Recommendations for the development, implementation, and reporting of control interventions in efficacy and mechanistic trials of physical, psychological, and self-management therapies: the CoPPS Statement
BMJ 2023; 381 doi: https://doi.org/10.1136/bmj-2022-072108 (Published 25 May 2023) Cite this as: BMJ 2023;381:e072108- David Hohenschurz-Schmidt, research associate, senior research fellow12,
- Lene Vase, professor3,
- Whitney Scott, lecturer4,
- Marco Annoni, researcher5,
- Oluwafemi K Ajayi, patient partner &, doctoral candidate6,
- Jürgen Barth, professor & director of research7,
- Kim Bennell, distinguished professor and director8,
- Chantal Berna, associate professor & director9,
- Joel Bialosky, clinical associate professor10,
- Felicity Braithwaite, postdoctoral fellow11,
- Nanna B Finnerup, professor and director12,
- Amanda C de C Williams, professor13,
- Elisa Carlino, associate professor14,
- Francesco Cerritelli, chair15,
- Aleksander Chaibi, senior researcher16,
- Dan Cherkin, director of research17,
- Luana Colloca, professor, director of Placebo Beyond Opinions Center, adjunct professor, MPower professor18,
- Pierre Côté, professor, director, Canada research chair in disability prevention and rehabilitation19,
- Beth D Darnall, director & associate professor20,
- Roni Evans, director, associate professor21,
- Laurent Fabre, researcher22,
- Vanda Faria, associate professor23,
- Simon French, professor24,
- Heike Gerger, postdoctoral researcher25,
- Winfried Häuser, adjunct professor26,
- Rana S Hinman, professor27,
- Dien Ho, director and professor of philosophy and healthcare ethics,28,
- Thomas Janssens, postdoctoral research fellow29,
- Karin Jensen, associate professor30,
- Chris Johnston, patient partner31,
- Sigrid Juhl Lunde, postdoctoral researcher3,
- Francis Keefe, professor32,
- Robert D Kerns, professor33,
- Helen Koechlin, postdoctoral research fellow34,
- Alice Kongsted, professor35,
- Lori A Michener, professor36,
- Daniel E Moerman, professor emeritus37,
- Frauke Musial, professor38,
- David Newell, professor39,
- Michael Nicholas, professor40,
- Tonya M Palermo, professor41,
- Sara Palermo, postdoctoral researcher42,
- Kaya J Peerdeman, assistant professor43,
- Esther M Pogatzki-Zahn, professor44,
- Aaron A Puhl, chiropractor45,
- Lisa Roberts, professor46,
- Giacomo Rossettini, researcher47,
- Susan Tomczak Matthiesen, postdoctoral researcher3,
- Martin Underwood, professor48,
- Paul Vaucher, professor49,
- Jan Vollert, postdoctoral researcher50,
- Karolina Wartolowska, clinical research fellow51,
- Katja Weimer, postdoctoral researcher52,
- Christoph Patrick Werner, postdoctoral researcher53,
- Andrew S C Rice, professor and scientific officer1,
- Jerry Draper-Rodi, senior research fellow and director254
- 1Pain Research, Department of Surgery and Cancer, Faculty of Medicine, Imperial College London, London, UK
- 2Research Department, University College of Osteopathy, London, UK
- 3Department of Psychology and Behavioural Sciences, School of Business and Social Sciences, Aarhus University, Denmark
- 4Health Psychology Section, Department of Psychology, Institute of Psychiatry, Psychology and Neuroscience, King’s College London; INPUT Pain Management Unit, Guy’s and St Thomas’ NHS Foundation Trust, London, UK
- 5Italian National Research Council, Interdepartmental Centre for Research Ethics and Integrity, Rome, Italy
- 6Department of Arts and Music, College of Human Sciences, University of South Africa, Pretoria, South Africa
- 7Institute for Complementary and Integrative Medicine, University Hospital Zurich and University of Zurich, Switzerland
- 8Centre for Health, Exercise and Sports Medicine, Department of Physiotherapy, University of Melbourne, VIC, Australia
- 9Centrer for Integrative and Complementary Medicine, Pain Center, Division of Anesthesiology, Sense Institute, Lausanne University Hospital, Lausanne University, Lausanne, Switzerland
- 10Department of Physical Therapy, University of Florida, Gainesville FL, USA; Brooks-PHHP Research Collaboration, Jacksonville, FL, USA
- 11IIMPACT in Health, University of South Australia, Adelaide, SA, Australia
- 12Danish Pain Research Centre, Department of Clinical Medicine, Aarhus University, Aarhus, Denmark
- 13Research Department of Clinical, Educational & Health Psychology, University College London, London, UK
- 14Department of Neuroscience Rita Levi Montalcini, University of Turin, Turin, Italy
- 15Foundation COME Collaboration, University of Chieti-Pescara, Italy
- 16Department for Interdisciplinary Health Sciences, Institute of Health and Society, Faculty of Medicine, University of Oslo, Oslo, Norway
- 17Osher Center for Integrative Health, Department of Family Medicine, University of Washington, Seattle, WA, USA
- 18Department of Pain and Translational Symptom Science, School of Nursing; Department of Anesthesiology, School of Medicine; University of Maryland, Baltimore, MD, USA
