Archive for April, 2008|Monthly archive page
Can Health Information Systems be Simple and Complex? yes.
Leave a CommentI came across this interesting article by Panatazi and colleagues (2006). They argue that health information systems cannot be usable and useful at the same time. Information systems are of two parts: a user interface and a problem solving engine. One of the goals of the problem solving engine is to be useful through solving medical problems. Medical problems tend to be complex. This complexity will eventually creep into the the user interface making it less simple and though less usable. There are many examples in reality. If you look around you will notice that usable systems tend to solve simple problems only. Calculators are highly usable and simple but only solve simple problems. Once you get into the more advanced scientific calculators you loose the simplicity and uablility.
The authors make a vary good argument, and shed light on some solutions that need not yet present technologies. But, they miss on one critical factor. That is the user brain. I believe the best way to solve this “usability usefulness paradox”, as called by the authors, is to allow the user brain bridge the gap between usability and usefulness i.e. have a simple and usable interface that allows that user to solve complex problems. The other part of information system, the problem solving engine, would only perform jobs that would not interfere with the user cognitive processes and would not add complexities to the user interface.
A Summary: Getting to the point: developing IT for the sharp end of healthcare
First I need to explain what is meant by the sharp end of health care. Health care systems can be looked at as a wedge with a sharp edge and a blunt edge. The blunt edge is where management lies. Clinicians directly caring for patients are at the sharp edge. Cognitive work at the sharp edge is more complex and less supported by IT. Unfortunately, there is simplification of these complexities. Changing and conflicting guidelines and varied patients response to treatments are two examples of ignored realities.
Clinicians are the most adaptive elements at the sharp edge end yet their needs must be supported. Developing user centred systems in this environment is oversimplified. Actually, this is an impossible job. The variability between clinicians prohibits reductionist approaches. Yet sciences as psychology, human factors and human-computer interaction should help. Naturalistic decision making approach can be used.
At the sharp edged, the coordination between human and machine needs better understanding. This is inlight that IT can help as can hinder cognitive work.
The authors call to improve the clinicians ability to perform despite changes and challenges; in other wards, to improve resilience.
Goodbye “Single- Sensor- Single- Indicator” Displays
This study used a display method developed using iterative design method. The goal is to replace the numbers displayed on ICU patient monitors with representational graph. The current number based systems use “single-sensor-single-indicator” display paradigm. This requires clinicians to integrate sequential, piecemeal data to reach an understanding of the patient’s condition. This is complicated by the fact that 67-90% of the generated alarms by monitor devices are false positive.
Using graphs generated by multiple sensors would aid the situation awareness and though the prompt response by clinicians. It has been shown that monitoring systems that increase situation awareness shorten the time between the occurrence of unexpected events and correction of the events.
This study is one of multiple that evaluates similar displays. This study focuses on this sort of displays in ICU settings. The study consisted of observation and questionnaire submission. The result shows that clinicians did observe this display and they think it is a desirable addition to current ICU monitors.
The vital thing in such displays is that they generate an easy to understand graphs that pull data from different sensors and though decrease the cognitive load. This should lead to better decisions.
Alerts as the Ones in Computerized Physician Order Entry Systems Harm Physicians
Heath information systems are plagues with alerts. In studies done on Computerized Physician Order Entry systems it has been shown that physicians are flooded with alerts that hardly every change the previously intended action. These experiments in this interesting article prove that these mainly useless alerts have and added negative effects on memory! Alerts decreased the memory accuracy of subjects in this study.
Everyone knows that mistakes in prescribing medication are done daily, and can lead to the loss of human lives, and human suffering to the minimum. But, alerts are not the way to do it. The old way of pharmacist checking on physicians prescriptions seems the methods we should be using till better ways of improving our prescription practices are found.
The Importance of Addressing the Environment When Designing Healthcare Information Systems
Follows is my review of an interesting article titled: artifacts and collaborative work in healthcare: methodological, theoretical, and technological implications of the tangible.
There are things when you read, your brain just refuses to let go of them. This article by Dr. Yan Xiao is one of these. Surprisingly, Dr. Xiao was able to squeeze this article in six pages!
This article is a call to reconsider of how IT is implemented in health care. We humans communicate in explicit and implicit methods. We can appreciate if the other person is paying attention or is upset implicitly by looking into their eyes and listening to their voices. IT focuses on explicit communication. And, unfortunately poorly designed systems impede the vital implicit communication. This is obvious in the technique of work-flow abstraction. Data and information flow is plotted with total ignorance to the physical environment. This physical environment includes us humans and the artifacts we use. Humans over years have devised ways to use these artifacts to optimize communication and cognition. The use of books, paper, and white boards are examples. Dr. Xiao gives examples as the white boards used in emergency rooms and operating theaters. He also mentions and example presented by Boguslaw and Porter (1962) present in restaurants’’, the “spindle wheel”. Read more »
