Grassroot Innovation in Humanitarian Health, From Student Work to Disaster Response: How Improved Data Collection Could Impact Austere and Transient Clinical Environments Globally
Accurate information is a foundational requirement for the delivery of health services, for planning and evaluating public and global health initiatives, and for identifying societal needs. The advent of electronic medical records (EMRs) offers distinct advantages over paper based systems: transportability, accessibility, individual and collective analyses, and protection from physical damage (e.g., loss, water damage, fire, fading).
While electronic medical records (EMRs) have been amply proven as tools for improving healthcare delivery, there is a paucity of health data collected in the aftermath of disasters (Kubo 2018; Kubo et al., 2019) and in regions lacking sufficient healthcare infrastructure, such as low and middle income countries (LMICs)(Oliver, 2020). This absence of data means that immediate decisions are not likely to be based on empirical evidence, potentially adversely affecting treatment and the course of disaster recovery (Anderson, 1999), and that analyses of contributing factors, morbidity and mortality, and individual and community needs are reduced to supposition. Previous attempts at implementing an EMR system in areas that lack connectivity involved technologies that require technical experts and plenty of bulky equipment--an unrealistic expectation at times and in many places in the world (Fayaz-Bathsheba, 2013).
It is said that “necessity is the mother of invention.” The 2010 earthquake in Haiti helped frame the obligation to collect health data in austere circumstances. Student volunteers who had deployed to Haiti recognized the absence of inter-institutional collaboration. In response, they began development of fEMR—“Fast Electronic Medical Records”—in 2011 by implementing the project into Computer Science classes at their home university. fEMR is an open-source electronic medical record system designed specifically for mobile medical teams volunteering in resource-limited environments with limited access to the Internet or electricity. The system comes with hardware that creates a closed intranet signal through which one may connect to the software using their own smartphone, tablet, or laptop. The “plug and play” nature of this equipment is designed so that any clinician or volunteer can easily set up the system in less than ten minutes, and no IT specialists are required in the field. The closed Intranet signal created by fEMR’s lightweight networking equipment allows for rapid health data storage even when all communications are down, as was the case in the aftermath of the 2010 Haitian earthquake. As networks were flooded with aid agencies trying to reach the population via SMS and citizens tried to reach loved ones, networks became congested and failed in the early days of the disaster response (International Federation of the Red Cross, n.d.). Trying to connect a traditional EMR to the Internet or use a mobile app would not have worked in this setting. fEMR software was also created with a streamlined user interface so as not to disrupt clinic flow in what is often a high-volume, chaotic, and stressful environment. After its first deployment in 2014, fEMR has been deployed in some of the most remote areas of the world, most recently in refugee camps along the U.S./Mexico border during the COVID-19 pandemic.
To fully accommodate the needs of a disaster or resource-constrained environment, however, systems should facilitate seamless transition of information from first responders to first receivers (heath staff taking in patients) (O'Neill, 2006) and subsequently to other providers, sponsoring organizations, and ministries of health. Clinical and demographic information coming from the field could serve as critical points of reference for receiving clinicians to appropriately allocate resources and engage an incident command system in the most efficient manner; a step that has traditionally been missing in mass casualty incidents (MCIs) and austerity operations (Klein et al., 1991; Mahoney et al., 2005; Feliciano et al., 1998). A strictly offline field EMR system, while conducive to retroactive data analysis, would not achieve this goal as readily. Therefore, the latest release of the fEMR system employs distributed ledger technology (DLT), a collection of data that is synchronized and shared across multiple locations (Mercy Corps, 2019), in an online version called fEMR On-Chain. The online version is built upon blockchain technology to ensure security and immutability of patient records, especially in areas with highly sensitive patient information. Blockchain technology, most commonly recognized in the context of digital currencies like Bitcoin, is a tamper-resistant way to keep ledgers in a distributed manner (Yaga et al., 2018). Records are redundantly stored with encryption, and with internal permissioning that strictly secures access. Upcoming releases will see the two systems connected, such that when communications are back up in the aftermath of a disaster, the offline version can detect internet signal, back up patient data, update software, and generally switch between online and offline versions of the EMR.
Team fEMR has been supported since its inception by volunteer developers and undergraduate and graduate computer science students from across the United States. Students from Wayne State University, the University of Texas, Florida State University, and California Polytechnic State University have contributed to software development since 2011. Team fEMR’s current California Polytechnic academic partnership has teams of students spending the entire academic year developing fEMR software. The larger goal of the fEMR system is to use the data collected by the system to build capacity in a variety of ways, depending on the environment. Therefore, in addition to the nation-wide open-source development community, there is also a nation-wide team of volunteer medical coders who assign codes from the International Classification of Diseases, 10th Revision (ICD-10) to all de-identified medical records collected in the fEMR system during and after each deployment.
Data collection in disasters and other dangerous environments allows for retrospective analyses of costs, effective patient care, better allocation of limited resources, better coordination between responding agencies, and potentially more informed policy decisions. Indeed, real-time data collection, institutional coordination, and ICD-10 coding to facilitate research are all supported by the American College of Emergency Physicians in their 2016 statement on data collection in disasters (American College of Emergency Physicians, 2016). However, fEMR has also been used in non-MCI humanitarian aid scenarios since 2014. For example, in 2019, changes to U.S. immigration policy resulted in thousands of asylum seekers living in a tent city style refugee camp in Matamoros, Mexico while they awaited case adjudication in the American court system (Stab, 2019). A few months later, the COVID-19 pandemic swept across the globe. Given the obvious hygienic challenges associated with living in a refugee camp, the patient population and the clinicians treating them were at high risk for a number of potential disasters, not least among them the constant threat of the SARS-CoV-2 virus.
