MotoMeds: Telemedicine and Medication Delivery Service

Our solution is working towards solving the problem of unnecessarily high rates of childhood mortality due to lack of access to primary healthcare for common childhood illnesses in resource challenged settings. In 2021, there were 5 million deaths in children under five years. More than half were considered to be avoidable with access to established prevention and treatment methods. Globally, the under 5 mortality (U5M) rate was 38/1,000 live births, however there is great disparity between countries. For example, the 37 countries classified by the World Bank as ‘fragile and conflict affected’ had an U5M rate of 75/1,000 liver births, triple the rate of all other countries. The three most common causes of death in children 1 month to 5 years are malaria, acute respiratory illness, and diarrheal disease. Early treatment with anti-malarial medications, oral antibiotics, and rehydration solution can reduce mortality associated with these illnesses by 99.7%, 32%, and 93% respectively. However, families face significant barriers to accessing timely care, including physical access, poverty, and availability, especially in resource challenged settings or fragile contexts. When families are unable to access care illnesses that were once treatable with simple solutions transition into emergencies that require more complex care that is often unattainable due to financial and/or access barriers.

Our solution is being built and evaluated in Haiti, classified as having ‘high institutional and societal fragility’. Political instability rooted in century old problems accelerated in 2018. The absence of a stable government presence made way for gangs and private interest groups to take a stronghold on large portions of the capital (May 2023 estimates are 80%). Haiti is a very centralized country and almost all government offices and activity, institutions of higher education, and the international airport are located in the capital, Port au Prince. The city is also the central point of most of Haiti's economic activity with the majority of imported good (including fuel) arriving there and a significant portion of national commerce either originates arrives at or passes through. 

The 'takeover' of Port au Prince has led to an almost full country wide collapse of public and private services. This includes the healthcare system,  that was already tenuous before the current events. In the capital some of the most well known healthcare institutions had to close or reduce their services because of violence, kidnapping of staff and medical personnel, severe fuel shortages, and excessive danger to patients who travel to the facility. Healthcare institutions outside the capital have also been closing and altering their services due to supply chain disruptions, severe fuel shortages, and an inability for medical personnel to physically travel to their places of employment. 

MotoMeds is a pediatric telemedicine and medication delivery service designed to fill the gap in access to pre-emergency care during ‘off hours’, defined as the time period when local clinics are closed, typically nights and weekends. 

The TMDS model involves a centralized call center where healthcare professionals receive calls from families with children who are acutely ill. A TMDS healthcare provider screens for eligibility; the TMDS currently serves children 0-10 years. The provider then uses evidence based clinical decision-support tools to triage the child as mild, moderate, or severe (life-threatening). The tools have recently been digitized into the MotoMeds 4.0 application for use on Android devices. If the triage is severe, the child is referred to seek hospital level care. If the triage is not severe, the provider gathers basic clinical findings and medical history from the parent over the phone to generate an assessment and treatment plan. If the child lives within a delivery zone, a TMDS driver is dispatched to transport medications/fluids to the child’s home. In some instances where the patient is ‘at risk’ (typically triaged as moderate) a provider accompanies the delivery to perform an in-person exam. Families outside the delivery zone or with illnesses beyond the clinical scope of MotoMeds receive consult alone. All families receive a follow-up call at 24 hours and/or 10 days to ascertain current health status and whether care was sought.  

MotoMeds was launched in Haiti in September 2019 with both the call center and delivery zone located in Gressier, a semi urban and rural region located approximately 30 Km west of Port au Prince. In response to the need for increased healthcare services following the 2021 earthquake in Les Cayes we opened a satellite delivery zone in the urban area of Haiti's 4th largest city. The service is operational from 6pm to 5 am seven days a week, 365 days per year. 

The initial rationale to function during the nighttime was because of the complete lack of options for pre-emergency care during this period to avoid disrupting the network of public and private daytime providers. However, as the situation in Haiti deteriorated we realized that working at nighttime allowed us to operate when daytime providers were forced to shut their doors for security reasons. During these periods of 'lockdown' or protests families were unable to leave their homes during the day to visit their local clinics, leaving MotoMeds as one of the only non-emergency healthcare services that was operational. We soon realized that the TMDS model was very versatile and operating at night with a minimal physical presence was ideal for fragile situations in conflict. 

