This article by Dawn Smith about OpenMRS was originally posted in the December 2010 issue of Open Source Business Resource (OSBR), which focuses on Humanitarian Open Source projects.
Healthcare is a distinctly information-centric business. The care environments that effectively capture, process, and re-use clinical information are typically more efficient and produce better healthcare outcomes. These functions are even more critically important in the resource-poor areas of the world, which are most often forced to support higher disease burdens per capita with fewer human and infrastructural resources. The onset of the HIV/AIDS pandemic in the 1980s spotlighted the disparity of resources between developed and developing countries. Resources such as basic medical care, human capacity, and physical infrastructure were absent from the countries that also carried the heaviest burden of infectious diseases, which often led to dramatic drops in life expectancy.
As a response, initiatives such as the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR) and The Global Fund to Fight AIDS, Tuberculosis, and Malaria provided aid to help establish health systems in environments without infrastructure. Consistent with developing health systems, there grew an increasing focus on the development of health information technology (HIT) which supported routine clinical care processes such as clinical and hospital visits.
It was within this context that the collaboration that led to OpenMRS began. In 2004, early independent work by the Regenstrief Institute in Indianapolis and Partners in Health (PIH) in Boston encouraged both groups to find collaborative partners for developing patient record systems. In each team’s work with health-system-strengthening initiatives (Regenstrief’s support of the Academic Model Providing Access to Healthcare in Western Kenya and PIH’s support of environments in Rwanda and Haiti), there was a shared recognition that an information system’s ability to grow would only come with the sharing of expertise, experience, and limited resources. This commitment to each other’s success was manifest in a deep health informatics collaboration between the two groups that led to shared designs and software development which unknowingly served as the formation of an open source community. The decision to use freely available online tools to facilitate collaboration between the two groups made it easier for other organizations to learn more about the work and participate at all levels in the community.
As a result, since 2006, OpenMRS grew from utilization in Kenya, Rwanda, and South Africa into a community-driven medical record system platform implemented in over forty countries throughout the world. This platform is comprised of a growing number of software components and collaboratively developed designs that can be mixed and matched to develop a patient record system application. The community involved in this work consists of hundreds of people who work together collaboratively to meet the massive demands for basic HIT in resource-poor settings. They come from all walks of life: from clinicians who work directly with patients in Africa to technologists who have no direct healthcare experience but want to do something philanthropic with their time. OpenMRS in many ways is a vibrant example of what is now more commonly known as a humanitarian free and open source software (HFOSS) organization. What started out as a partnership formed out of need has evolved into its own not-for-profit, which has a community-coordinating mission in support of sustainable enterprise patient record system deployments.
Health Information Business Ecosystems
The OpenMRS community believes deeply that the key to sustainable patient record systems in poor environments is indigenous local health informatics capacity. All too often, aid-supported environments do not have health information workers readily available, which leads to externally implemented and supported information technology, much of which is proprietary in nature. In turn, this leads to all sorts of problems as the healthcare practices naturally evolve within these environments, necessitating the technology’s evolution as well. “Western” organizations struggle to understand the cultural contexts that make information systems fit within clinical work flows, and health systems have rapidly evolving demands as they become increasingly information fluent. These tensions lead to high levels of failure and health systems that are poorly empowered to take advantage of their own health data. Additionally, this model of business is not consistent with the goals of humanitarian aid, which seek to empower self-reliance in emerging health systems and support the sovereignty of countries as they emerge into middle and higher-income economies.
We see humanitarian FOSS as a remarkably positive disruptive market force as it relates to HIT in aid-supported environments. Using the OpenMRS example, the community proactively makes all of its proceeds freely available through the Internet and other means and puts considerable effort into capacity development. So not only does it serve free food, but it also gives away the recipes alongside all of the ingredients as well. It also holds regular cooking classes. By helping establish local informatics “cooking” capacity, new forms of business opportunities form. These newly skilled workers can then leverage an international community for help and peer mentorship while participating directly in the massive infrastructure building activities that are taking place within their environments.
HFOSS empowers health systems to collaboratively participate in the construction and maintenance of their own medical record system, whether through hiring local health information workers directly or in collaboration with newly formed HIT-based service organizations. In a vibrant community built upon an open platform, individuals and organizations “compete” with one another based on the quality of service or knowledge they bring to an environment, as consumers can choose from a variety of workers who all work from the same technical means. It also gives these health systems more control of their own health information.
