e-Health Expert System for Primary Care
Development of an expert system specific in aiding preventive treatment of cardiovascular diseases among patients
As the costs of health care grow virtually exponentially, efforts have begun to explore alternative care models in traditional primary care. One area that has yet to be exploited to any meaningful extent is the devolution of certain aspects of care to the highly informed patient. We propose that supervised web-based care is readily feasible for many rigorously-defined preventative care endpoints if relevant & natural Human-Computer Interfaces (HCI) can be developed. This asynchronous mode of care will free traditional synchronous encounters for those issues which truly require direct physician patient interactions. As a proof of concept module, we focus on cardiovascular (CV) preventative care in view of the extent of the cardiovascular disease prevalent in developed and developing nations and the rigor of the relevant evidence base. CV diseases represent the single largest cause of mortality in the developed economies and are rapidly reaching epidemic proportions in the developing world. A large proportion of total primary care outpatient appointments are currently occupied by CV prevention.
A usual Cardiovacular prevention appointment consists of an initial encounter with the physician where a standard panel of health information is collected along with some basic physiological parameters such as weight or blood pressure. The patient then undergoes a number of blood tests (e.g. for diabetes, cholesterol, inflammation, & other common risk factors), potentially additional evaluations such as an ECG or stress test, and with the results sees the consultant again, who, based on the results of the physical exam & the laboratory tests, will often prescribe one of several classes of medications, e.g. an aspirin, a statin, an ACE inhibitor or an Angiotensin receptor blocker. These basic medication strategies have been shown to be very effective, their use is generally undisputed & standardized algorithms are used to identify those who are likely to benefit. Indeed some of these therapies are currently being considered for over the counter sale. In addition, the side effects & toxicities of each of these medications are well known, & strategies to monitor these effects of the medications (e.g. follow-up evaluations & blood tests). Furthermore, there are also many lifestyle modifications which complement the therapeutic efficacy. Finally, in today's healthcare environment, comprehensive prevention or intervention requires multiple synchronous visits, as there is usually insufficient time to deal with data acquisition, decision making & prescription across all modalities on a single visit.
Our new approach : Since the indications for the above mentioned medication regimens are so well defined, we are proposing to put in place an intelligent expert system that can identify the appropriate medications & follow-up from a combination of patient input & test results. This approach would extend a basic system developed by John Moore at MIT Media Lab that can interactively replace the initial appointment with the Physician by an online appointment with a virtual agent acting as the doctor.The proposed system could thus be used for automatically conducting patient pre-visit interviews. It will not replace a human doctor, but would be used before an office visit to prepare the patient, deliver educational materials,triage care &preorder appropriate tests, making better use of both doctor & patient time. The new approach could make the consultation look like:
- Virtual Agent based Consultation: which could be initiated by a phone call to the hospital; delivered online to a TV or computer, and asynchronous to the Doctor. Health information would be collected from the patient via a speech or touch-controlled interface. The result will be (an expert system) recommendation for further testing or an immediate referral to doctor based on initial indicators.
- Tests: The result from the virtual agent based consultation can be printed and the tests offered by Pharmacies or even Supermarkets. The Test results will be given to the patient and directly feed into the supervising provider's information systems (i.e. NHS Trust’s patient record system or e.g. Google Health)
- Test Analysis: The expert system will analyse the test results in consideration with the input from the initial virtual consultation and will produce a recommendation of either: a specific treatment regimen, lifestyle modifications & both an asynchronous & a synchronous follow up plan; or, an immediate referral to the doctor.
- GP/Consultant visit: In the final stage, patients started on any treatment regimen will have an appointment with a Doctor, who will explain the regimen & follow-up plan.