Speaking to the versatility of the model, we launched a MotoMeds TMDS in Ghana in December 2022 in partnership with their National Ambulance Service. The objective was to assess the portability of the model to other settings and to learn how we can incorporate malaria into the clinical workflow. To date, this pilot expansion has met expectations.

Our solution serves families with children who are acutely ill with common childhood illnesses. These illnesses include fever, respiratory complaints, dehydration, vomiting, diarrhea, skin, ear and urinary tract infections, as well as some other complaints such as parasites. MotoMeds is operational in the cities of Les Cayes and Gressier Haiti and the community of Jamestown/Usshertown in Accra Ghana. These communities currently do not have access to pre-emergency healthcare at nighttime. We know this because we conducted needs assessments in both locations and once clinics are closed families are isolated from advice and care. It is also very rare for them to have ‘over the counter’ aids such as paracetamol (Tylenol) or oral rehydration solution (Pedialyte) stocked at home.

MotoMeds impacts families by 1) preventing common childhood illnesses from transitioning into emergent situations, 2) enabling families to address their pediatric healthcare needs early when they can be treated with affordable solutions, 3) literally engaging families in the care of their children by asking them to participate in virtual exams, 4) eliminates the transportation barrier that many vulnerable parents face to access care.

Each of our team’s members offers a unique perspective and approach to designing and building our solution. The Team Lead had lived in the community where MotoMeds operates for 12 years, six of those years with young children who were occasionally in need of healthcare. She had many physician contacts in the healthcare community due to the nature of her work and would reach out to them for virtual care when her kids were sick.  This was a convenient, affordable, effective, and low stress way to manage her children's’ illnesses. She recognized that most families don’t have the contacts to access healthcare in the same way. 

A series of human centered design initiatives and a formal needs assessment were conducted within the communities where MotoMeds now operates to learn if a lack of access to healthcare was actually a problem for the community, and to understand possible solutions. As the idea for a TMDS was solidified and the project was built formal qualitative research studies on TMDS providers/staff and user feedback have remained regular aspects of the work. Concrete examples of how feedback from staff and employees has shaped the development of the solution have included; the hours of operation, whether to charge a fee for the service and how much to charge.  

The MotoMeds TMDS is evolving through a series of research studies in Haiti (H) and Ghana (G) called INACT (Improving Nighttime Access to Care and Treatment) with the following objectives: INACT1-H: understand the need through identifying healthcare seeking behaviors (completed 2019), INACT2-H: determine clinical safety and feasibility of implementing a TMDS in a resource-limited setting (completed 2021), INACT3-H: assess scalability of the TMDS by conducting 80% of patient assessments by phone only and servicing two geographically distinct delivery zones with a single call center (completed 2022), INACT2-G: testing the portability of the TMDS model to a disparate setting with high rates of malaria (Ghana), and INACT4-H: build and evaluate a digital clinical decision support tool to improve guideline adherence and workflow metrics (in progress 2022-2023). In order to accomplish our research goals we have loosely controlled call volume. However, from September 2019 to May 2023 we have provided virtual advice and care to the families of over 2370 children, performed 567 in-person exams, and completed 1569 medication/fluid deliveries.

Our team has extensive experience in research, pediatric clinical care, software development, and project management and evaluation. However none of us have experience as social entrepreneurs and we lack the wisdom and direction to efficiently and effectively grow our solution. Our most pressing need is to develop a business model as we transition from research based activities to expansion and growth. We are also in search of coaching on networking strategies and outreach methods.

In summary our team feels that we have created a life changing product but we don't know how to effectively disseminate it.

Our solution addresses the problems of high and disparate rates of under five mortality and lack of universal access to healthcare through implementation of a nighttime telemedicine and medication delivery service. The idea for the solution is innovative because it combines the following; 1) telemedicine coupled with medication delivery to reduce healthcare access barriers, 2) a focus on preventing pre-emergent illnesses from transitioning into emergencies during 'off hours' when typically the only care available is emergency care, 3) uses technology to promote the uptake and adherence of evidence based clinical guidelines, 4) the versatility of hte TMDS model in terms of how, where, and when it can be deployed.