We are seeing these very patterns emerge in multiple countries throughout Africa. For example, in Rwanda, the local Ministry of Health has chosen to begin a project to implement OpenMRS broadly as the electronic medical record used in outpatient clinical settings. As a result, they have developed their own functional requirements specifications, and have requested external aid to help build capacity in implementing and developing OpenMRS-based applications. This training takes many forms, the most exciting of which is a formal training partnership between PIH and the Kigali Institute of Science and Technology. Graduates from these training opportunities are then either hired directly by the Ministry of Health or are increasingly being paired with business-savvy individuals and seed capital from the Rwandan Development Board to help them establish their own HIT-implementation organizations. These workers will then be called upon to carry out both the initial implementation and long term support of the more than 400 clinics located throughout the country.
The perhaps more entrepreneurial nature of Nigeria is encouraging a more emergent business ecosystem. We are aware of more than a half dozen new, self-identified HIT companies that have formed in support of local health systems that are using OpenMRS as a foundational piece of their strategy. Additionally, self-forming local collaboratives in Nigeria, such as the one led by the Institute of Human Virology, are creating the demand side of the business by bringing health systems to the HIT workers. In Kenya, a local enthusiast group is forming virtually, in response to a recent publication of electronic medical record functional requirements by the Ministry of Health, and it is their intention to create a Kenyan-specific “distribution” of the OpenMRS software consistent with those requirements. There are other examples of such pro-business ecosystems, each forming in a way that is consistent with the socio-cultural contexts of the environment.
The cases in Rwanda, Nigeria, and Kenya represent only a few examples of the emerging successes OpenMRS has seen where a community, even in a resource-poor environment, can harness and build local health-informatics capacity to create a sustainable system. Yet we have to ask ourselves: What are the lessons we can learn from the success of others to create a sustainable model for all communities? From the perspective and experiences of OpenMRS, much of the growth seen within these environments extends from a focus on the following lessons:
1. Enable community ownership: If the community does not have ownership of the tools needed to succeed, then the community sees little, if any value, in the project. Volunteers in the community contribute their best ideas, time, and talents to the organization and development of the software. Given that OpenMRS asks these contributions to be freely shared with the rest of the world, we believe in the importance maintaining a culture of shared ownership.
2. Focus on real, not perceived needs: Potential solutions come from individuals who have an idea of the problem; sustainable solutions come from individuals who know the problem. In resource-poor environments burdened heavily by infectious diseases, it is vital to respond to the real, not perceived, needs of that community. End users and developers collaborate through various forums to prioritize design features of the software. This healthy design tension ensures that end users receive the features they need for daily workflows and that developers can scale the code appropriately.
3. Re-use what works: Redundant efforts are hazardous in resource-constrained environments. Prior to FOSS, many organizations re-created ideas and code that already existed but were inaccessible due to copyright and licensing restrictions. Redundancy and inefficient responses in many HFOSS environments cost more than time; they also cost patients’ lives. OpenMRS built its software upon the underlying design developed by the Regenstrief Institute. Blending this design with the work from open source software projects allowed OpenMRS to develop efficiently and quickly, and in turn, OpenMRS hopes to serve as a foundation for other open source projects in the future.
4. Promote active transparency: Active transparency establishes earned trust within the community. As an organization, OpenMRS openly shares both its successes and failures. In doing so, the response is two-fold. Community members actively work to build upon and improve the success of the software while preventing the reoccurrence of the same failure, and in turn, the same individuals share their own stories and encourage others to do so as well. What we have seen is the community connecting through various forms of communication (face-to-face meetings, weekly conference calls, online forums such as Internet Relay Chat, and message boards), which has taken on a movement of self-organization.
The lessons described in this article represent some of the core values and experiences of OpenMRS, but they certainly do not represent the only lessons that have been learned. Whether in the foundations of an emerging business ecosystem or in the daily workflow of an HFOSS organization, these lessons contribute not only to the success and sustainability of those entities, but to the health informatics capacity in resource-poor environments as well. The success experienced by such systems represents a tide that raises all ships; when the community succeeds, we all succeed.