Connecting the unconnected Slum dwellers

There is a significant knowledge gap in the health status and medical history of the underserved population living in urban or peri-urban areas of Bangladesh. Lack of proper hygiene awareness also puts a large percentage of the overall population facing a high risk of contracting COVID-19, and makes it increasingly difficult for relevant authorities to make informed decisions while crafting policies on health safety and region-wise mobility of this population. 

We will offer a two-step solution via the use of a mobile app. Firstly, the app will facilitate the collection of data regarding the health status of urban slum dwellers. Secondly, the data will be analyzed to find out the percentage of the population that is most at risk of contracting COVID-19. This data will ultimately help tracking and monitoring urban slum regions to make informed decisions on possible mitigation plan

People living in congested slums and low-income settlements are particularly vulnerable during any pandemic given the lack of adequate WASH facilities and hygiene awareness that have to be maintained to mitigate the situation. Moreover, in the congested slum context, it is challenging to identify and track the percentage of the people infected or at risk of being infected by a contagious disease, and take necessary steps to quarantine individuals or isolate potential patients. The informal nature of slums limits the government’s ability to reduce the risk of further spread within and beyond slums. Lack of any suitable digital platform to track, analyze and store COVID-19 related household level information to support and facilitate government’s decision making for responding is also absent. However, a prompt, and supportive collaboration between the government, citizens, and health experts, along with international assistance, can enable the country to minimize the impact of the pandemic. Our  solution will bridge this very gap, between the lack of a proper method of collecting data, identifying those at risk, spreading knowledge in the slum areas - all powered by a strong technological platform, and intervention of relevant authorities to take prompt decisions to minimize the impact of the pandemic.

The solution package jointly developed by WaterAid Bangladesh and SELISE, brings in expertise of two diverse yet vision aligned organizations. From the onset, the COVID-19 spread in Bangladesh noted that the low-income communities are often neglected when it comes to collecting relevant data about the pandemic as a whole. Despite representing 55% of the total urban habitat, these COVID-19 hotspots are often ignored as a source of data which leads to poor decision making. To mitigate this a two-step technology solution is proposed. Firstly a web and mobile application has been developed which will track patients with COVID-19 symptoms, their movement and track their previous health record. These data would be stored in a central database. In the second step the data collected would be analyzed to help inform projection of the pandemic and enable public health experts to make decisions about which areas should be brought under lockdown, or how many patients need to be relocated into isolation units. This mobile based application would also be used by health clinics as a “passport” of the service recipient based on a unique code or a QR code which would indicate about the most recent health status.

he solution package jointly developed by WaterAid Bangladesh and SELISE, brings in expertise of two diverse yet vision aligned organizations. From the onset, the COVID-19 spread in Bangladesh noted that the low-income communities are often neglected when it comes to collecting relevant data about the pandemic as a whole. Despite representing 55% of the total urban habitat, these COVID-19 hotspots are often ignored as a source of data which leads to poor decision making. To mitigate this a two-step technology solution is proposed. Firstly a web and mobile application has been developed which will track patients with COVID-19 symptoms, their movement and track their previous health record. These data would be stored in a central database. In the second step the data collected would be analyzed to help inform projection of the pandemic and enable public health experts to make decisions about which areas should be brought under lockdown, or how many patients need to be relocated into isolation units. This mobile based application would also be used by health clinics as a “passport” of the service recipient based on a unique code or a QR code which would indicate about the most recent health status.

Pandemic and yearly disease outbreaks such as dengue in Bangladesh puts millions of people at risk. These problems prevail in Bangladesh due to the absence of an effectively designed health information collection mechanism, especially from the most severely affected low income communities. The proposed application offers solution of collecting information which would fit into a periodic health tracker, promote low cost solutions like installing a hygiene point (e.g. hand washing station), leverage the Community Based Organizations (CBOs) to collect information and most importantly work closely with the policy makers in decoding and analyzing the data for effective health management planning.

Multiple web and mobile based applications have been developed and being  launched in Bangladesh that focus on the immediate response for COVID-19. Although many of these apps provide useful data for relevant public health agencies to make informed decisions, they are yet to develop a complete system loop by which other related disease information can be captured. The innovation perspective of the proposed solution package is as follows:

1.  The application will act as a patient’s “scorecard”. The scorecard is  a matrix of an individual’s  risk of contracting the disease, versus the individual’s risk of infecting others in a community. An individual getting an overall high score will mean he/she is susceptible of contracting the disease, and therefore should be placed under quarantine. An individual scoring low can have a report, or a “passport” for further mobilization.

2. The parameters for the matrix will be set based on a few key indicators, such as  medical history, transmission routes, pre-existing conditions, immune system, age, gender, travel history, living conditions, income, etc. Data on these indicators will be analyzed by doctors and epidemiologists for giving the “score” in the scorecard.

