Albert Einstein, or rather a life-size cardboard figure of Ulm’s famous mathematician, is looking over Franz Porzsolt’s shoulder. One of Einstein’s famous aphorisms is attached to the cardboard Einstein: “Not everything that counts can be counted, and not everything that can be counted counts.” The quotation is also a very fitting description of the work of Porzsolt, a 64-year-old clinical economist, specialist in internal medicine and oncologist who enjoys turning accepted truths on their head. Many colleagues regard him as an outspoken critic of others for his tendency to refer to science as theory and clinical settings as practice. Porzsolt criticises evidence-based medicine as it is currently understood and applied.
Porzsolt does not reject the “evidence-based medicine” tool per se, but sees it as a subjective decision rather than a factual-normative element. If evidence-based medicine is correctly understood and conscientiously applied, it guarantees doctors the freedom of choice, expanding rather than restricting the possibilities in decision-making, says Porzsolt in answer to his critics.
The major problem is that we "repeat what other people have said, without really thinking about what we are saying. All fields where progress is achieved require people who try to get to the bottom of things." And this is what Porzsolt, experienced clinician, long-standing scientific secretary of the Tumour Centre in Ulm, and leader of many clinical trials, has done: he scrutinises, looks for mistakes and errors and has developed a growing scepticism of the dogmas of medicine.
This process, which turned against scientific reporting that was entirely centred on success, and rarely considered errors and mistakes, started around twenty years ago. Porzsolt questioned whether "everything that patients had to accept, and even put up with, from doctors, was returned to them as benefit." "Without realising it, I slid into an economic analysis of listing the costs on one side and the consequences on the other and eventually questioning whether alternative therapies were available and could have been used."
Porzsolt has recently submitted a paper to the European Journal of Clinical Oncology whose results he presented at an international conference in Milan a couple of months ago. These findings will turn 50% of oncology on its head,” Porzsolt predicts. He is shaking existing dogmas to the core by critically analysing the so-called surrogate parameters used in oncology.Porzsolt explains that new therapies (with new substances) are normally tested by comparing them with existing therapies. The survival rate of people undergoing these two types of therapies is determined and compared. However, these measurements usually take a very long time, which is why progression-free survival (PFS) or time-to-progression (TTP) are taken as surrogate endpoints for median overall survival. This procedure is standard in the field of oncology and has also been standard in granting marketing approval to drugs for the last 20 years. In a metastudy carried out with some colleagues from Ulm (Weber, Kirchheiner, Muche), Porzsolt found out that these surrogate parameters are not suitable for predicting the overall survival rate. The results of this study are also becoming more influential due to the FDA’s current practice for granting marketing authorisations to drugs, Porzsolt explained. The FDA cleared nine out of 22 anticancer substances purely on the basis of “time-to-progression” data, without knowledge of the survival rate. “When our data are published, this type of approval practice will become obsolete,” said Porzsolt.
Over the last 20 years, the ‘rebel from Ulm’, as Porzsolt is sometimes called, has identified numerous ambiguities in how evidence is generally understood. This led him to develop an approach which he calls ‘clinical economics”. And he freely admits that he has fallen between all stools. He remains true to his conviction that it is more important to scrutinise the benefit of a certain therapy before talking about its costs.
What is clinical economics? It is based on three pillars – clinical epidemiology, economy and ethics. Porzsolt understands clinical economics as an attempt to achieve the best possible results with the least effort. In contrast to health economists, clinical economists look “through the eyes” of doctors and see themselves as patient advocates rather than defenders of their own discipline. Porzsolt does not deny that economic assessments have an important role to play, but the starting point is nevertheless an analysis of therapies rather than costs.In a recently published paper (“Folgt auf die ”Evidence-based Medicine“ eine ”Confidence-based Medicine“?, in: Medizinische Klinik 2010; 105:560-566), Porzsolt and one of his colleagues from Ulm, Heiner Fangerau from the Institute of History, Theory and Ethics in Medicine, summarised their concern about evidence-based medicine and listed this concern under five aspects.
Porzsolt and Fangerau believe that the term “evidence-based” is often used as a phrase that lacks real meaning and kills all discussion: those who use evidence-based medicine in a professional context are well aware that it is only a tool for rationally substantiating decisions and that it lacks any normative power. It has come to the point now that doctors initially use their own, internal evidence to solve certain problems before looking for and critically assessing external evidence, said Porzsolt.Criticism no. 2: With regards to content, the validity of evidence-based medicine is not understood and unknowingly, is not taken into account. For example, Porzsolt and Fangerau regard it as a false conclusion to believe that life can be prolonged when an early diagnosis of a disease is made. This assumption does not take into account that early diagnosis potentially only moves forward the time of diagnosis itself without having any effect on the end of a person’s life.Some statements lack validity even in renowned journals. Porzsolt refers to one of his studies carried out in Germany and Brazil (TACT: Twin Assessment of Clinical Trials) that clearly shows this. He found that 30% of all studies published in top-class journals do not even mention the objective of the trial, i.e. post an accurate lead question. This result was achieved in a doctoral thesis carried out under his supervision. Furthermore, Porzsolt criticises the fact that many guidelines lack transparency. In addition, such guidelines are influenced by the interests of the association that publishes them, and hence vary considerably. The risk of such a “guideline apoptosis” can only be reduced if everyday, systematically collected experiences are taken into account. Porzsolt also believes that randomised controlled studies – the current gold standard for collecting scientific evidence – have ethical and empirical limits and are unsuitable as sole source of insight, for example because they are not suitable for describing therapeutic interventions in real-life situations.
