New Times,
New Thinking.

  1. Science & Tech
10 June 2019updated 08 Sep 2021 8:01am

Placebo surgery raises ethical concerns – but this was not placebo surgery

This is a response to Professor David Papineau’s article questioning whether Placebo Surgery can ever be ethical. You can read that here.

By Andy Carr, David Beard and Julian Savulescu

David Papineau raises some important points about deception and therapeutic misconception in medical research, but this is the wrong trial to pick on to make that point. Three of the authors of the ethics article in the Journal of Medical Ethics which he approvingly cites were also investigators in CSAW. We are well aware of the importance of an honest, transparent and intelligible consent process. It is because of concern with the ethics of such a trial that one of our number, Andrew Carr, invited Julian Savulescu – a professor of practical ethics and expert in medical ethics – to be involved at the stage of trial design.

The first point he brought up was that there must be no deception. They together wrote an article with another investigator to explore more deeply how surgical placebo controlled trials should be conducted ethically and believe this trial conformed to those standards. This built on a widely cited systematic review of the results of all surgical placebo controlled trials published in the BMJ led by the same authors. It is hard to imagine a more rigorous process of ensuring a trial is designed and conducted ethically.

The term placebo can be defined in various ways: we chose, after careful consideration, to define it for the purposes of trial design and description as “surgery with the critical therapeutic element omitted”, in this case the removal of the bone spur. The established rationale for the operation is that a spur of bone is causing pain due to mechanical contact with tendons and soft tissue during movement and that removal of the spur will cure the problem.

The patients were very carefully and clearly made aware in the consenting process that if randomised to surgery they had a 50 per cent chance of having the spur removed and a 50 per cent chance of it being left untouched with no tissue removed. The trial design, consent process and patient information leaflet were considered and approved by an independent NHS Health Research Authority Research and Ethics Committee. These committees, which include patients and lay members as well as clinical trial experts, have the dual mission, firstly, to ensure the protection of the rights, dignity and wellbeing of research participants and, secondly, to promote ethical research that is of potential benefit to participants, science and society.

Patients involved in the CSAW trial were not “left in the dark” nor was there any “sleight of hand”. They were informed that there was uncertainty (equipoise) between three interventions:

Arthroscopy Only, where the patient undergoes a general anaesthetic has a telescope introduced into the shoulder, a diagnostic inspection is performed involving a fluid wash out but no tissue is removed. After the surgery the patient rests and undergoes postoperative physiotherapy.

Arthroscopic Subacromial Decompression – identical to Arthroscopy Only but with the addition of the surgical removal of the bone spur, the “critical therapeutic element”.

Give a gift subscription to the New Statesman this Christmas from just £49

Active Monitoring and Observation Only, where no treatment is given and the patient is observed to determine if natural history might result in an improvement in symptoms.

In this trial, the arthroscopy only group, where the bone spur is not removed, could have had some potential benefit as a consequence of the potential diagnostic value of the arthroscopy alone, from washout of the joint, from the rest associated with the surgery, from analgesic medication associated with the surgery and from post-surgical physiotherapy. While it would not be routine clinical practice to do arthroscopy alone, it is not an inert intervention; it is different to a sugar pill. For this reason, we have consistently and clearly identified that the term placebo is problematic in surgical trials. This is why the patient information leaflet was worded as it was.

There was genuine uncertainty about the effectiveness of this commonly and increasingly used procedure. This uncertainty proved to be justified and this trial has resulted in major policy changes globally with reduction of unnecessary risk to patients and costs to health care providers. While there were benefits to patients outside the trial, the trial participants who received “placebo” were benefited: they were spared a useless surgical intervention that has risks of its own. They were not used purely for some utilitarian purpose.

Significant effort was made besides providing an accurate and informative plain language statement to avoid deception or therapeutic misconception during the surgeon-patient interaction, which Papineau does not discuss. Patient representatives were involved in the design, conduct and reporting of the trial. All recruiting surgeons and sites were visited and trained in recruitment. The trial team included the QUINTET group from the University of Bristol, who have expertise in recruitment to challenging clinical trials.

As part of their contribution to the trial audio recordings were made of consultations and the consent process to understand and improve the recruitment process. This process identified key points which were made available and emphasised to all the trial surgeons and sites:

Following diagnosis, explain that there is no clear evidence on the best way to treat their condition (impingement);

Inform the patient that the centre is participating in a study comparing three options for treatment to find out the best way to treat impingement;

Describe the three study arms (emphasising that there are advantages and disadvantages to each);

Briefly describe the randomisation process;

Invite questions from the patient;

Ensure that each of the three treatment arms is presented in a balanced way. Occasionally, the active monitoring arm was “over-explained”, which may lead to uncertainty from patients about this option;

Explore the reasons for any patient preferences, so that you are clear that these are based on accurate information, before mentioning the patient preference survey;

Present the study in an enthusiastic (non-apologetic) manner, as patients often gave altruistic reasons for taking part in CSAW, consider presenting the benefits of doing so – for example, assisting with finding a solution to a treatment dilemma. 

The CSAW trial featured in a recent BBC Horizon Programme “The Placebo Experiment can my brain cure my body”, first shown in October 2018. This included an interview with a patient who took part in the CSAW trail (CB).  We believe it is clear from this interview that there was no deception. The same patient (CB) was also interviewed by Xian Rice for the article in the New Statesman that Papineau refers to.

CB has read the Papineau piece and comments:

“The Papineau paper puts forward no supporting evidence but only a series of hypotheticals. It misconstrues consent to the trial as a limited paper exercise, whereas in reality for CSAW it was a multifaceted process involving a dialogue between patient and medical practitioner as well as further members of the trial team. As a patient participant but also a lawyer with some interest in medical law (capacity and consent), I can attest that my experience with the CSAW recruitment and consent process adhered to the highest standards possible. At no time did I feel deceived, confused or misinformed about the meaning of ‘placebo’ or the design and objectives of CSAW. I am delighted to have been included in this important study, whose results had beneficial effects for both for me personally and, I believe, the NHS and wider medical community.”

The trial was also supported throughout its design conduct and reporting by a patient representative (D F-H) who was also a member of the trial steering committee. D F-H has also read the Papineau piece and comments:

“Having had surgery on both shoulders following a serious road accident, I have been involved in shoulder trials for more than nine years as a patient representative.  From initial design of the CSAW trial, through conduct, reporting and the aftermath, ethical parameters have been paramount.  The charge of patient deception therefore is a figment of Professor Papineau’s imagination.

“It is refreshing to find so many surgeons admitting that they no longer know the best way to repair damaged shoulders.  From the patient’s perspective, without more trials such as CSAW, potentially harmful and costly procedures are likely to continue unchallenged around the world.  Such a state of affairs would surely be unethical“.

The core message of Papineau’s article is very important – but he has unreasonably and incorrectly criticised the CSAW trial. This piece very unfairly represents what actually happened in the CSAW trial. Patients were not deceived. Such a piece stands to jeopardise well designed ethical placebo-controlled surgical trials which benefit present and future patients.

Andy Carr is Professor of Orthopaedic Surgery and head of the Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences at the University of Oxford. Professor David Beard is professor of Musculoskeletal and Surgical Science & Fellow of Kellogg College, Oxford. Professor Julian Savulescu is director of the Oxford Uehiro Centre for Practical Ethics.

Content from our partners
How the UK can lead the transition to net zero
We can eliminate cervical cancer
Leveraging Search AI to build a resilient future is mission-critical for the public sector