population of the villages.
Key desired specifications related to the technological tools included: low cost, inclusion of data security protection (eg, password protection and remote wipe capabilities), ability to transmit information to cloud-based databases, ease of use, flexibility (to edit forms as needed), and capabilities to work in the given environment (eg, protection during the rainy season and store-and-forward capabilities for areas with limited cell service):
Yes, I think using [a] mobile phone app for data collection will be a perfect choice for the...CHWs...It is easy to carry and once it is filled there is no need to worry about the loss of information. It will be done at once so time saving...The outcome of data will be more accurate. It is more simple to understand on mobile.
Person-based design needs pertained to user skills and motivations. Different user needs or skills to accommodate multiple languages, have limited literacy, and little to no experience with mobile technologies:
One of our huge obstacles has been the translation...they really genuinely don't read each other's dialects, don't always follow each other's dialects, and so on. So, again, it's not just illiteracy, it's overcoming language barriers that would make some sort of pictorial and touchscreen thing easier.
[The] mobile phone thing is very new. Last year (2019) we have got this telephone tower and people start[ed] using phones.
Another person-based theme involved understanding and leveraging potential motivators for community health work and data collection. Avenues for motivation overlapped with other work system components such as reimbursement or payment provided by the local NGO, respect for other villagers and desire to help, excitement related to new technologies (ie, mobile phones), and personal altruism. It was also viewed that connecting the data collection activities to the larger purpose of the different organizations (both the NGO and village) would serve as a motivating factor:
We also have to explain why we make them use this app...because we would like [to know about] their village and how the patients are being taken care of by them, and then it’ll be...a strong connection between the community and us [the NGO].
Information-related themes touched on data quality and data structure. In general, data quality was perceived to be higher when using the mHealth tool over paper forms as it eliminated issues with reading handwriting. Structured data entry was also viewed as helpful in reducing issues related to limited literacy, although local team members wanted unstructured data entry to account for atypical cases:
Some of the CHWs are not so educated so they do not know the exact words to fill sometimes.
Actually, here “others” is something that involves nothing that we have here...If it's out of this topic and you see something other than this or that, please describe it. That's how we taught them.
Population health surveillance form for collecting village census information. Village names have been redacted but included the village name in Burmese characters as well as the unique, numeric identifier.
Approaches implemented for the design considerations derived from interview themes. Work system domain.
Particulars and design considerations |
Approach implemented | ||
|
|||
|
Lack of electricity for charging devices |
Included solar charged batteries in budget to supplement existing solar-based systems |
|
|
Mountainous topography, transportation challenges owing to weather |
Choose a mapping software to find villages in need |
|
|
|||
|
Planning of government vs NGOa ownership of materials |
Developed a plan where NGO would purchase and own devices, developing memorandum of understanding with the government |
|
|
Need to balance data collection with demands of daily life |
Created simple-form design so data collection is as efficient as possible |
|
|
|||
|
Detection of problematic public health patterns |
Chose a mapping platform, which allows geographic tracking and data summarization |
|
|
Program measurement |
Chose a mapping platform, which allows geographic tracking and data summarization |
|
|
|||
|
Low cost |
Found a program usable on Android (ie, for mapping and survey) offering nonprofit pricing |
|
|
Flexibility to edit forms |
Used a web platform for NGO staff members to edit forms easily |
|
|
Work in the given environment |
Purchased waterproof cases for devices |
|
|
|||
|
Easy to learn and use |
Leveraged an iterative participatory design process to create data collection forms |
|
|
Multiple languages |
Included Burmese text in forms |
|
|
Limited literacy |
Included icons in forms |
|
|
Data structure (unstructured vs structured) |
Made most fields structured Included open-ended text boxes for additional notes not conforming to structured data collection elements |
aNGO: nongovernmental organization.
The prototype displayed in
Feedback from the study in the field (third round interviews) provided a proof of concept for electronic (mHealth-based) population health surveillance data collection and underscored the need for store-and-forward capabilities for villages that may not have sufficient cellular service for real-time data upload. For example, in one of the follow-up interviews, an NGO-employed nurse described the following:
So far, it's very good and it is very easy to use...I think compared to the paper form...I prefer these Survey123 survey data forms.
Demonstration of how the survey platform from Survey123 (left) integrates with the mapping platform (right).
Our formative, iterative design approach allowed us to develop an mHealth tool for population health surveillance in an extremely rural, low-resource, limited-literacy environment with minimal cellular infrastructure in western Myanmar. We first summarized our results and illustrated how they were translated into concrete next steps, as the implementation of our findings may be informative for others conducting similar work. We then described our findings in the context of the work system model. Finally, we discussed the overlap of our findings with those of other studies and highlighted how our results extend currently published knowledge.
Key design needs were organized based on environmental, organizational, sociocultural, technological, and task-, people-, and information-based constructs. Many of the design considerations derived from the interviews were addressed before the field study (
As shown in
Previous studies regarding the integration of electronic health surveillance and mHealth integration into low-resource settings are similar to ours, but others may not yet be applicable in particularly underdeveloped settings. For example, McIntosh et al [
Considerations derived from interviews implemented in subsequent efforts.
Work system domain and design considerations |
Implemented approach | |
Environmental considerations derived implemented in pilot study |
Considerations derived implemented in pilot study |
|
Developed videos for data collection forms to providing ongoing training |
||
|
||
|
Considerations derived implemented in pilot study |
Considerations derived implemented in pilot study |
|
Secure data collection and storage |
Obtained software that may be locked (with a passcode) and wiped remotely |
|
Ability to transmit information to database |
Purchased cords for data transfer so data could be stored and uploaded later in villages with poor cell service |
|
||
|
Easy to learn and use |
Iterative participatory design |
|
Multiple languages |
Added text or audio from additional local languages (eg, Mara) |
|
Limited mobile device experience |
Incorporated mobile device basics into data collection training |
|
Motivating data collection |
Connected data collection to larger purpose of the work in training Providing needed supplies to community health workers Created a data plausibility review process by NGOa staff |
aNGO: nongovernmental organization.
Our study has limitations related to (1) sample size for the iterative design process, (2) field study, (3) transferability of findings, (4) ongoing COVID-19 pandemic, and (5) military coup in the region. First, our interviews involved 5 participants and 7 total interviews, which are smaller than sample sizes recommended in some qualitative guidelines but appropriate for our approach—an iterative, formative design process [
We conducted formative, qualitative interviews with key stakeholders to develop an mHealth-based system for population health surveillance in a low-resource setting in Myanmar. Design considerations involved various work system domains including physical environment, organization-, task-, technology-, people-, and information-related issues. Members of our NGO partners were able to use the mHealth-based population health surveillance tool to establish an initial proof of concept in a feasibility study of 33 villages. Future work will involve evaluating the plausibility of the data collected and expanding data collection beyond simple census information to disease and maternal or child health outcome information. The February 2021 military coup in Myanmar presented a difficult political environment for the employment of this approach in our continued work that will also need to be addressed, moving forward regarding safety and security reasons.
mobile health
nongovernmental organization
Systems Engineering Initiative for Patient Safety
This work was supported by the Anna-Maria and Stephen Kellen Foundation and the Eunice Kennedy Shriver National Institute of Child Health and Human Development for their support. (R21HD103053; principal investigator: RMC)
None declared.