Titin.net

I am reading a very interesting article by Johnson, Johnson and Zhang (2004). This article has suggestions on how to redesign the interface of an existing healthcare system. The trigger to write this article is as stated “numerous health care systems are designed without consideration of user-centered design guidelines”. They point out the need to understand your users. They then go and divide users into expert users and novice users. Expert users need the rapid response time and shortcuts and the novice users need a lot of feedback. I do not think this sort of discrimination is necessary. Any user needs feedback, rapid response time and easier ways to do repetitive tasks (shortcuts). If you check Johan Redström winning article I commented on, there is no such thing as “a user” before the the system is in operation. There are only people we think would use the system we are developing. Discriminating between people in such a way is just to give excuses for developing suboptimal interfaces. The ground breaking iPhone interface proves this. There could be and interface that any person can understand and use.

The only way I see that a similar discrimination makes sense is if you focus on the tasks that the system will perform. There are systems that are meant to be used by system administrators and database administrators. These are mainly servers. These systems avoid the overhead of a Graphical User Interface (GUI). These systems need to squeeze the fastest response time possible out of the available hardware. But, the division here is based on the system task and not the people that will be using the system.

Ignoring Fitts’s law can exacerbate one of the unintended consequences of information technology in health care. This is juxtaposition errors.  This is when a user clicks the item next to the intended one. So, a physician can prescribing medicine to the wrong patient, because s/he clicked on the wrong patient. This is not theoretical. This was one of the errors elucidate by Ash, Berg and Coiera (2004) when they were discussing the errors induced by the highly interruptive environment in heath care.
You can listen to Jared Spool on the Usability Tools Podcast describing the use of Fitts’s law in computer interface design. In short, things you need users to click on should be big and close.
Surprisingly Fitts’s law is quite old. Paul Fitts, an American psychologist, published his paper in human motion in 1954. He developed his theory while trying to improve aviation safety. He may have realized that “juxtaposition errors” when made by pilots can be deadly! This is another development in aviation safety that health care should catch up with.

check my other related post

It is challenging to develop systems for the future. It is as hard to make worthwhile research. Everything in IT is changing. The technology is changing fast. It is al most hard to predict the future. It seems the only way to predict it is “to make it or to look for unevenly distributed technologies or phenomena.” (These two quotes are not mine)

We humans are also changing. We develop new skills and improved or loose our current ones. Even our brains’ networks are changing change.

Before mentioning my thoughts on design research, I must guide you to this superb description of design research on isworld.org .

Design research is somewhat similar to any other research. The only difference is that researchers come up with an artifact then test it as apposed to coming with a hypothesis.

I do believe that design researchers should focus on the concept and not the design itself. This may require just a mockup instead of a prototype. So, there should be a balance between the focus on the question needing an answer and between how elaborate is your design. For a question as: how much information should be displayed on the main page to support decision making? would not require a prototype. A mockup with multiple screens should allow us to perform required tests.

Design research is not good for every thing. But, it looks like a good tool to be used when the goal is to narrow the gap between what is known and practice.

Ontology, Epistemology and Methodology

This is a summary of my readings from module one in my course. Ontology, epistemology and methodology are three related concepts. Ontology is the building blocks of reality, or the nature of reality. Ontology can be objective, as the speed of light and can be subjective where different perspective are present as when we start our senescence with “it dependence”. Realist treat reality as objective and interpretivits treat reality as subjective. (Tansley 2004) We use epistemolgy to reach new knowledge that is based on our building blocks (ontology). Methodology is more practical; it deals with methods, systems and rules we use to conduct inquiries. This definition makes methodology somewhat similar to epistemology. Epistemology is more of a philosophical construct that would answer questions as “How can we be sure that we know what we know?” (Guba & Lincoln 2006) So, Epistemology is the “how”.

Paradigm

Paradigm is another concept. Paradigm is a collection of related concepts or a philosophical genre. (Weber 1997)

We should not design the use. Unfortunately, this was an unintended outcome of usability research. The Author argues that we should go back and focus on the designed object itself. We should aim at designing objects that people will probably use. People who become users of this object should be able to use it whatever way they please. These users will then have a user experience. This is opposite to the designs that start by designing the user experience and ending up with the design.

This is the 2006 Design research society design studies award. This study is by Johan Redström and is published in Design Studies Vol. 27, No. 2, pp 123-139.

Based on this, a suggestion to design the perfect clinician interface system is to first list all types of data that is needed as labs, imaging, problem list. Then, develop a way to view and access these. The opposite, that is if we focus on the user first, is to develop use-case scenarios. Then design the clinician interface based on these scenario.

Lately I been having a thought: do we need health informatics? Could Information Technology (IT) professionals do the job? Some may say that heath informatitions are IT professionals. But, health informatics is different. You can realize this if you look into informatitions career paths. I have no statistics, but a lot have a medical background. This is clear if you focus on the presence of Nursing Informatics as a recognized specialty at some colleges.

I started raising this question as I am studying system analysis (for my health informatics degree). System analyst’s duty is to analyze and design new information systems. So, a system analyst could go on a journey to explore healthcare institutions and, then, design the perfect healthcare system. Systems are designed by teams. These teams will have users. In our situation users will be healthcare professionals. These teams will also have programmers. Do we really need health informatitions with expertise in computer and healthcare?

Information systems are improving. In the past Information technology professionals were more involved in the day to day operation. Now, users expect a “perfect system” that seamlessly provides them with the information they need (with no IT person bugging them). They expect an internet like experience. What I am trying to say here is my understanding of what was put nicely the authors of System Analysis and Design seventh edition (Sally, G., Cashman, T. and Rosenblatt, H. 2008): the IT group became a supplier of information technology, rather than a supplier of information.

If your answer to my question (do wee need health informatics?) is yes we do need it, then we need a different and demanding form of heath informatics. Health informatitions will be needed during the development of systems. Health informatitions will need to have more computer science expertise. I believe that only someone expert enough in healthcare (knowledge domain) and computer science can know what is possible and what is not possible in healthcare. Those developers who don’t understand how medicine is practiced go overboard with technology designs. And, healthcare professionals who don’t know what is possible don’t know what to ask for.

Note: in my post health informatics is medical or clinical informatics and not biomedical informatics.