Porzsolt’s survey, in which he interviewed 275 staff members at Ulm University Hospital (from advanced medical students to senior physicians) on their thoughts on evidence-based medicine, shows that scientific evidence does not always reach clinical settings. The survey included eleven questions on issues such as the objective of clinical trials, trial design, patient risk profiles, blind studies and conflicts of interest. The survey showed that the results obtained when critically assessing clinical trials varied considerably if not all or none of the validity criteria were fulfilled. Porzsolt is convinced that the results from Ulm are transferable to any other German hospital.Porzsolt was not at all surprised about the results. It takes a lot of time to evaluate clinical trials thoroughly and critically. The evaluation also requires people to be interested in methodological details and to have practical and theoretical experience. Porzsolt thinks that clinical experts frequently lack all of these three things but he is convinced from his own experience that motivated and specifically trained medical students can fulfil these requirements.
Porzsolt believes that a health economy approach that seeks to reduce increasing healthcare costs without taking into account potential consequences is a fatal knee-jerk reaction. Franz Porzsolt has come to the conclusion that the decision to introduce health economics into the curriculum of medical students was wrong. He finds it more important to teach medical students the basics of economics, which must then be translated into the theory of medicine. Porzsolt teaches his medical students the principles of clinical economics under the guise of health economics.
Porzsolt has fallen out with one of his American co-authors (an economist) of the book entitled "Optimizing Health: Improving the Value of Healthcare Delivery" due to an unresolvable dispute: Porzsolt's co-author is convinced that clinical economics needs to be oriented around the welfare of the state, while Porzsolt is firmly convinced that the welfare of every single individual is of primary interest.
Porzsolt is well aware of the problem of assessing the benefit of healthcare delivery as "two aspects need to be taken into consideration: costs and consequences". He also knows that it will be tremendously difficult to transfer the economic approach to patient treatment and believes that a thorough cost-benefit analysis would require closing down all intensive care wards or abolishing rescue helicopters. He believes that it is possible to save a person's life with less expensive options.
In Porzsolt’s approach towards more beneficial, productive medicine, conflicts seem to be inevitable every time an established medical treatment has been proven to have no practical value. In order to make this transition from useful to useless socially acceptable, Porzsolt believes that options to enable doctors to change how they treat patients need to be made available. Porzsolt envisages clinical economists providing instruction to specialists on how to work according to the principles of clinical economy in their respective fields. This ‘pilot’ function assumes that hospital directors have a high level of interest in this approach, and Porzsolt knows that they are interested very well from his own experience. Scientific evidence is suitable for showing the effect of preventive, diagnostic or therapeutic principles (under the idealised conditions of a clinical trial) or their efficacy (under everyday conditions). Porzsolt believes that effect and efficacy are unimportant in terms of pricing, which instead depends on the value or benefit of preventive, diagnostic or therapeutic measures.
Standardised methods that take into account the aspects mentioned in the preceding paragraph are still lacking. Porzsolt therefore suggests “submitting innovative products to further tests three to five years after a drug has been granted marketing authorisation and a decision on reimbursement has been made”, something that is standard in France. This would help to show whether newly developed drugs only have a medical effect or whether they are actually useful. Porzsolt believes that current thoughts on the new regulations for the German pharmaceutical market (AMNOG; Arzneimittelneuordnungsgesetz), which envisage the analysis of benefits immediately after a drug has been granted marketing authorisation, are meaningless.Porzsolt thus addresses a dilemma associated with the granting of marketing authorisation. Generally speaking, the only data available when a company applies for marketing authorisation are data obtained from randomised clinical studies, carried out under artificial conditions and only involving a small number of patients. Moreover, the acquired data are assessed at a time when no evidence is available. Porzsolt believes that in contrast to England, France or Scandinavia where the regulatory authorities deal a lot more pragmatically with the degree of evidence obtained in scientific studies, German regulatory authorities base their decisions on a rather purist assessment of evidence. Franz Porzsolt is currently designing larger studies that focus on finding out whether an effect that has been shown in idealised trials can also be found under everyday conditions. He has already collected initial evidence, but declines to disclose any more at this stage: “We are generating totally unrealistic data under the idealised conditions of a randomised study.” Franz Porzsolt is firmly convinced that his way of thinking has a great future because he believes that more and more data will substantiate his beliefs. However, he is unsure whether he will be able to harvest the fruits of his 20 years of work. At least, the 64-year-old is once again part of the University of Ulm. Since 1st October, Porzsolt has been a member of the department of Holger Fangerau, director of the Department of History, Theory and Ethics in Medicine. The two are very keen to publish their thoughts.Porzsolt is unsure whether or not the University of Ulm will be able to create a unique selling point through the establishment of the ‘clinical economy’ course. However, he is secure in the knowledge that he has the backing of the Society for Medical Education. At the society’s recent annual meeting, its members expressed their enthusiasm for Porzsolt’s concept.
Literature: Porzsolt F., Pressel H. et al.: Appraisal of health care: from patient value to societal benefit, in: J Public Health (2010) 18:297–302DOI 10.1007/s10389-009-0294-1Porzsolt, F., Fangerau, H.: Folgt auf die "Evidence-based Medicin" eine "Confidence-based Medicine"?, in: Med. Klinik, 2010; 105:560-566 (nr. 8).Porzsolt, F., Schreyögg, J.: Die wissenschaftliche Evidenz und der Preis für Innovationen im Gesundheitssystem, in: Med. Klinik, 2009; 104; 622-30 (Nr. 8)Porzsolt, F. Kliemt, H.: Ethische und empirische Grenzen randomisierter kontrollierter Studien, in: Med. Klinik 2008; 103; 836-842 (Nr. 12)