Enhancing the QUAlity and Transparency Of health Research

Why did we develop TIDieR-Placebo?

23/09/2020

Why did we develop TIDieR-Placebo?

The TIDieR reporting guideline helps authors fully report their interventions, so that they can be used by others. However, TIDieR only explicitly covers active interventions, not placebo controls. The developers of the new TIDieR-Placebo extension explain why it is needed.

Placebo ingredients influence how effective or harmful the active treatment, with which the placebo is compared, appears. The idea the placebos and placebo ingredients are all inert is a myth, so reporting what’s in placebos is necessary to appraise placebo-controlled trials and implement their results. TIDieR-Placebo (1) is a new guideline that helps researchers report what’s in placebos or sham controls with the same rigour that active interventions are reported.

Placebo ingredients influence apparent treatment harms

In some COVID-19 vaccine trials, participants in the control group (the group receiving a placebo) are injected with a saline solution. In other trials, such as the one developed by the University of Oxford, the control group receives a meningitis and septicaemia vaccine (2) as a placebo.

Using an actual vaccine as the control is good because it will cause similar infection-site reactions as the COVID-19 vaccine. Participants and trial staff won’t get an unexpected clue about which treatment they receive, preserving blinding.

In unblinded trials where participants know whether they are getting the real treatment, their expectations can make them get better (a bit) faster (3). And if they know they are getting the placebo, they could drop out of the trial (4) because they know they aren’t getting the actual treatment. Adding an actual vaccine to the placebo control helps the trial remain blinded, and thus prevents bias arising from differing expectations.

The problem with active ingredients

The main problem with including something active in the placebo, such as another vaccine, is that it can confuse researchers when they measure side-effects. We determine whether an active treatment has a particular side-effect (like an injection-site reaction) by comparing it to a placebo. In the same way that we conclude that an active treatment works if it is better than a placebo, we conclude that it is harmful if it has more side-effects than the placebo. However, if the placebo is designed to cause a side-effect, then the normal way of detecting side-effects doesn’t work. Sticking to our example, the fact that an active vaccine does not have more infection-site reactions than a placebo that was produced to cause infection-site reactions, doesn’t mean that the active treatment doesn’t cause the reactions.

We rarely know how to interpret side-effect information in trials because researchers rarely report what’s in them. Reporting placebo ingredients in general (5) and specifically in vaccine trials, is not common (6). This lack of information makes it difficult to tell what the true harms of the vaccine will be. The same applies to most treatments tested in trials with unknown placebos.

Placebo ingredients influence apparent treatment benefits

What’s in placebos can also affect the apparent benefits of the tested treatment. Olive oil was previously used in ‘placebo’ controls (7) for cholesterol-lowering drugs before it was known that olive oil has cholesterol-lowering properties.

A study of megestrol acetate as an intervention for anorexia associated with cancer (8) showed an unexpected benefit of the drug compared with a lactose placebo in reducing gastrointestinal symptoms. Because lactose intolerance is common (9) in cancer, the lactose placebo’s adverse effects may have contributed to the appearance of benefit (10) from the drug.

Supporting better reporting of placebos

Placebo controls are rightly the gold standard against which new treatments are measured. If a new treatment proves to be better than a placebo, it is taken to be effective. Otherwise, it isn’t. The problem is that there has been no standard for properly reporting placebos in trials. The TIDieR reporting guideline helps researchers fully report their interventions but does not explicitly mention placebo controls. We have now remedied this gap. Our new guideline for placebo control standards has been published in PLOS Medicine (1) and is archived here.

We’ve known about the failure and need to report what’s in placebos for 15 years (10). Using TIDieR-Placebo will make placebo-controlled trials easier to interpret and the benefits and harms of active comparators easier to appraise.

