EHR Scalability

One of the questions that comes up when considering an Electronic Health Record (EHR) system is “can this EHR scale up to my sized practice?” A presumption is that “smaller” EHRs, which cater to small group and solo practices, would not fare well when deployed in a group of, say, 20 physicians, or 100 physicians (plus the attendant additional users in the practice, like front-office staff and nursing). Larger groups frequently engage IT consultants, who may reinforce this kind of thinking. Is this a valid assumption?

To understand the scalability of a health IT system, it is important to understand the kind of technological platform it is build upon. Over the past several decades, technology has undergone a series of evolutionary stages, as hardware has gotten more powerful and cheaper, and software has leveraged these improvements - mainframes were replaced by client/server systems, which allowed distributed microprocessing environments; and the subsequent shift to web applications and web services opened networking through the Internet. The next phase, still in its infancy, is cloud computing, which leverages these prior waves of technology (computing and networking), but embraces deep innovations in storage/data management in order to tackle Big Data.

Most “big vendor” legacy EHRs were built as client/server systems, and rely on local networks in order to function. Though touted as “scalable,” these systems require larger and larger networks in order to function at larger scales – and it is the service and maintenance of these local networks that ramps up cost. The result is exponential cost inflation, and failure of the software to deliver on its promises – UCSF Medical Center, after investing over a billion in its GE Centricity EHR, has pulled the plug on its system. Kaiser invested over $4 billion in its system (a customization of Epic), and although robust, its EHR is local to the Kaiser system – no one outside Kaiser can access it.

Further, these legacy systems often have stiff and clunky interfaces, which can actually cause errors that result in worse medical care. Recall that the previous legacy EHR certification process (CCHIT) specifically did not measure Usability as a certification criterion. The failure of these legacy systems to deliver on its promises has led Iowa Senator Charles Grassley to confront HIT vendors on these failures, and has sent a letter of inquiry to Cerner calling on them to respond.

So how can EHRs scale upwardly without breaking the bank? The key is to recognize that building larger and larger local networks is not scalable – the solution is to recognize that there is already a massively-scaled network in place: the Internet. Web-based applications are built intrinsically for scale, while maintaining safe and secure segregation of data – when you use Yahoo mail, you only see your own mailbox (and no one else’s), and you can access your email box even if several million other users are using the system simultaneously.

Practice Fusion is an EHR built on a web platform, moving to a cloud platform. Scalability is not really an issue – all 21,000 users of the system can be using the web to access their EHR, without any platform-dependent limitations. The organization of practices that enroll and utilize the Practice Fusion platform is not limited by scale – a practice of 1 or 2 physicians can use the system as easily as a practice of 1,000 physicians. Local network escalation is not the issue, as Practice Fusion leverages one of the largest cloud environments in the world, Salesforce’s platform.

The only thing, then, that determines the utility of a given web-based EHR for a particular practice is whether the features and interfaces address the real-time workflows faced by that practice – the question of extensibility to multiple sites, multiple providers, multiple specialties, and shared patients, is not an adoption-limiting consideration. There are some web-based EHR applications focused on very specific audiences – and have interfaces adapted for the workflows faced by those specialties. Practice Fusion emerged as an ambulatory EHR particularly well suited for primary care practices, and its adoption to date has been most avid among smaller practices (who are loathe to adopt expensive, local network-based legacy client/server EHRs). However, Practice Fusion has also been adopted by a variety of specialties, as well as by larger practices and networks, and continues to build features and interfaces that address the kinds of workflows encountered by these various specialties.

The bottom line? When choosing an EHR system, the serious limitations of legacy client/server systems should be avoided – the cost escalation inherent in scaling-up a local network becomes prohibitive, and the interfaces are often problematic. Instead, one should look at the new generation of EHRs based on web technologies, which overcome the cost and difficulties of scaling-upward by leveraging web services and the Internet. Though still young, these companies (like Practice Fusion) are rapidly building features and interfaces that address all the workflows encountered in the various settings of medical care.

Robert Rowley, MD – Chief Medical Officer, Practice Fusion, Inc.