- 19Faculty of Health Sciences, Institute for Disability and Rehabilitation Research, Ontario Tech University, Oshawa, ON, Canada
- 20Stanford Pain Relief Innovations Lab; Stanford University School of Medicine, Department of Anesthesiology, Perioperative and Pain Medicine, Stanford, CA, USA
- 21Integrative Health & Wellbeing Research Program; Center for Spirituality and Healing, University of Minnesota, Minneapolis, MN, USA
- 22Centre Européen d’Enseignement Supérieur de l’Ostéopathie, Paris, France
- 23Department of Psychology, Uppsala University, Uppsala, Sweden; Smell & Taste Clinic, Department of Otorhinolaryngology, TU Dresden, Dresden, Germany; Brain and Eye Pain Imaging Lab, Pain and Affective Neuroscience Center, Boston Children’s Hospital, Harvard Medical School, Boston, MA, USA; Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Harvard Medical School, Boston, MA, USA
- 24Department of Chiropractic, Faculty of Medicine, Health and Human Sciences, Macquarie University, NSW, Australia
- 25Erasmus MC, University Medical Centre Rotterdam, Department of General Practice, Rotterdam, the Netherlands; Care and Public Health Research Institute (CAPHRI), Maastricht University, Maastricht, the Netherlands
- 26Department Psychosomatic Medicine and Psychotherapy, Technische Universität München, Munich, Germany
- 27Centre for Health, Exercise & Sports Medicine, University of Melbourne, Melbourne, VIC, Australia
- 28Center for Health Humanities, School of Arts and Sciences, Massachusetts College of Pharmacy and Health Sciences, Boston MA, USA
- 29Health Psychology, KU Leuven; Ebpracticenet, Leuven, Belgium
- 30Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
- 31BC Patient Safety & Quality Council's Patient Voices Network; Health Research BC's Partnership-Ready Network; Health Standards Organization’s Emergency Management Technical Committee & Working Group
- 32Duke University, School of Medicine, Durham, NC, USA
- 33Departments of Psychiatry, Neurology, and Psychology, Yale University, New Haven, CT, USA
- 34Division of Psychosomatics and Psychiatry, University Children’s Hospital Zurich; Division of Child and Adolescent Health Psychology, Department of Psychology, University of Zurich, Zurich, Switzerland; Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA
- 35Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark; Chiropractic Knowledge Hub, Odense, Denmark
- 36Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles CA, USA
- 37College of Arts, Sciences, and Letters, Behavioral Sciences, University of Michigan, Dearborn, MI, USA
- 38National Research Centre in Complementary and Alternative Medicine, Department of Community Medicine, Faculty of Health Science UiT, Arctic University of Norway, Tromsø, Norway
- 39AECC University College, Bournemouth, UK
- 40Pain Management Research Institute, University of Sydney Medical School (Northern) and Kolling Institute of Medical Research at Royal North Shore Hospital, Sydney, Australia
- 41Department of Anesthesiology and Pain Medicine, University of Washington, Seattle, WA, USA
- 42Diagnostic and Technology Department, Neuroradiology Unit, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy; Department of Psychology, University of Turin, Turin, Italy
- 43Unit Health, Medical and Neuropsychology, Leiden University, Leiden, the Netherlands
- 44Department of Anaesthesiology, Intensive Care and Pain Medicine, University Hospital Münster, Münster, Germany
- 45Able Body Health Clinic, Lethbridge, AB, Canada
- 46University of Southampton and University Hospital Southampton NHS Foundation Trust, Southampton, UK
- 47Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Campus of Savona, Savona, Italy; School of Physiotherapy, University of Verona, Verona, Italy
- 48Warwick Clinical Trials Unit; University of Warwick, Coventry, UK; University Hospitals Coventry and Warwickshire, Coventry, UK
- 49School of Health Sciences Fribourg, HES-SO University of Applied Sciences and Arts Western Switzerland, Switzerland
- 50Pain Research, Department of Surgery and Cancer, Faculty of Medicine, Imperial College London, London, UK; Division of Neurological Pain Research and Therapy, Department of Neurology, University Hospital of Schleswig-Holstein, Campus Kiel, Germany; Neurophysiology, Mannheim Centre of Translational Neuroscience, Medical Faculty Mannheim, Heidelberg University, Germany; Department of Anaesthesiology, Intensive Care and Pain Medicine, University Hospital Muenster, Germany
- 51Nuffield Department of Clinical Neurosciences, University of Oxford, John Radcliffe Hospital, Oxford, UK
- 52Department of Psychosomatic Medicine and Psychotherapy, Ulm University Medical Centre, Ulm, Germany
- 53School of Psychology, Faculty of Science, University of Sydney, Australia; Department of Clinical Research, University Hospital Basel, Switzerland
- 54National Council for Osteopathic Research, London, UK
- Correspondence to: D Hohenschurz-Schmidt d.hohenschurz{at}imperial.ac.uk
- Accepted 5 April 2023