3. Critical data will be stored in a database that can also be used for future purposes, such as making an overall stronger health policy regarding urban slum dwellers.  

The solution would also ensure preventative measures suggested by the application by means of predetermined selection options determined by national plans such as National Preparedness and Response Plan for COVID-19, Bangladesh.

The app contains a blend of data collection, self-survey for health assessment, reporting and a knowledge platform. The two-step solution package involves utilizing a mobile and web based application which would capture the health information of the target beneficiaries with the aim of populating a health system database. For feature phone users, a self-assessment can be conducted through interactive voice response (IVR) calls. Self assessment and volunteers will determine risk levels on the triage scale – red, amber, and green depending on their health conditions. Based on the data collected, the next step includes analysis of the data and tracking the patient. For instance, if the person is affected then the application would alert the nearby authority if s/he tries to move out of lockdown locations, through geo-fencing. 

Secondly, the application will maintain a patient’s “scorecard”. The scorecard will be the result of a calculated matrix of an individual’s risk of contracting the disease, versus the individual’s risk of infecting others, based on a on a few key indicators, such as transmission routes, medical history, pre-existing conditions, immune system, age, gender etc. With the help of machine and deep learning , classification algorithms will regularly update the matrix of individuals with new data collected everyday, as a form of pattern recognition. Additionally training sets will be used to improve accuracy and precision so assessment of key indicators can be generated in real time. Open data sources will also be used to enrich findings by tracking population movement and hospital findings.  

Machine learning, used in the proposed application, has been used in the techniques and tools that can help in the diagnosis of diseases. It is used for the analysis of clinical parameters and their combination, for the prognosis example prediction of disease progression, extraction of medical knowledge for the outcome research, and for patient monitoring. These are successful implementations of the machine learning methods. It can help in the integration of computer-based systems in the healthcare sector. It is also used for classification, in which an individual is placed under study in many classes. It is also used for classification and extraction of unstructured data.

The Center for Systems Science and Engineering (CSSE) by John Hopkins University in Baltimore, Maryland of the United States, on January 22, first shared an interactive web-based dashboard to visualize and track reported cases of COVID-19, in real time. It was developed to provide researchers, public health authorities, and the general public with a user-friendly tool to track the outbreak as it unfolds. Although during January all data collection and processing were done manually, from February 1, the center decided to adopt a semi-automated living data stream strategy to maintain the dashboard. The center started using data using an online platform (DXY), run by members of the Chinese medical community, which aggregates local media and government reports to provide cumulative totals of COVID-19 cases in near real time at the province level in China and at the country level otherwise.

The theory of change that we want to bring about is as follows:

Inputs: The data collected from the slum areas by applying the mobile application and record specific data

Activities: Activities will include collection of data on a ground level through the application or trained volunteers, which will later be stored in a central database.

Immediate Output: The database will be used for such activities as data analytics, projection of pandemic movement, pockets of community in need of immediate humanitarian support and similar actions. This would help policymakers and public health experts take the next steps in public health security.

Longer-term Outcomes:

1. The application will also act as a patient’s “scorecard”, which is a matrix of an individual’s own risk of contracting the disease, versus the individual’s risk of infecting others in a community. An individual getting an overall high score will mean he/she has high susceptibility of contracting the disease, and therefore should be placed under quarantine/isolation. The parameters for the matrix will be set based on a few key indicators, such as  medical history, transmission routes, pre-existing conditions, immune system, age, gender, travel history, living conditions, income, etc. Data on these indicators will be analyzed by doctors and epidemiologists for giving the “score” in the scorecard. 

2. An individual scoring low can have a report, or a “passport” for further mobilization. The “passport” will be given in a form of a unique ID or a QR code that can be validated by law enforcement/other entities as other entities after checking the individual’s potential risk factor. 

3. Validated news can be shared through the app during times of crisis such as the one we are currently facing.

The assumption is that when the steps mentioned above follows their own path and executes properly, it will enable the decision makers and end users to have a proven mechanism for determining the future of health information system

Currently, the proposed solution package has been rolled out in one low income community in North Dhaka with a plan of reaching out to 4,000 people. This current plan is a packaged solution deal which aims to bring in low income community members under the mobile application coverage by capturing their COVID-19 related health details and also other relevant WASH related infrastructure. The medium term future plan for the project within a year’s time looks to serve 3 additional low income communities by which approximately additional 30,000 people will be covered. With appropriate adjustment of the solution based on learnings captured, in the next five years, the solution aims to cover 1,000,000 people by capturing their health data. 

