All of us who are in the field of informatics are well advised to pay attention to voices like these: Op-Ed Contributor – The Computer Will See You Now – NYTimes.com.
So many times have I heard the whines of those who develop applications for the clinicians. The refrain goes something like this, "The doctors are technology averse", "The clinicians are not interested in learning new ways of doing things".
There are innumerable examples of doctors embracing technology when it helped them achieve the goal of delivering better or more efficient care. I remember the days when ultrasound first arrived on the scene. The ultrasound machines in those days were kludgy pieces of equipment. The screens were tiny, with no shades of gray, the text messages displayed were cryptic, the buttons on the console were ill-organized. In short, they were the epitome of user-unfriendly technology. Yet, the clinicians took to them with gusto. Why? Because they gave them that extra edge in looking at soft organs. Obstetricians, whose notoriety as technology-averse is a legion, took the lead in adopting the technology. So, let’s just agree that the clinicians do not like technology is a myth promoted to conceal the failure of the discipline of Information Technology in meeting the needs of the clinicians.
A question that is relevant for us who are working with Proteus in the Semantic Data Capture Initiative (SDCI) is how much will the clinician feel shackled by the pre-defined processes which are created based upon guidelines? Does the clinician really need to enter all the data in exactly the same order as prescribed by the process or the guideline? We have taken an interesting approach to address this. In the web interface we are developing to provide clinical decision support from the Proteus Engine, we will allow the clinicians to enter any data that any of the templates for the guideline/process require, at any time. However, they will only be able to submit the data of each individual clinical context (defined by the knowledge components) only when the executing process reaches that particular point. This allows us to have best of both worlds: constraining data submission based on the needs of the decision support system while allowing the clinician to freedom to type ahead if they feel like it.