Hi All, Great discussions regard OpenEHR and Patient-Provider contacts. The question I have is what are the goals and objectives of these Patient-Provider contacts, especially from an information and systems theory perspective? The focus seems to be mainly on the Provider.
Information Systems Theory (Includes Cybernetics, the science of communications and control for) would address the Patient and Provider as peers and attempt to optimize information flow between both. It would not restrict the time-of-transfer to Patient-Provider communications. Communications can be unilateral. Systems theory, approaching the Patient from the Provider position, would view the Patient as a black/gray/white box approach: black: I'll ask questions, receive some responses and generate a diagnosis gray: I have some information from the Patient and along with my queations and their responses I'll generate a diagnosis white: I have lots of information on the Patient and along with my questions and their responses I'll generate a diagnosis based upon the maximum amount of information available. Reference: Norbert Weiner: Cybernetics Practice: Advanced Systems and Networks Diagnosis On the Patient level, when encountering the opportunity to hear what has been entered into my medical report, the number of errors, omissions and problem seem to indicate that the record belongs to another Patient. The absence of feedback and attempts to verify what is in the record is interesting but a concern. Since the distance between medical diagnosis and systems diagnosis, in my opinion, appears relatively short (probably because we humans built them; check the von Neumann architecture for computers) , and because one would like to optimize the amount of information that flows into and out of EHRs, the focus on recording Provider impressions will likely produce a result that is somewhat less than 50% efficient. Whether the Patient-Provider interface can be modeled as a black/gray/white box interface determines how much less than 50% the efficieny is. The 5-8 minutes allocated per Patient has a tendency to push this down to the low levels. OPINION: -The 5-8 minute Patient-Provider interface will not support an effort to gather all appropriate information and render a complete, lasting diagnosis, e.g., try a Patient who is borderline present for whatever reason. Note the models presume that the Patient and Provider are operating on all cylinders and are 100% efficient communicators. How much information can be transferred in 5-8 minutes. -There must be associated EHRs that can be integrated with other EHRs that support both the Patient and the Provider, e.g., short- and long-term recovery. Reminds me of the need to describe current aliments and the originating/root cause (often lost in the records). From the Providers viewpoint I would expect to have available information on today's aliment, the root cause, related progress, other conditions/problems, and all the intermediate diagnosis that may or may not have been appropriate/correct/missed-the-mark. In short, I would need to know what the prior outcomes had been before venturing into the forest. I haven't seen studies that report on the efficieny of Provider diagnosis for particular individuals and classes of Patients. An analogy might be a complex manufacturing process that produces widgets that drop into a socket in another process, e.g., Henry Ford. One doesn't optimize and control the processes producing widgets standing at the end of the production line. Neither does one determine the efficiency of the process by a quick visual tour. There are many Healthcare organizations that are keenly interested in outcome-determination in Healthcare, e.g., the NHS. One gets there by gathering as many bits and pieces as possible and using as much as possible in today's and tomorrow's analysis and diagnosis (great data mining/evaluation/performance application). An analogy would be AOLworldwide operations on a daily basis. GOALS/OBJECTIVES: -Information/System Theory -Maximize data retrieval/storage/analysis -Make room for multiple Provider diagnosis regardless of when rendered -Make room for Patient input and feedback -Make room for freeform/unformatted input to support unformatted/unstructured data -Use object-oriented information design/structure to keep this data together -Thomas Clark - If you have any questions about using this list, please send a message to d.lloyd at openehr.org
