In its recent April 18, 2009 special report on health care and technology, The Economist magazine offers a nice high-level summary of both the progress and struggles that have taken place in this space, and concludes that the digitization of medical records is getting closer through health information technologies (HIT) such as electronic health records (EHRS). While the terminology in this space is still evolving, the article entitled “Hit or Miss” in this report defines EHRS as “digitized versions of all the bits of paper usually kept in files by all the doctors a patient sees regularly”, and EHRS as “all the hardware, software, and other kit needed to make sense of the data and to give remote access to them”. (Additional, related terms are often used interchangeably, with electronic health records (EHR) often assigned to public or global records, and electronic medical records (EMR) often assigned to private or local records; the term EHRS used here is being used as a universal term that includes both of these domains.)
According to this article, while “Father of HIT” Morris Collen of Kaiser Permanente became convinced of the promise of “medical electronics” back in the early 1960s, he has also witnessed numerous failures of such efforts in the four decades since. Those of us familiar with this space also recognize that such failures are not limited to one or two firms, but widespread. Some of the stumbling blocks to progress that this article sights are privacy concerns and resistance from physicians, although encryption software has improved greatly over the years and legal changes such as a law passed in the United States that stops insurers or employers from discriminating based on genetic information have taken place. In addition, financial incentives being planned via the recently passed stimulus package to help encourage physicians to go digital are also forthcoming.
An additional topic discussed is the choice between bottom-up and top-down approaches to tackling this problem.The Economist magazine sites the Connecting for Health project by the National Health Service (NHS) of the United Kingdom, and it is argued that the top-down approach utilized for that project did not take into consideration the concerns of physicians and hospitals, or the philosophy that patient control might be beneficial. In contrast to such a bottom-up approach, however, the writers indicate that centralization is likely to be better when the setting of such areas as security standards and data sharing protocols is needed. The writers argue that some compromise between these two approaches might have helped the NHS succeed on the Connecting for Health project, now reported to be 4 years behind schedule. For example, Simon Eccles, medical director at Connecting for Health, “concedes that his system could have provided more choice to local hospitals”. Morris Collen, however, goes one step further by meanwhile insisting that patient empowerment is critical to success, and that his original vision can be supported by technology that has increased in reliability over time, noting that “the patient has lived with his medical problem and often knows it better than the doctor”.
Various architectural and business quality benefits associated with digitization of medical records (and somewhat commonplace within industry literature) are alluded to in this article, including usability, performance, and cost and benefit. In addition, Peter Neupert, Corporate Vice President of Microsoft’s Health Solutions Group, is quoted as saying that we need to “remember that HIT is not like railways, where the guages had to match perfectly for interoperability”. And a recent report by the Institute of Medicine is sited as warning that “medical information must be free to move about on rival software systems”.
Patient empowerment combined with simultaneous implementation of phase one and phase two of HIT reforms in the United States (which are planned to consist of health information interoperability exchanges and adoption of EHRS, respectively), is discussed as a possible approach moving forward. Thomas Lee, CEO of Partners Community Healthcare, a large Boston-based healthcare provider, for example, acknowledges that he “openly Google[s] things he [doesn't] know in front of [his] patients”, but also indicates that he “want[s] the provider to control records behind a firewall, and to let patients peep into them as necessary”. Aurelia Boyer, CIO for New York-Presbyterian Hospital, views healthcare as being “paternalistic in nature”, and that instead of waiting for healthcare integration she wants to give patients data access immediately and link information later. New York-Presbyterian Hosptial recently launched a pilot EHRS using Microsoft Health Vault to do just that.
As someone who has experience developing clinical systems in the public healthcare space, as well as private portal software and integration of systems in both realms, it is clear that the interoperability discussed in this article is not a trivial matter, and is actually one of the most important architectural quality risks, having a profound impact on business qualities. Industry research has repeatedly sited system interfaces as being very high in risk, and projects that address interfaces late into development typically experience implementation delays which lead to cost overruns. While Peter Neupert may be correct in that systems do not need to exist in “perfect” harmony with each other, as practitioners in this space we do need to realize that systems need to communicate well enough so that data is not misrepresented, especially since the interpretation of this data is often relative.