The checklist and guide

Howick J, Webster RK, Rees JL, Turner R, Macdonald H, Price A, Evers AWM, Bishop FL, Collins GS, Bokelmann K, Hopewell S, Lamb S, Madigan C, Napadow V, Papakikitas AN, Hoffmann T. TIDieR-Placebo: A guide and checklist for reporting placebo and sham controls. PLoS Med. 2020;17(9): e1003294. https://doi.org/10.1371/journal.pmed.1003294.

Authors

Jeremy Howick combines philosophy of science, ethics and clinical epidemiology and is internationally known for his research on evidence-based medicine, placebo effects, and empathy. He has published 3 books and almost 100 peer-reviewed articles, and is the director of the Oxford University Empathy Programme. He has appeared on the BBC, ITV, and Channel 4, and his research has been featured in The Guardian, The Times, Men’s Health, the Daily Mail, the Huffington Post, and The Conversation. He speaks regularly at Oxford and internationally.

Claire Madigan is a senior researcher in behavioural medicine at Loughborough University and has worked in the UK and Australia developing and managing clinical trials of investigational medicinal products and behavioural trials.

References

  1. 1. Howick J, Webster RK, Rees JL, et al. TIDieR-Placebo: A guide and checklist for reporting placebo and sham controls. PLoS Med. 2020;17(9): e1003294. DOI: https://doi.org/10.1371/journal.pmed.1003294. PMID: 32956344
  2. 2. Folegatti PM, Ewer KJ, Aley PK, et al. Safety and immunogenicity of the ChAdOx1 nCoV-19 vaccine against SARS-CoV-2: a preliminary report of a phase 1/2, single-blind, randomised controlled trial. Lancet. 2020;396(10249):467-478. DOI: https://doi.org/10.1016/S0140-6736(20)31604-4. PMID: 32702298
  3. 3. Howick J, Moscrop A, Mebius A, et al. Effects of empathic and positive communication in healthcare consultations: a systematic review and meta-analysis. J R Soc Med. 2018;111(7):240-252. DOI: https://doi.org/10.1177/0141076818769477. PMID: 29672201
  4. 4. Howick J (2011) The Philosophy of Evidence‐Based Medicine. DOI: https://doi.org/10.1002/9781444342673
  5. 5. Webster RK, Howick J, Hoffman T, et al. Inadequate description of placebo and sham controls in a systematic review of recent trials. Eur J Clin Invest. 2019;49(11):e13169. DOI: https://doi.org/10.1111/eci.13169. PMID: 31519047
  6. 6. Doshi P, Bourgeois F, Hong K, et al. Adjuvant-containing control arms in pivotal quadrivalent human papillomavirus vaccine trials: restoration of previously unpublished methodology. BMJ Evid Based Med. 2020;bmjebm-2019-111331. Published Online First: 17 March 2020. DOI: https://doi.org/10.1136/bmjebm-2019-111331. PMID: 32184277
  7. 7. Golomb BA. Paradox of placebo effect. Nature. 1995;375(6532):530. DOI: https://doi.org/10.1038/375530a0. PMID: 7791863
  8. 8. Loprinzi, Ellison NM, Schaid DJ et al. Controlled trial of megestrol acetate for the treatment of cancer anorexia and cachexia. J Nat Cancer Inst. 1990;82(13):1127-32. DOI: https://doi.org/10.1093/jnci/82.13.1127. PMID: 2193166
  9. 9. Osterlund P, Ruotsalainen T, Peuhkuri K et al. Lactose intolerance associated with adjuvant 5-fluorouracil-based chemotherapy for colorectal cancer. Clin Gastroenterol Hepatol. 2004;2(8):696-703. DOI: https://doi.org/10.1016/s1542-3565(04)00293-9. PMID: 15290663
  10. 10. Goloumb BA, Erickson LC, Koperski S. What’s in Placebos: Who Knows? Analysis of Randomized, Controlled Trials. Ann Intern Med. 2010;153(8):532-5. DOI: https://doi.org/10.7326/0003-4819-153-8-201010190-00010. PMID: 20956710

 

Blog archives

Latest guest blogger