.jpg)
Ironically, medicine is one of the most technologically advanced and, at the same time, technologically antiquated fields in our society. Researchers are using nanotechnology to target specific intracellular processes, transforming mature skin cells into functioning heart cells by applying genetically active transcription factors, using imaging to recreate three-dimensional models of internal processes, and applying advanced computer processing power to integrate millions of bytes of information into patterns of gene expression. Clinicians are conducting complex cardiac ultrasonic evaluations with a portable hand-held device, monitoring electrocardiograms through a cellular phone application, performing robotic surgery and scanning thousands of articles to arrive at the most recent information in mere minutes. The accomplishments in both areas are fast-paced and too numerous to begin to catalogue here.In the face of this impressive rate of progress, the profession has been remarkably slow to adopt information systems that are commonplace in other industries. One does not think twice about going to a machine in Chicago, or Paris or Bangkok to withdraw cash from ones bank account in New York. Walmart revolutionized the retail business by using informatics to track inventory, sales and productivity. Songs, movies and TV shows (and virtually everything else) are readily available on our hand-held portable devices (phones, tablets, etc) Yet if I go my local hospital emergency room, who knows how long it will take to find my previous hospital records—as for my complete personal health records from my doctors’ offices and other institutions? Not even a possibility.
Prior to the government incentives (soon to followed by penalties) for electronic claims submission and more recently documentable electronic “meaningful use”, which includes electronic submission of prescriptions, only 5-10% of physician offices/practices actually used an electronic medical record. Now the figure is in excess of 50%. But the situation is complex. There are literally hundreds of systems available. And these systems are distinct from one another and from those used by the hospitals, which, of course, are not linked in any way to those used by the government or the payers.
What are the challenges? First, the nature of the information is somewhat complex. Although it may be straightforward, perhaps even intuitive, to digitalize the largely numerical information contained in a bank record, it is much more challenging to attempt to digitalize the vagueries of patient’s symptoms or complex differential diagnoses. Second, accompanying the complexity of subject matter is the absence of an agreed-upon digital language. Such efforts do exist, but they are imperfect and not uniformly accepted. Even simple bits of information such as age might be recorded as age in one place and calculated from date of birth in another. Is shortness of breath the same thing as dyspnea on exertion? Are swelling and edema the same thing, etc? And how do we digitalize impressions? Does the patient seem to have congestive heart failure with a component of COPD? Suppose that impression is wrong, but is now digitally recorded in the patient’s record; how does it get removed? (i.e. what is the mechanism for distinguishing between presumptive or speculative diagnoses and more definitive diagnoses, and how does the former get removed after arriving at the latter, and do we really want to remove the historical record of what was the clinical impression at the time???) Thirdly, as might be expected, the first and even second generation of electronic medical records are not particularly user-friendly. Computer enthusiasts in the 1960’s and 70’s didn’t mind generating those index cards to program each function, but doctors are interested in taking care of patients not in making new medical records work. Many in the profession have been using paper records successfully and comfortably for a long time and are reluctant to change from something that works to something that, at least initially, doesn’t seem to work as well. Fourth, medical care is basically local. Therefore, just as there is no generally acceptable digital medical language, there is not a uniform system for recording, not to mention transmitting information. Fifth, although not foreign to the industry, especially when it comes to financial records, medical records require a high degree of confidentiality. The ease of access and transmissibility challenges the ability to secure the privacy of the information. Lastly, informatics need not only support point-of-care clinical decision-making, it should also be able to translate into outcomes assessment. Unfortunately, the registries being developed by the professional societies each have their proprietary definitions and elements which do not necessarily readily correspond to the electronic data in the medical record. The process of data extraction requires a level of semantic logic which is beyond current capabilities without the insertion of a sentient person to extract the appropriate data from one place and put it into another.
Despite the slow pace of progress, there is definitive progress—incremental and definitive. The ability to integrate information, coordinate care and improve processes is already manifest within certain local health systems. This is why FHRI has been involved for several years in the use of advanced computing technology to decipher the potential impact of genetic signaling in the transmission of disease. The ability to integrate personal information, such as the blood glucose levels that you monitored with your portable phone app into your medical record, or to integrate your personal genome information with the differential effectiveness of a given medication or treatment will open a whole new era of medical care. These efforts correlate nicely with FHRI’s impressive record of community education in making people aware of their personal cardiovascular risk. Our “homework” still needs to keep up with our vision.