There are many examples of real world use cases of the individual components of our solution, however to the best of our knowledge MotoMeds is the first program to integrate them into a single product that is being systematically evaluated and iterated through a series of research studies.

Individual components.

The MotoMeds clinical guidelines are based on the universally recognized and widely implemented World Health Organization Integrated Management for Childhood Illness (WHO IMCI) guidelines. To the best of our knowledge these guidelines have not been previously adapted for telemedicine use. 

Many community health worker programs use digitized versions of the WHO IMCI guidelines to promote their uptake and adherence. However, these use cases involve in-person interactions and not telemedicine. (Example- ThinkMD)

There are telemedicine services operating in low-resource settings but their services are restricted to adult patients (Example- Babyl Rwanda) or they do not offer medication delivery at nighttime. (Example- PendaHealth Kenya).

There are pharmacies in low resource settings that offer 24/7 delivery but they are not coupled with evidence based telemedicine. 

Versatility.

The TMDS model has established proof of concept in the following circumstances; 1) in rural mountainous areas, 2) in densely populated urban areas, 3) at nighttime, 4) in two separate countries with different healthcare systems and cultures, 5) as part of the response to a natural disaster, 6) within the context of extreme insecurity. We have not yet implemented daytime deployment but considering the difficult circumstances where implementation was successful, daytime deployment should also work.

In terms of having broader impact on those working to improve the provision of pediatric healthcare our team believes it is important to acknowledge that children become sick at all hours and that waiting until the morning could be waiting too long. We also believe telemedicine coupled with medication delivery is a very effective mechanism for overcoming healthcare access barriers related to transportation, distance, security, competing household needs, limited finances. Hopefully these ideas could resonate with others who learn about the MotoMeds TMDS model of healthcare delivery.

1- year impact goal

Completion of a rigorously evaluated clinical decision support tool that covers common pediatric complaints and can be used offline for virtual or in-person exams.

We are currently testing the alpha prototype of our clinical decision support tool in an interrupted time series study and through qualitative means with end-users. Over the next 6 months we will use the research findings to iterate the tool and create a beta version, at which point we will be ready to share the tool with others. 

5-year impact goal

1) Have an established open access toolkit including a digital clinical decision support tool accompanied by a framework for medication delivery that facilitates healthcare institutions to incorporate a TMDS within their existing healthcare infrastructure. 2) Have deployed the toolkit to at least 15 organizations with sustained use of at least two years.

Our team's end goal is not to implement the MotoMeds TMDS model at the global scale ourselves, but to facilitate others who provide pediatric healthcare to adopt the model and integrate it into their existing services. We can envision this happening at the level of a country wide deployment by a Ministry of Health, at the organizational level of an international NGO, or by local institutions (private or NGO supported). To do this we are working on creating a 'transferable package' that contains all of the components for adopting the model; the clinical decision support tool, the logistics framework, training materials, publicity materials, implementation guide, etc. 

Because our team is building and evaluating the solution within the context of research our indicators are a bit different than if we were solely providing a healthcare service. Some of the indicators we are tracking are shared below.

# of nights the TMDS is operational. We measure this to assess the robustness of the operational structure. In Haiti an increasing number of healthcare institutions have been closing or modifying their services over the past couple years because of the worsening security and infrastructure challenges. It is important that we are able to provide un-interrupted service to the families who rely on the clinical care and the staff who rely on the employment income. 

Call duration. We measure this because we want to provide care as quickly as possible and we do not want to burden the caller whose cell phone connection and battery charge is tenuous.

Time to delivery. We measure this because we want to provide care as quickly as possible and our goal has been to reach families in under 2 hours, as we are confident that the clinical picture is unlikely to change during this timeframe. 

Guideline adherence metrics. We measure this because we have gone to great lengths to create solid clinical resources and it is important that we deliver on them, even amidst the high stress environment. Providers have the ability to deviate from the guideline as there are many 'gray' areas in clinical care, however we have set parameters for these types of deviations that are also important to adhere to.

