Cartesian versus confusion
e all know that medicine has nothing in common with rational thinking. When I started working in the profession, however, I thought differently. One of the main reasons I entered radiology was my perception that it was a rational and logical medical discipline. I thought that there are rational approaches to medicine, but I was misguided.
I was prompted to consider this matter following a conversation at a recent meeting. It is unusual to discuss philosophy during conference coffee breaks, especially when your background in philosophy is rather limited. The conversation started with the influence of Descartes, Leibniz, and Spinoza on medical life, and the Cartesian approach. But I had not the slightest idea what such an approach might be. This term reminded me of folding models I used to play with as a child. In philosophy, however, the Cartesian approach refers to Descartes, not to cardboard castles and ships.
A conversation between a Cartesian thinker and a philosophy student might run as follows:
"We are Cartesians. We are logical. We are not emotional. Unless we can see the solution, we are not interested."
"You start with the solution?"
"Yes, of course."
"But surely you have to start with the problem?"
"This is the mistake people make. You must start with the solution."
"And work back to the problem and therefore understand it?"
"But how do you know you are working back to the problem correctly and not making a mistake?"
"Because you can't. If you are being logical, you will always work back to the problem. Don't you see that there is a direct line between problem and solution?"
"But surely a problem can have lots of solutions."
"Agreed. But only one correct solution. Logic will give you the correct solution."
You have a solution and find a problem for it. I sometimes feel that this approach has become common to radiology over the past ten years.
I tried driving a BMW SUV for a few weeks.
I ended up using public transport.
Another conversation on an apparently unrelated topic set me thinking about parallels in radiology. This time, I was discussing my choice of new car with a U.S. psychologist, who started out by explaining the psychology of selling and buying cars. I asked her what she thought of me buying a sports utility vehicle.
"You are not the right type of person to drive an SUV," she said. "Psychological and market research conducted by the companies producing SUVs shows that the people buying them are different from you."
Internal research by Daimler-Chrysler has apparently concluded that people buying SUVs tend to be "insecure, vain, self-centered, and self-absorbed, who are frequently nervous about their marriages, and who lack confidence in their driving skills." This viewpoint was supported by a chief engineer with General Motors, who noted that SUV owners' primary anxiety is "I wonder how people view me.” 
SUVs are fashionable. The bigger and bulkier, the better. Their owners must be wealthy, too, because these cars are expensive. By buying an SUV they are showing off their wealth. Why buy a four-by-four vehicle for use in the city? Many buyers argue that they want to drive a safe vehicle, but these cars are less safe. They are three times more likely to roll over in a crash than a normal car. They have poor visibility and are extremely dangerous to pedestrians and other animals .
I tried driving a BMW SUV for a few weeks, and then a Volvo. It felt good, although I got a little seasick in the BMW and never found a parking place in the city with the Volvo. I ended up using public transport.
I asked the psychologist whether medical imaging equipment manufacturers have performed similar market research on their customers' personality profiles. She did not know. I watch new developments in diagnostic imaging with increasing skepticism. Why do we need 3T or even 7T MR imagers? Why have 64-slice spiral CT scanners? Do we have to perform patient examinations consisting of several thousand images – and store all these images in our PACS?
There are two approaches to performing radiology, although they go hand in hand. Number one, the patient is the focus of all considerations. Number two, radiology as a medical discipline is the center of thought. In the latter case, the patient becomes an object, rather than the center of attention. This becomes understandable when one considers that non-medical people are increasingly influencing radiology. Their primary interests are productivity, workflow, and lowering costs, not taking care of patients' medical needs and soothing their fears.
Prognoses and proposals for future developments are usually made by leading physicians (called opinion-makers) who work in tertiary healthcare, by which I mean university hospitals and medical schools.
Medical life in a university hospital is often different from primary healthcare, as delivered by general practitioners, and secondary healthcare, provided by community and general hospitals and specialist practices. People in tertiary healthcare have a different agenda.
Prognoses presented by parties with commercial interests are, in many instances, biased. They can also be Cartesian, if we use the definition described above. They have solutions for which they search for problems. This is not meant as an attack, but rather as a description of the distribution of human professions. It is not the task of companies to decide what kind of equipment would aid a rational approach to healthcare. In radiology, the radiologist makes this decision. If radiologists state that certain machines or techniques are pointless, there will be no market for them.
By the way, how can a radiologist be expected to read 20,000 images a day? I could hardly cope with 1500 images a day, and just reading those took me about three hours. The rest of the day was spent doing administration. Patient throughput and the number of images per patient have risen dramatically over the past decade. Whereas 10 years ago you had six or seven seconds to evaluate each cross-sectional image, today you have one or two seconds. Computer-aided detection will become essential if the number of images per examination increases further. But isn't CAD just a manifestation of the Cartesian approach?
Cartesians have solutions for which they search for problems.
CAD software extracts possible diagnostic information hidden in volumes of data generated by CT, MR, or other imaging equipment. It is the solution to the problem of having to read too many images. Why do we have that problem? Because people are pushing data technology. Numerous applications exist that offer new ways of visualizing disease instead of detecting and diagnosing it. Is this medical progress? Where is the advantage for the patient?
There is a difference between creating patient benefits through new developments and making life easier for doctors, nurses, and bureaucrats. One day somebody will stand up and ask whether all these technological advances are necessary. I do not expect this to happen in Europe, but most likely it will happen in the U.S. Should the U.S. Senate hold an official enquiry, nobody would be able to prove the advantages of such equipment. You can show with clinical evidence that smoking is dangerous. No such outcome studies exist for 32- or 64-slice CT, however.
Organizers of this year's European Congress of Radiology in Vienna held a symposium for hospital administrators entitled "Investing in medical technology and information technology innovations." Prof. Dr. Maximilian Reiser, chair of radiology at Ludwig-Maximilians University in Munich, Germany, and co-moderator of the symposium, summarized the position of management and radiology within healthcare today .
"Hospital work flow and processes must be organized to achieve high effectiveness at reasonable cost, so that medical teams can concentrate on their job – patient care – and this only will be possible if management, doctors, and nurses cooperate in a trusting environment where antagonism makes way for corporate identity ..."
I have now decided to adopt a Confucian approach to new developments in radiological technology – through pure confusion. But I believe a better approach should be available, as Reiser seemed to indicate.
"Being a doctor has become imperceptibly – but increasingly – less attractive, due to an overkill of legal and bureaucratic requirements, as well as escalating requirements for documentation," he said.
Disclaimer. I do not mean that radiologists at university hospitals and medical schools are not lucid and that no outstanding healthcare originates in their departments. Saying so would involve shooting myself in the foot. The same holds for companies.
1. Thompson D. Crash-obsessed dummies. Guardian Weekly 2004; 170 (23): 18.
2. Reiser MF. Investing in medical technology and information technology innovations. What's the value? ECR Today 2004; 5 March: 30.