My output is a design proposal for a fictional EHR system, called Barnett. My proposal is shown on the homepage of this website. It is a combination of what I think are the best solutions to each of my problem areas.
Format. The file format is a modular, extensible format, which is both consistent but easily adaptable to user needs.
Authentication & Privacy. Barnett uses a flexible authentication system to adapt to whatever is appropriate. 3rd party apps can access information through OAuth logins, and statistical tools are provided for research.
Network. A distributed network simulates a single record per patient - while providing a robust, privacy conscious implementation.
User Interface. An extensible interface provides a framework which can be adapted to the context of care, and individual user.
I think Barnett has two unique design philosophies behind it.
Barnett isn't focused on the needs of any specific organisation, instead, it is focused on the needs of the problem - the management of health records. It provides an agnostic way to store, share, and interact with health records. This means it has limited scope of functionality compared to EMR systems used in Primary Care - for example, it doesn't provide any appointment management or invoicing tools. It's designed to be used in unison with other tools which will fill this gap.
I believe that by splitting up the functionality provided by EMR systems, there will be more opportunity for product designers and developers to improve the experience of health care professionals, and specialise on hard problems.
At its core, Barnett accepts that healthcare is varied. The needs of users change dramatically depending on context, institution, region, and the individual. Instead of working against variety, Barnett is designed to adapt to variety. The format of the records themselves can be easily adapted, and shared. Likewise, the interface is extensible, and customizable down to a very specific detail.
Extensibility is a long celebrated factor of good software. In fact, The Internet Engineering Task Force (the group of individuals who oversee the creation of the protocols and technologies that power what we know as the Internet), note that extensibility is a key factor in a protocol being 'Wildy Successful' (Aboba & Thaler, 2008).
Due to the varying nature of healthcare, I believe that adaptability was a necessary design philosophy for Barnett.
Designing the system is one challenge, but explaining it to a visitor, who may not have much experience in Healthcare, or Software Design, is another challenge.
To solve this I used an iterative method of testing with friends and fellow students, and then redesigning. This is appropriate as I wanted the design proposal to be approachable by anybody, not just people with a background in healthcare. The design went through five iterations before settling on the current output. Full size here