Clinical status at 10 days. We place follow up calls to all users/study participants to query about the child's clinical status at 10 days post care. We want to learn about any adverse outcomes

Qualitative feedback at 10 days. During the 10 day follow up call our team also asks users for feedback on the service and delivery to learn about any abnormalities, or user concerns that could help us improve.

Our theory of change is based on the idea that children need early access to simple and cost-effective treatments for common childhood illnesses to decrease the need for emergency services and to prevent unnecessary morbidity and mortality. In many resource challenged settings, traditional methods of providing families with basic healthcare are limited especially during the nighttime hours, forcing families to seek unqualified care or wait until the morning. A pediatric telemedicine and delivery service fills this nighttime gap in pre-emergency healthcare access. The delivery model overcomes the distance barriers to accessing healthcare that are common in these settings. In the specific deployment of Haiti nighttime telemedicine allows us to bypass security barriers that present during the daytime.

The technology that powers our solution is an electronic clinical decision support tool. The architectural foundation draws from the ‘Outbreak Responder; Rehydration Calculator’ we built previously for the management of diarrheal diseases. The existing architecture for diarrheal diseases was broadened to include the six most common pediatric chief complaints that are within the scope of a TMDS. Five input pages are navigated via a stepper flow interface to enter data into the calculator. The user then presses ‘calculate’ to generate an output page with a case summary (weight, severity, diagnosis, danger signs), a fluids and medications section, a disposition (hospital vs household) and a follow-up plan. The child's weight is estimated from the WHO 50th percentile weight for age. The output page is configured for printing or forwarding for to facilitate the in transmission of treatment plans. 

The tool is designed to be used offline and is agnostic to mobile and web interfaces using Flutter. Flutter represents an opportunity for the mHealth community because it enables a cross-platform application in a single codebase for Android, iOS, Linux, macOS, Windows, Google Fuchsia, and web browsers

We have established that the paper-based clinical decision support tools used to create the electronic clinical decision support (eCDS) tool are effective through a congruence analysis (manuscript out for review and posted on medRxiv). In addition we established the larger TMDS workflow is feasible to implement and meets basic safety standards (Journal of Pediatrics). 

Validation of the TMDS model and workflow.

 It was important to formally evaluate the clinical guidelines and the workflow to answer the following questions.

• Is the TMDS implementable?
• Will the target population want to use it? 
• Can untrained parents adequately relay their child's clinical symptoms over the phone?
• Is the TMDS clinically safe? 
• Is the TMDS clinically effective?
• Will users be satisfied? 

Validation of the clinical guidelines.

In brief a prospective cohort study was conducted at a TMDS in Haiti; children ≤10 years were enrolled. Among non-severe cases, paired virtual and in-person exams were conducted at the call center and household. The primary outcome was the performance of the virtual exam compared to the in-person exam (reference standard).

A total of 391 cases were enrolled. Among 320 cases with paired exams, no general World Health Organization (WHO) danger signs were identified at the household; problem-specific danger signs were identified in 6 cases (2%). Cohen’s kappa for the designation of mild cases was 0.78 (95%CI 0.69-0.87). Among components of the virtual exam, the sensitivity and specificity of a reported fever were 91% (95%CI 87%-96%) and 69% (62%-74%), respectively; the sensitivity and specificity of ‘fast breathing’ were 47% (21%-72%) and 89% (85%-94%), respectively. Kappa for ‘no’ and ‘moderate’ dehydration indicated moderate congruence (0.69; 95%CI 0.41-0.98). At 10 days, 95% (273) of the 287 cases reached were better/recovered.

This study, and resulting guideline, is a formative step towards an evidence-based pediatric telemedicine guideline built on WHO clinical principles. However, the study exposed limitations in the paper decision support tool. The tools were cumbersome, likely contributing to a long call duration of 20 mins (range 3-60mins) and avoidable guideline deviations. Long call times hindered response to subsequent calls, burdened the caller who pays by the minute, compromised battery charge on the caller’s phone, and delayed time to delivery. We concluded that an eCDS that digitizes the clinical guidelines and includes a medication dosing calculator, would expedite the workflow, increase guideline adherence, and improve operational outcomes (e.g., reduce call duration, time to delivery).

Validation of the eCDS.

