Monday, 18 June 2018

Vinay Prasad: Why is 40% of clinical practice wrong?

On May 18, Vinay Prasad offered a conference in Barcelona as part of the 5th "Right Care" Conference of the Clinical Management Section of the Catalan Society of Health Management (SCGS), where we had the opportunity to invite him to explain why he had created (with Adam Cifu and other collaborators) the list of 146 clinical practices that would have to be reversed and what are the criteria they had been used.

What is medical reversal?

According to Prasad, a medical reversal is the need to stop a clinical activity because a well-done study, usually a clinical trial with finalist indicators, shows that in fact, the desired results are not achieved, or that the adverse effects do not compensate the benefits. The speaker gave some very diverse examples, such as the Swan Ganz catheterization to monitor the hemodynamic balance of patients in shock, the hormonal treatment for post-menopausal women in order to reduce coronary or cerebral vascular risk and the placement of coronary stents in patients with stable angina to reduce the risk of infarction, increase survival or even to delay the effort angina. In all three examples, consistent clinical trials have shown that these were clinical activities that, in practice, did not meet the set objectives, and in addition had side effects, which were not unimportant.

How it have been chosen the 146 practices to be reversed?

After reviewing 2,044 articles published in the New England Journal of Medicine in the "Original Articles" section for 10 years (2001-2010), researchers found that there were 363 that dealt with evaluations of established clinical practices, of which 146, 40.2% (hence the title of the conference), should be reversed because the published results contradict the evidence known to date. The list includes practices related to medicines, diagnostic and surgical procedures, implantation or use of devices, cancer screenings, etc., which shows that the reversals affect the entire spectrum of medicine.

How do you explain that the phenomenon is so widespread?

The deep causes of the phenomenon are three, said Prasad: a) The human psychology itself that is always ready to act. Not doing anything doesn’t appeal; b) People are optimistic by nature, we always hope that what we believe in will go well; and c) The entire health system, especially the US, is designed to encourage clinical activity and not the value it provides. It seems that by acting everyone wins.

Reversal is widespread because it’s very common to adopt innovations without consistent previous studies and, at this point, the speaker presented some favourable circumstances, such as when the novelty is accompanied by a convincing psychopathological argument, with the addition of some study based on anecdotal cases, or the appearance of epidemiological studies that are not well adjusted, or retrospective research that is not very rigorous.

To favour the launch of a new product, it’s very common to notice clinical trials that would not pass any validation filter, with inappropriate patient selections, insufficiently supported dosages, local limitations that prevent later demonstrating that these studies are replicable or that indicators of surrogate outcomes (such as the size of a tumour or the perception of pain, but not finalists) are used or even by misrepresentations and unclear publications.

Some of the conclusions of Vinay Prasad

Reversing is not without difficulties: on the one hand, it often takes more than ten years to achieve it, but on the other, there is a lack of confidence in the system. Prasad explained the case of the change of criteria of preventive mammograms from 40 to 50 years in the US and the bitter reaction of many groups that experienced it as a cut of serious consequences for their life expectancy.

The other aspect highlighted by the speaker is that many of the things that are done, also the most prevailing, often have negligible and modest results, especially in the medium and long term. In fact, seen in perspective, and taking into account the history of medicine, it’s not bad, but we should remember that nothing in clinical practice should be exempt from being evaluated.

Finally, Vinay Prasad said that the history of medicine is written by people who only think of the successes and that on the other hand, the failed procedures are not part of the ever thirsty for great discoveries scientific narrative. In this sense, he said, we should have the humility to analyze the results that don’t turn out as expected so we can learn from them. Prasad, with an optimistic final message, affirmed that the future of medicine will only be understood if we advance progressively in reaching higher standards of evidence and effectiveness.

Proposed solutions

According to Vinay Prasad we must advance in three points:
  1. Elaborate clinical trials before introducing any novelty, the technological ones included. We must bear in mind that the belief in a technique, if it also has an economic incentive, is a very addictive drug. 
  2. Priority should be given to the evaluation, with consistent studies, of common practices not tested, especially the most expensive and the most frequent ones. This strategy could save large amounts of money to healthcare systems. 
  3. Evaluative studies should be designed and conducted by neutral agencies. We must eliminate the current model in which regulatory agencies are forced to approve products based on studies that didn’t have sufficient control of neutrality. 
"To contradict my proposals, many tell me that clinical trials are too expensive and too slow," says Prasad, "and this is true under the current circumstances." For example, in the US clinical trials cost an average of $25,000 per participant, but, given its importance, we notice that there are possibilities of conducting clinical trials directly from practice, such as the case of TASTE that has randomized 60% of patients with coronary angioplasty at a cost of $50 per patient”.

Link to the Vinay Prasad conference: "Why is 40% of clinical practice wrong?"

Jordi Varela

Monday, 11 June 2018

Radiologists and incidental imaging findings

A group of radiologists from several American university hospitals (Massachusetts General, Cleveland, Brigham and Women's, etc.) started a debate in the Journal of the American College of Radiology about the eventuality that radiologists would stop reporting the incidental imaging findings lacking clinical significance. "The traditional role of the radiologist," they say, "is to warn of everything they see, leaving the interpretation of the findings’ relevance to the referring physician”. However, we now open the opportunity to go further, and not just intervene by saying, for example, that an observed abnormality is benign, but also taking the decision not to report the milder ones, given the possibility that our opinion generates confusion and ends up causing excessive medical actions".

Regarding level I renal cysts of the Bosnian classification

The radiologists who authored the article used the findings of renal cysts, which are very frequent with a prevalence of 36% in patients over 80 years of age, in order not to inform of renal cysts of level I of the Bosnian classification in their reports, in accordance with the following criteria: a) the cyst is not the reason for the examination, b) doesn’t generate local problems, c) has no malignant potential, and d) is not likely to generate a polycystic kidney disease.

Monday, 4 June 2018

Coronary Bypass and Hemodynamics: the amount matters

In the article "Comparing hospital performance within and across countries: an illustrative study of coronary artery bypass graft surgery in England and Spain", signed by a Spanish-English team in which Sandra García Armesto (IACS) and Enrique Bernal (REDISSEC) participated, it was concluded that the Spanish hospitals of the study operated in general with a smaller number of cases than the English (it was coronary bypass). Therefore, it is suggested that the number of cases intervened should be a tracer that could explain why mortality from this process is twice as high in Spain as in England.

European Collaboration for Healthcare Optimization (ECHO) is a European network of administrative databases for the analysis of clinical practice variations. In the following article: "Hospital Surgical Volumes and Mortality after Coronary Artery Bypass Grafting: Using International Comparisons to Determine Safe Threshold", carried out by almost the same authors as the previous one, based on data from the ECHO project, confirms that for interventions of coronary bypass there is a clear relationship between volume and mortality and concludes that the minimum limit of interventions of a cardiac surgery team, if you want to safeguard the safety of patients, should be 415 per year. In the following graph (from the previous article) it is observed how the Spanish hospitals that participated in the study (dark spots), generally underwent fewer coronary bypasses (many did not reach 200) and showed greater mortalities.