The tremendous effort of caring for this vulnerable population with limited resources during a pandemic fell on not only a number of humanitarian aid groups but also on the refugees themselves, some of whom were medical doctors and nurses (Jordan et al., 2020). This scenario—where health information was of the utmost importance, but with very little resources and no IT professionals in the field—is exactly why the fEMR system was designed. Accurate, timely, and well-recorded medical records were vital not only during this time, but also after the eventual dissolution of the camp. In 2021, U.S. immigration law changed again, which then allowed the refugees back into the United States. Preliminary internal reporting via fEMR as well as anecdotal evidence from the field suggest that while the camp was established, there was a surprisingly low prevalence of COVID-19. If that is in fact the case, retroactive analysis of what worked, what did not work, and how much it cost will be useful knowledge to contribute to humanitarian aid communities. Research is currently underway by both Team fEMR and partners who were on the ground delivering aid to produce these publications.
In recent decades, systems designed to respond to disasters and humanitarian emergencies have vastly improved. fEMR has become a scalable and open-source solution to recording patient information in some of the most remote areas of the world. While technological advances have improved as well, there are still many scenarios that lack adequate electronic tools to collect minimum data set standards set forth by international organizations. Efforts should be made to implement systems like fEMR, as appropriate, to provide more granularity to lessons learned in the aftermath of disasters and other humanitarian crises. Such implementation should include consensus on interoperability, so as to preclude standalone systems that cannot communicate with other applications. Further implementation research should be conducted as well, especially examining health outcomes and continuity of care associated with improved data flow in post-disaster and austere environments.
Figure 1: A refugee camp in Matamoros, Mexico where fEMR has been deployed since 2019.
Figure 2: The aftermath of the 2010 earthquake in Haiti.
Figure 3: A typical, non-MCI deployment with the fEMR system.
American College of Emergency Physicians (2016). Disaster data collection. Annals of Emergency Medicine. 68(3), 407. https://doi.org/10.1016/j.annemergmed.2016.07.017
Anderson, M.B. (1999). Do no harm: How aid can support peace or war. Boulder, CO: Lynne Rienner Publishers, p. 161.
Fayaz-Bathsheba, A. & Sharifi-Seder, M. (2013). Electronic medical records in a mass-casualty exercise. Prehospital and Disaster Medicine; 28(6).
Feliciano, D.V. et al. (1998). Management of casualties from the bombing at the centennial Olympics. American Journal of Surgery; 176:538–543
International Federation of the Red Cross (n.d.). Haiti case study. https://www.ifrc.org/en/what-we-do/beneficiary-communications/haiti-case-study/
Jordan, M., & Ferman, M. (December 19, 2020). Stranded on the border, this migrant became the camp doctor. New York Times. https://www.google.com/amp/s/www.nytimes.com/2019/12/22/us/migrant-cuban-doctor- mexico.amp.html
Klein, J.S., & Weigelt, J.A. (1991). Disaster management lessons learned. Surgical Clinics of North America; 71:257–266.
Kubo, T. (2018). Disaster Medical Record/J-SPEED—Towards national implementation and international sharing of lessons learned from the Great East Japan Earthquake and the Kumamoto earthquake. Prehospital Emergency Care; 31:64–68.
Kubo, T. et al. (2019). Health data collection before, during and after emergencies and disasters-The result of the Kobe Expert Meeting. International Journal of Environmental Research and Public Health, 16(5), 893. https://doi.org/10.3390/ijerph16050893
Mahoney, E.J. et al. (2005). Lessons learned from a nightclub fire: Institutional preparedness. Journal of Trauma and Acute Care Surgery; 58:487–491Mercy Corps (2019, January 11). 7 tech trends that are transforming humanitarian aid. https://www.mercycorps.org/blog/tech-humanitarian-aid#:~:text=%207%20tech%20trends%20that%20are%20transforming%20humanitarian,technology.%20Distributed%20ledger%20technology%20%28DLT%29%20is...%20More%20
Oliver, J. (2020, January 30). LMIC health systems in danger of losing their own history. International Health Policies. https://www.internationalhealthpolicies.org/featured-article/lmic-health-systems-in-danger-of-losing-their-own-history/
O'Neill, P.A. (2005). The ABC's of disaster response. Scandinavian journal of surgery; 94(4), 259-266.
Stab, M.R. (December 26, 2019). What Christmas was like in an asylum-seeker tent city on the border with Mexico. Vice News. https://www.google.com/amp/s/www.vice.com/amp/en_us/article/pkedb7/what-christmas -was-like-in-a-migrant-tent-city-on-the-border-with-mexico Accessed April 21, 2021.
Yaga, D. et al. (2018). Blockchain technology overview. https://arxiv.org/abs/1906.11078
We would like to express our gratitude towards Global Response Management for their ongoing partnership in the work in Matamoros.