Discussion about entry: Clinical Decision Support for Triage of Violence and Suicide

is needed, because people do not become patients by deciding that they have psychiatric disorders, psychological distress or medical/surgical diseases. They do not feel normal - i.e. stomach cramps for no reason, that can be surgical emergency or somataform manifestation of Major Depression, or chest pain with trouble getting their breath, that could be Panic Attack or Cardiac event. They cannot triage themselves to either "psychiatric evaluation service", ER, Urgent Care Center or their PCP. Likely, Major Hasan had clinical encounters that had to be documented with templates of the Army's EHR, "AHLTA". Did he go to a Psychiatrist, because he knew he was unraveling? Most likely not; references to his progressive mental disorganization and identity diffusion may have been alluded to by any provider, whether Psychiatrist, Psychologist, Surgeon, PA, NP, Counselor, Medic etc, etc. All of these healthcare professional disciplines are on the front line of Army health care and they both diagnose and enter their opinion re "presenting problem", along with text describing it. The latter text is usually formatted as a "SOAP" note. (Subjective, Objective, Assessment and Plan) The Digital Clinician (R) integrates Neuropsychiatric rules and knowledge base for interviewing, working up and triaging patients via rules and knowledge base of all Medical and Surgical Specialties, because neither God nor patients sorts out to whom and where they first present. And, that first presentation can be the most important - either getting it right the first time, or anchoring the patient with invalid diagnosis that obstructs effective intervention for a long time - even a lifetime.
But The Digital Clinician (R) is not only a cross-specialty tool that disciplines all clinicians to see the whole patient during that critical first encounter, but its language must be clear and precise enough for multidisciplinary parsing for an EMR. Different health care disciplines use different terminology for the same elements of a clinical presentation. Professionals talking as a team can filter out the irrelevant from the relevant, but computers are not intelligent enough to do so. Thus, the lexicology and taxonomy of Computerized Clinical Decision Support (CCDS), such as The Digital Clinician (R), must facilitate the filtering out of multidisciplinary "noise". For example, a statement from a medic like, "he's out cold", must be parsed into "Assault" and "Head Injury" and/or "Altered Mental Status with Loss of Consciousness." John Liebert

The only way I know how to test validity of Knowledge Base and Underlying Inference Engine is via a controlled clinical study comparing outcomes of patients being Triaged the traditional way - "seat of the pants and intuitively from personal experiential base" - vs patients being triaged with computerized clinical decision support (CCDS) of The Digital Clinician(R).

Unfortunately, too many clinical IT applications have been hyped as if proven to both reduce medical errors and reduce healthcare costs - a simplistic and seemingly sensible solution to an enormously complex problem. That problem and challenge today is Service Optimization in The Practice of Medicine and The Multidisciplinary Delivery of Health Care.

I am getting such a controlled study now through a highly respected, combined School of Nursing and Medicine project. If it fails to show improvement in outcomes utilizing The Digital Clinician (R), then these questions must be addressed specifically.

But, for now, I only know how to be pragmatic and find out, using evidence based rules of time and epidemiologically-informed clinical decision-making, whether such Clinical Decision Support actually improves outcomes. John Liebert

It has been proven that doctors need more information every day of their practice than they know that day, but I know of no studies that show such apparent deficiency results in worse - rather than best practices supported by Clinical IT. And, recently we have had disastrous cases of malignant psychopathology slipping through clinical services - Cho was involuntarily committed; Major Hassan was observed by peers and supervisors to be going over a slippery slope. In such cases, Computerized Clinical Decision Support could discipline medical judgments by forcing clinicians to align elements of clinical presentations with evidence-based diagnostic entities - i.e. Major Depression with Psychotic Features. But, clinicians may have good reason to keep diagnoses open by reducing such diagnostic specicity - i.e. Depression NOS (Not Otherwise Specified), as long as they expect to stand by it later, if necessitated. The latter was Cho's diagnosis, and proved to be both invalid and disastrously minimizing of his psychopathology; how many students meet the minimal criteria of "Depression NOS" on any given Sunday on our college campuses? Alot, I suspect, and Cho was no ordinary guy on campus. But, increasing patient volumes; decreasing fee structures; more disciplines involved with same patient's presentation over time and increasing administrative work by clinicians all require intelligent user-interface in order to make Clinical Decision Support of Interview, Workup and Triaging effective. That means both cost, as well as clinical outcomes. This is what we are striving for with both training and Clinical IT; namely "Service Optimization". The EHR/EMR is inevitable for nearly all practitioners in the near future. And, most already use handheld devices for some of their clinical functions every practice day. The goal should be to make the user interface effective in the real world of clinical practice - not the ethereal world of textbook medicine translated via Computer Engineering to work stations. CDS, therefore, must support and direct all users interacting with the same patient presentation to the same place efficiently and correctly over safe time for intervention, whether emergency, as with Altered Mental Status in a Natasha Richardson - or, chronic, like most high utilizers of primary care services. That is a tougher challenge than may be realized today with the noise of political rhetoric in Health Care Reform. Current and future stakeholders are wary, including legacy EMR systems already installed at high cost. I explain my views on this in "Unity in Health Care the Toyota Way; Can Lean Engineering Fix Health Care?", which I will attempt to attach. (King County Medical Bulletin, Seattle).
John Liebert