Understanding Application Programming Interfaces (APIs)

Application programming interfaces (API) are specifications that detail how software components should interact, and APIs provide the technology to power graphical user interfaces (GUI).  From a health care organization (HCO), health researcher, or patient perspective an API is the glue that enables the ability to connect systems and data.  If one uses their mobile phone to access their insurance claims, for example, that transaction makes use of an API.  If your surgeon, for instance, obtains x-ray images in the middle of surgery that operation is supported by an API.

According to a recent HIMSS Connected Health Survey, more than 81% of survey respondents indicated that they use or plan to use at last one connected device.  51% of survey respondents reported that they use or plan to use more than three connected devices.  Connected devices include vital sign monitoring tools, remote monitoring, and patient portals.  Each of these connected devices uses APIs to connect and share the information required to power connected devices. 

As an example, of the use of APIs, the Centers for Medicare and Medicaid (CMS) is one of the first health organizations to pioneer data publishing online. They have released many different datasets with documented APIs, including data about hospital and provider data, as well as data about pharmaceutical transactions. CMS is making their data more accessible online; they are also encouraging others to use and build applications, games, and other websites by using their APIs.

Earlier we talked about organizations who are publishing data and enabling API usage online the Centers for Disease Control and Prevention (CDC) is another organization that is encouraging data consumers to use the data they are publishing on chronicdata.cdc.gov. Through this website, a group called DiseaseCast was able to pull from CDC’s APIs to build a custom website that uses their publicly available data to predict chronic disease outbreaks.

As HCOs and others continue to improve their informatics and business intelligence capabilities, it is important to consider APIs.  APIs establish a standardized approach to sharing data.  The API replaces the manual processes that many organizations use to share data.  These manual processes often require custom programming and often result in inaccurate data.  Future newsletter articles will further articulate the processes necessary to design and architect APIs.

About the Contributors
Stuart Rabinowitz currently serves on the HIMSS Clinical and Business Intelligence Committee and is the Director of Strategic Alliances for Socrata, specializing in health data.  Socrata is a cloud software company focused exclusively on democratizing access to government data, serving organizations like the City of New York and the World Bank. Stuart is an expert in enterprise architecture and data analytics.  Stuart holds an undergraduate degree from Temple University, an MBA from Lehigh University and is currently pursuing a Master’s of Science in Health Informatics from the University of Illinois at Chicago.

Amy Yeung is a Data Management Analyst that supports Socrata’s work in the non-profit and health sectors. She helps organizations understand how to manage, use, and disseminate data to various data consumers. Amy is passionate about health and international development; she is excited to help organizations utilize data to solve critical policy problems and help decision makers make more informed, data-driven decisions.