The project team aims to achieve two key goals in the next year:

1. To adapt the software design of the application to cope with the government’s COVID-19 response plan and

2. Adapt the software to enable government to track other vector-borne disease.

The aim is to ensure that the process followed and the design of the software made is not limited to only a few diseases. Rather the aim is to ensure that the mobile application has an impact both in scale and in systemic change. The two goals will be achieved by means of:

1. Capacity building of local community based organizations (CBOs)

2. Training on the usability of the software for Health information management specialists 

3. Advocacy and influencing for the government policy makers and 

4. Knowledge product development, research and learning

Densely populated country like Bangladesh provides rich opportunity for impacting millions of life if the government can support the roll out of the application. Since WaterAid already has MoU with important health department agencies like Institute of Epidemiology, Disease Control and Research (IEDCR), the first step to reach scale and create impact among millions is already in place

The presence of the COVID-19 pandemic is the most obvious barrier of operation at the moment. Operating and rolling out the usage of the application by volunteers is also a challenge as they would need to be in presence of potential patients while collecting data. The behavioural change expected to adopt such an innovative technology may still be in a rudimentary stage among the slum dwellers and government which may affect quick adoption of the program. 

During the days of COVID -19 in Bangladesh, implementing the project poses some risks which are outlined below:

1. Data collection in highly contagious locations: The application is aimed to be used by both smart phone and basic feature phones. However, all users may not have the same skill to report the correct information. This would require trained volunteers to get into the crowded slums to collect data. This increases the risk of contracting the virus even with PPE.
2. Data Input: Despite the correct training, inputting the correct data by the volunteers in the application could be in error. If this error is high, the ultimate data output could give a wrong reading.
3. Late adaptation by government: If the successful implementation of the model is slowly adopted by the government then there remains a risk / challenge of missing out real time data capturing. Delayed adaptation and scale up would risk the health informatics database as there would be gaps which may not be fulfilled

The project aims to mitigate the barriers by adopting the following steps:

1. Training of the volunteers on proper data management. The data entered in the application will follow a two step authentication process and reviewed by a specialist for accuracy.

2. Influencing and advocacy exercises would be conducted among the community based organizations in the low income communities to explain the purpose and ultimate benefit of the health data collection information.

3. Relevant government agencies who are already supporting the application would be used as an advocacy partner to convince the policy makers about the importance of the application.

4. The beneficiaries would also be supported with medical and treatment support if found to be affected by COVID-19 symptoms, by the other donor resources allocated for emergency response. The purpose of this is to ensure the confidence and trust of the beneficiaries on the implemented project.

5. The volunteers who would be collecting the data would also be provided with suitable personal protection equipment while visiting the slums.

A total team of 8 staff, comprised of people from both organizations, will be involved in working out the solution:

1. ICT Expert and Lead Trainer (WaterAid) - Full Time

2. Program Management and Partnership Expert (WaterAid) - Full Time

3. M&E and Knowledge Management Expert (WaterAid) - Part Time

4. Finance and Administration Specialist (WaterAid) - Full Time

5. Fullstack Software Developer (SELISE) - Full Time

6. Data Science Expert (SELISE) - Full Time

7. Deep Learning Expert (SELISE) - Full Time

8. Business Analyst (SELISE) - Full Time

Established in 2011, Secure Link Services AG (SELISE) is a multinational software development, outsourcing and IT consulting company. SELISE has development centers in Zurich, Dhaka (Bangladesh) and Thimpu (Bhutan), as well as a consulting location in Dubai. SELISE strategizes, develops and maintains digital platforms for leading businesses across the world in sectors of health, insurance, retail, financial services, logistics, consumer goods, sports, telco and industry. The company provides enterprise cloud solutions, facilitates omnichannel ecommerce and serves application management. In the last nine years, the company has grown exponentially, and employs over 300 staff, has built over 200 applications, and has served over 120 customers in 90 countries. While Dhaka and Thimphu host intensive, multi-year talent development programs, Zurich and Dubai primarily employ senior staff. 

WaterAid, on the other hand, is a global name in the area of promoting sanitation and hygiene. With the aim of promoting clean water, decent toilets and good hygiene as basic human rights, WaterAid has already made a mark in Bangladesh, especially in the time of COVID-19. The application for tracking patients of COVID-19, developed by WaterAid has been recognized through a national award given by the ICT Division of the Government of Bangladesh. We convince governments to change laws, link policy makers with people on the ground, pool knowledge and resources and rally support from people and organizations around the world, making lasting change happen on a massive scale. Proven track record of delivering results under demanding conditions makes us a popular choice among donors.

WaterAid and SELISE, in Bangladesh, have reached a contractual arrangement to promote the mobile based application to capture baseline health data in order to contain the spread of COVID-19 and any future health risks. Both the organizations have brought forward their own expertises. WaterAid Bangladesh will be providing the frontline service in capturing the data from the slums, training the volunteers and managing the overall field operation. SELISE, on the other hand will be utilizing the data captured to analyze them and improving the software’s performance in providing real time updates, which will later be used by the relevant authorities to make informed decisions and policies. SELISE would also explore the business viability of the model among interested clients.

