Hey all. I’m just back from holiday and hoping that you all are enjoying your summer! As promised, here’s an update on our Ebola response efforts, along with some personal reflections. (In case you missed it earlier, my previous blog, which describes the drivers and goals of the effort in detail, provides some additional background. Enjoy!)
The declaration of a Public Health Emergency of International Concern (PHEIC) for the Ebola outbreak in the Democratic Republic of the Congo has finally occurred as of July 17. Better late than never, I suppose, but new resources are yet to arrive on the ground and the current response remains starved for funding, lacks adequate security, and has yet to put the right tools in the right place at the right time.
It is this last element that is our focus. We remain committed to our goal of a full electronic records management system for use in treating Ebola patients, especially inside the Biosafety Level-4 treatment areas.
In my last post, I mentioned the systems testing about to be performed at Médecins Sans Frontières (MSF) Operational Center Geneva (OCG). The testing went strikingly well, and the result was a request to get the system ready for field testing and deployment as quickly as possible. These results are impressive and should be informative, given that such a system was re-developed over a period of a few weeks by a team of 7-8 people, many of whom are unpaid volunteers. Looking in more detail at what the systems does, many will be impressed with the comprehensive level of functionality already achieved.
As I described in my last blog post, the key issue to be solved is how to get data out of the hot zone. Sharp utensils are too dangerous and paper cannot be carried out due to infection-control risks. Further, the PPE – the “personal protective equipment” that must be worn by health workers - greatly hinders manual dexterity and the transmission and reception of sound. Any electronic data capture system must have self-contained power and data network capabilities, as basic utilities and communications networks can be erratic-to-nonexistent during emergency outbreak situations, even in urban areas.
The Buendia system functions as a fully configurable clinical electronic health record (EHR) built around the WHO Ebola case definition data. Custom patient charts are derived from a spreadsheet that medics themselves design – there’s no need for programmers to configure charts or forms. Instead, a medical team leader in an Ebola center configures the app for the team’s specific context and needs. This approach truly puts the power of technology in the hands of clinicians.
To me, this is one of the most profound lessons I’ve learned from global health: large problems are best solved via targeted local interactions.
It’s important to realize that this effort is not an outlier. Buendia is just one example of how a “mini” and fit-for-purpose application (or coordinated set of applications) can improve clinician and patient interaction efficiency and quality. In late May of 2015, I was in Sierra Leone on what would be my last mission of that outbreak. The epidemic had subsided significantly, and the goals of the missions at that time revolved around “getting-to-zero” patients, tracking and following up on all known and probable contacts, and re-opening local healthcare facilities. It was a delicate time, as the people and infrastructure had weathered so much tragedy and trauma, but hope was clear and palpable everywhere you went. One of the most joyous sights for me personally was watching children dressed in bright colors going to and from the schools that had finally reopened.
But the shadow of Ebola remained. We understood all too well the painful lesson that healthcare workers, many of whom were very recently trained, were at the greatest risk of exposure, should an Ebola-infected patient present at a hospital or clinic.
The solution I designed and implemented started as an Ebola virus disease (EVD) triage application purpose-built to enable menu-driven care and assessment at the clinical triage center of a clinic in Kono, Sierra Leone. Ideally, the principle of protocol-driven care should be as easily automated as placing an order in a fast-food restaurant. When a patient presents, the health worker is guided by a set of multiple-choice questions driven by an algorithm based upon EVD case definition. These can be built to be run pictorially, such that health worker literacy does not become an issue. There are also excellent low-cost and open-source toolsets available, such as CommCare by Dimagi, that I used for these applications. The paper system at the clinic was simple and elegant, consisting of a single sheet per visit for a single patient.
Once the triage app was in place and trained, I quickly began to receive requests to automate other parts of the clinic. A better way to communicate test results from the lab to clinicians - check! Can we have a pick list of the World Health Organization (WHO) essential meds list – check! Within 2 weeks, most of the clinic’s essential processes had been digitized and the physicians running the clinic looked forward to further digitization and automation. Another significant benefit of the “light app” approach was training and local economic development. I was able to teach people to write and modify applications in just a few hours. This turned out to be life-altering for a number of young people who have gone on to become technology leaders in their local communities.
Taking the time to quiet our displays of knowledge, opinions, experiences, and prejudices may represent the most valuable contribution we can make to improving healthcare technology in our own society. It is time to be quiet and listen—and learn.
Watching how quickly solutions such as this can be developed and deployed, I cannot help but wonder what those of us working in “modern healthcare” are doing wrong. Five years and 19.2 billion dollars after the initial meaningful use deadlines of the American Recovery and Reinvestment Act of 2009 (ARRA), it’s difficult to find anyone in healthcare that feels this is all going well. There are steady streams of articles in the peer-reviewed literature and lay press decrying the ill effects of electronic medical records (EHRs) on patient safety, clinician efficacy and well-being, and the economics of healthcare. Pieces with titles such as Physicians' Well-Being Linked To In-Basket Messages Generated By Algorithms In Electronic Health Records, or The Patients versus Paperwork Problem for Doctors, or Why Doctors Hate Their Computers, or Continuing Patient Care During Electronic Medical Record Downtime, or The Impact of High Capital Costs of EHRs on the Medical Staff are so ubiquitous as to have become background noise everywhere except within the healthcare information technology trade press. No need to wonder why: indeed, the meteoric uptake of the @EPICEMRparody account on Twitter has become a salve, support system, and rallying cry for many who work in clinical care.
It’s easy enough to bemoan the situation and add to the noise level of discontent. Positive action and genuine solutions, however, are far rarer. Can mini-EHR systems like Buendia be part of the answer? In many cases, they already are, although this is often below the radar in many institutions. The number of clinical wards that are running apps or even mini-EHRs on top of their enterprise systems seems to grow every day with no end in sight, but the cost is additive. Healthcare institutions are under constant pressure to deliver value at lower cost, and more and more, healthcare technology is seen as a cost driver versus a value driver. This approach will need to change, and change significantly. While the solutions I described above are clearly limited in scope and scale, they should not be discounted as viable strategies for far more complex issues.
To me, this is one of the most profound lessons I’ve learned from global health: large problems are best solved via targeted local interactions. In the case of the Ebola outbreak in the DRC, solutions will only be found via hand-to-hand partnering in each unique village, chiefdom, community, and country. For healthcare in the United States, we need to start by re-imagining the single clinical encounter and admit that fundamental change is needed. The fact that the road will be difficult is no reason to avoid the journey, and the size of the challenge can’t be an excuse for not fixing it. Instead of monolithic medical records systems designed to view each clinical interaction as a fundamental element of financial productivity, we need systems that view those same interactions as a fundamental unit of quality care.
At the same time, it’s easy to forget that we have as much to learn as we have to teach during global health emergencies. The fact that a handful of talented developers can build and implement a highly-functional mini-EHR to digitize an Ebola clinic is not a lesson to be dismissed out of hand. The complexity of healthcare has been built by healthcare itself and not externally imposed. And since we have no one to blame but ourselves for this state of affairs, fixing it is also our responsibility. Taking the time to quiet our displays of knowledge, opinions, experiences, and prejudices may represent the most valuable contribution we can make to improving healthcare technology in our own society. It is time to be quiet and listen—and learn.
Back to Buendia, we are excited to be launching the field testing and pilots in the Ebola outbreak in DRC over the coming weeks, and we’ll keep you updated on progress.
Eric, Ivan, Ping & the Buendia Team