(First East European Conference on Health Care Modelling and Computation - HCMC 2005, held in Craiova, Romania, between August 31 and September 2 2005.)

Editor’s comment: Four mind map drawings were used to underpin discussion at the Craiova workshop. An elephant on a shelf; a basic set of tools, an umbrella in the sky and seven mountain peaks. The workshop discussion focused on the chasm to be crossed between sophisticated 21st Century data analytical tools and 19th Century measurement used in modern hospital practice. Below Roy Johnston comments, the mind map pictures are shown, and the Editor has the last word.

Buying the elephant. If you buy a china elephant, or any other object, you only need tools to put it on a shelf if you have nowhere else to put it. The new shelf must compliment your new possession, hold the weight and look good. If you don’t buy the elephant and / or need a new shelf, then brackets, screws, screwdrivers and electric drills will remain in the toolbox and never be used.

Different viewpoints. Everyone sees the world from the point of view on which they stand. Imagine an umbrella in the sky. What will passers by see? Three people, one underneath, one on a hilltop and one in an airplane will all have different views. Different people, different angles, different views of the same umbrella. Of the elephant. Of the tools we need to put it on a shelf.

Different questions. What tools reflect the elephant? Are the tools we choose suitable for the elephant? How can we modify the tools we possess to suit the problem and how can we build new ones taking into account the old ones? How will the elephant evolve?

Nosokinetics is our elephant. We want to build a better world, and think our tools are useful, how do we sell the idea, if no-one wants to buy? What are the problems and what are the possible solutions? How can healthcare issues be modelled optimally? Toolmakers from around the world, gathered at the round table, brainstormed some ideas.

What are the developers’ faults? The healthcare system depends on the country; what kind of tools do they build? Can their tools adapt to changes in the healthcare system? Are their tools more important than the problem? Are they more concerned in developing tools just for the sake of tools?

For us all: Do not “Mind the gap” in knowledge between toolmakers and doctors, but bridge it! Make a liaison between the people who make the models and the people who need the toolkits. Build the new science to make an interface between academics and buyers; that could also take care of the problem of language complexity.

Only when we achieve that, will our Nosokinetics elephant shine happily on the shelf so nicely and firmly pinned to the wall of our home.

Roy Johnston, who works with IMS and has a background in applied maths, gives his initial reaction:

Peter:
On the matter of the 'elephant' is this the best you can get out of that conference? I must say I do not think it is illuminating, with its dubious analogies and implication of 'white elephant' status for nosokinetics. What is more, it gives a wrong message, namely, that the stakeholders are the modellers and the medics.

There are at least 6 stakeholder-groups, namely the managements, the doctors, the nurses, the special service technicians, the suppliers of IT systems, and above all the patients. The latter are the most important in their own right, and they can also influence the managements via the political process.

I do not count the modellers as stakeholders; they are at the meta-level, and their problem is to sell the understanding of the problem structure to management and their associated medical service experts, their potential allies being the IT system suppliers, the patients and the politicians.

The stake of the IT system suppliers is related to their need to support management and clinicians in the avoidance of problem situations. That is why we in IMS are here doing this. So on the whole it would go against the grain from me to edit the 'elephant' paper without serious critical comment.

A final comment, on the nature of the modelling problem. It is evident that we have a strong stochastic component in the arrivals, and a variety of service time distributions. The servicing of a patient with a given condition is likely to involve a complex set of services, for each of which there can exist a queue. The challenge to modellers is how to adapt the existing extensive and excellent body of queue-theory to describe meaningfully this complex situation, and to lead management to an understanding of the extent to which they need to supply surplus capacity in critical service areas, given that in a stochastic environment, 100% service utilisation implies infinite queues.

Sincerely / RoyJ

Peter replies

Roy:
Thanks for your insights. I should have told you that the discussion was based on four mind map drawings. Clearly, the working paper struck a nerve. I thought about rewriting, however, it is the work of a PhD student and they are our future.

Figure 1. The elephant, tools, umbrella and knowledge gap used in the workshop.

With the mind map drawings the message is clear - and the message is an important one. If the mathematicians and computer scientists who are building methods of data analysis don't talk to / work with health care practitioners, they may create meaningless models. On the other hand, if the doctors cannot understand the language that model makers use and / or the tools are not user friendly, then no progress will be made, and models, useful or not, will languish in the dark and never enter the light of day.

It would be interesting to know what our readers think and I hope many will be stimulated by Ruxa’s paper and your response to write in and tell us their views.

Yours / Peter

Some navigational notes:

Copyright (c)Roy Johnston, Peter Millard, 2005, for e-version; content is author's copyright,