The proposed application is aimed towards providing a large scale impact for the under served slum dwellers of urban Bangladesh. The application has already been developed using resources generated by WaterAid Bangladesh’s own fund, donor’s contribution and in kind support from SELISE. Hence no additional fund is required to develop or revise the application itself. 

Additional fund if available would be directed towards expansion into other regions, training of volunteers, purchase of devices, pay for data collection and influencing and advocacy meetings and development of knowledge product and capturing critical learning. Formation of One Stop Service (OSS) and a Standard Operating Plan (SOP) is also a part of the scale up plan with the aim of supporting the government to take up the application for easier scale up.

The application will be serving primarily the government of Bangladesh and low income community members. But the data analytics part of the project, developed and created by SELISE could be packaged and sold for generating revenue to hospitals, health insurance service providers, airlines and other interested private agencies. In addition to the private agencies, the solution can also be packaged and sold for  revenue to the government. It is anticipated since the government of Bangladesh does not have any digital platform to trace and track health information, it would be a timely and suitable revenue generating avenue for the project

The joint organizations plan to sustain their business model by following these revenue models:

1. Sustained grants -  WaterAid, being a key development partner of the Government of Bangladesh and bi-lateral development partners, have access to grant revenue every year. It has already allocated committed funds for the pilot phase and explored other funding and grant opportunities to finance the project.

2. Service subsidization - The data analytical model, with all the collected and analyzed data in an anonymized form, will be sold to healthcare centers, private medical facilities, government bodies, diagnostic centers, research institutes, universities, non-government organizations, and other external markets to help fund the project. SELISE, which is a for-profit organization, will be in charge of selling the solution. However, it will also fund the project as part of a social program. WaterAid, a non-profit organization, will be funding the project solely based on sustained grants. Key success factors include leverage intangible assets, such as the huge pool of collected information, technological expertise, research findings, and others, 

3. Campaigning - A social media/promotional campaign will be run, like a Kickstarter campaign, to ask for donor participation and fundraising for the underserved community. 

The health data collection mechanism in Bangladesh, especially in the low income communities is still a very weak structure. The system's weakness has been brazenly displayed during the COVID-19 pandemic where lack of data has made the policy makers struggle to come up with realistic solutions at the right time when time is of essence. Under such circumstances, it has become very challenging. Furthermore, underdeveloped and rather dealt with limited priority, the health information system of Bangladesh needed innovative ideas which would create pathways for better planning mechanisms. 

Application for this Solve grant would enable the government of Bangladesh to re-prioritise the importance of the health information system. The application developed and planned to be scaled up by WaterAid and SELISE is expected to create firm and clear evidence of integrating a strong system based approach and process to strengthen health related data management.

Although the first phase of the two-phase proposed solution is fully operational, the second phase is still in the ideation and conceptualization stage. In order for the project to take off, and have a successful launch that will leave lasting impacts in the healthcare sector of Bangladesh, it is important to receive intellectual support and mentorship for effective campaigning of the project, along with future funding and revenue model establishment. Navigating through the often grey areas of regulatory matters, especially in a global context, would also benefit from the right kind of advice, mentorship or partnership. 

Like minded global health informatics and IT based platforms developing programs to collect health data would be relevant partners of the project. In addition, investors interested to capitalise on the model to promote into a business would also be potential partners. MIT Sloan Health Systems Initiative of Healthcare Analysis would be the perfect platform to test, evaluate and generate evidence for future data collection and management processes. These partners would help to improve the current status of the project and provide ideas for improved project performance on an even greater scale.

The mobile application will also contain a knowledge platform, through which WaterAid can instruct and alert people on WASH methods, as a trusted partner of the government of Bangladesh as well as Bilateral and multilaterla donors. Urban slum dwellers severely lack the awareness regarding the steps that need to be taken to stop the contraction and infection of COVID-19. However, the congested, densely populated slum areas serve as a prime hotspot for COVID-19 spread. Therefore, it is imperative to maintain a proper hygiene and sanitation routine to take effective measures against the spread of infection. The knowledge platform built inside our app, and also IVR response in feature phones, will offer an opportunity for the slum dwellers to learn, respond and follow proper WASH methods. At the same time, our trained volunteers/CBOs will also work towards building awareness and driving actions to solve this problem. Ultimately, we hope to create a better understanding of hygiene, sanitation and waste disposal among the underprivileged communities so that such a change in their collective behaviour can prevent the onset of the current and any future pandemic, and ensure higher health security. 

Therefore, the Elevate Prize will greatly help us to scale our work, impact more lives, and inspire others to tackle issues in their own communities. 

The application can be extended as a process towards informing women and young adolescent girls about Menstrual Health. The application can be adapted to include features such as Period Tracker and Sexual health and Well Being.