We have built an alpha prototype of the eCDS tool and are currently undergoing testing with users (TMDS providers) before deploying the tool in an interrupted time series study. The study will compare rates of guideline adherence (primary outcome) and logistical metrics (call duration) among providers using paper (control) vs electronic (intervention) decision support. The 6-month pre-intervention period is ending shortly and we are preparing to transition to using the eCDS tool for the 6-month intervention period. 

Our solution has been designed, built, and evaluated within Dr Eric Nelson's lab at the University of Florida. However we are quickly outgrowing academia and are finishing up our remaining research studies as we undertake the process to create a stand alone NGO registered in the US.

Our leadership team is comprised of researchers, clinicians, educators, project managers, and student volunteers. We have attempted to include  members of the business community in our leadership team but have yet to find success with this. 

Due to the fact that we are currently working in Haiti we actively recruited Haitian team members to advise clinically and culturally on the project. The project in Ghana just recently launched a few months ago but has taken the same approach. The operations in Haiti are run by a Haitian team, the clinical advisory committee is 50% Haitian and our research partner institutions are Haitian. Haiti does not have a strong research infrastructure environment so our team at the University of Florida (UF) has been very conscientious to build capacity amongst Haitian researchers.

Within or implementation team we promote the voices of all members by conducting regular feedback sessions with all levels of employees in an impartial setting. For example we just completed over 5 hours of focus groups with MotoMeds delivery drivers that were facilitated by two Haitian graduates of UF. The anonymous results are currently being synthesized by other team members to understand how we can better facilitate their work and improve the service for their community.

In terms of ensuring the design and implementation of the TMDS is inclusive of TMDS end users, families, we conducted many sessions using human centered design methods to understand the communities priorities and preferences. 

Our business model is based on the premise that children become ill at all hours and providing them with relatively inexpensive pre-emergency care in a timely manner will prevent the transition to a more complicated and costly emergent illness. This premise benefits the patient, the family, the community, and the healthcare system.

Patients and families- It goes without saying that when children are sick, they prefer to receive treatment as soon as possible. Families benefit from the remote nature of the service and can bypass the physical access barriers associated with seeking care- transportation access and costs, competing household priorities, and security. Families are relieved of some of the stress of having a sick child when they know there are medical professionals they can contact at any hour of the night for advice. Treating childhood illnesses early can be a significantly less complicated and less expensive experience for families to navigate.

Community and healthcare system- Reducing emergency illnesses conserves scarce emergency resources and personnel for situations that are less preventable (e.g., trauma). By providing care during off hours, early morning patient volume at area clinics can be decompressed and again resources can focus on more complex medical problems.

Our team conducted a formal cost-effective analysis (published at The American Journal of Tropical Medicine and Hygiene) comparing the TMDS to the only alternative to nighttime care in the study area, emergency services, and the results suggest that the TMDS is a cost-effective solution.

We will be undertaking two different activities that will likely have different paths to financial sustainability

Direct implementation in Haiti. We plan to continue implementing MotoMeds in Haiti while iterating the model through research into the near future (approximately 2-4 years). As long as we are conducting research, we will likely seek funding from traditional research mechanisms (e.g., The National Institutes of Health). Our goal in Haiti is to eventually nationalize MotoMeds in partnership with the Ministry of Health (MoH). This would likely be accomplished with a shared funding scheme derived from the MoH, TMDS user fees (this would generate a small fraction of costs), the Haitian diaspora, Haitian private sector partners (e.g., Digicel telecommunications), and possibly some grants and/or private donations.

Facilitation of TMDS adoption. We do not foresee our team directly implementing additional MotoMeds deployments other than the current deployments in Haiti and Ghana. Instead, we see our team taking on the role of consultants to initiate the adoption of the TMDS model by governments, international NGOs, or local private and charitable healthcare facilities. We will need funding to fine tune tools, build training materials, provide support, and . At the moment, our plan for funding this work will be through grants and private donations.  However, the main deliverable that we would hope to get out of a Solve partnership is assistance to explore creative and alternative business models and business plans.

From 2018 to present, we have funded our research and implementation in Haiti through grants from The National Institutes of Health, and donations from private individual and charitable organizations. The expansion to Ghana has been funded by USAID and in-kind support.