Randomised controlled trials (RCTs) that test the efficacy of a drug against a “placebo control” are well established in drug research. In principle, placebo controls mimic the treatment under investigation but lack its supposed active ingredient (for example, as pharmacologically inert sugar pills or saline injections).1 Such specifically designed control interventions control for the natural course of the disease and regression to the mean. Importantly, these interventions also account for context-dependent effects, such as those produced by patient-provider interactions and treatment related rituals. Researchers can use such interventions to blind trial participants, treatment providers, and outcome assessors to trial participants’ group allocation, concealing whether the investigational treatment or the control is administered. Through the indistinguishability of drug and control, all trial participants are exposed to similar treatment contexts, which should lead to comparable expectations of treatment benefit.
In non-pharmacological trials, methodological criticism of control interventions (often called “sham” controls, representing the equivalent of placebo controls) contributes to concerns about research quality.234567 Ultimately, these concerns impede adequate consideration of many therapies for clinical guidelines or reimbursement. While some of these concerns are due to a lack of methodological guidance on the use of control interventions, others stem from a failure to consider the nature of non-pharmacological trials and its impact on control methods. For example, a sometimes unavoidable lack of therapist blinding leads to a downgrading on the PEDro scale,8 a commonly used risk-of-bias scale in physiotherapy research, irrespective of other valid attempts to mitigate related bias risks.
Efficacy trials help to understand intervention effects in experimental or ideal settings. Such trials are important to study intervention mechanisms and causal effects on outcomes. In a pluralistic framework of complex intervention research (such as the 2021 Medical Research Council guidance for complex intervention development and evaluation), efficacy trials complement research designs that are more implementation-focused.9 Although control interventions are a central feature of efficacy trials, existing guidance for control intervention design focuses on individual therapies, such as psychotherapy,10 behavioural interventions,11 rehabilitation,12 sports and exercise,13 physiotherapy,14 and manual therapy.15 These guidelines provide no quality checklist and few generalisable principles; moreover, they often disregard problems of intervention complexity.9 Consequently, various specialties follow different approaches to fundamental questions of control intervention design, such as how closely the control should resemble the study intervention. Although there is a trend towards high similarity controls in some research areas (notably spinal manipulation), in other areas (such as other physical therapies and psychological therapies) control interventions often do not resemble the study treatments or efficacy trials are avoided altogether.1617 A lack of consensus on relevant issues exposes the field to justifiable criticism due to concerns over bias1819202122 and leaves questions of treatment mechanism unanswered. Finally, the only reporting checklist for control interventions is the TIDieR-Placebo (template for intervention description and replication for placebo and sham controls), developed for both drug and non-drug studies. However, it may not be sufficiently comprehensive to reflect the challenges of control interventions in all types of physical, psychological, and self-management (PPS) intervention efficacy trials.
To fill this gap in guidance, we present the CoPPS Statement (recommendations for the development, implementation, and reporting of control interventions in efficacy and mechanistic trials of physical, psychological, and self-management therapies). This guidance is dedicated to PPS interventions, which present unique challenges for blinding and control interventions. The PPS term includes all forms of manual and physical therapy; exercise and rehabilitation therapy; conversation based and psychological therapies; mind-body, spiritual, religious, and other non-material healing practices; and educational interventions. We will not consider surgical, needle based, or meridian interventions, devices, drugs, or nutritional interventions in this discussion, as there are alternative options for creating “sham” controls (such as treatment under anaesthesia, use of non-acupuncture points, or device deactivation).
Importantly, this guidance is intended for researchers who have decided that a controlled efficacy or controlled mechanistic RCT is appropriate for their research question. Viewing clinical research as a spectrum from explanatory to pragmatic,2324 we suggest the control interventions discussed here are more useful for reducing bias when testing the efficacy of a given intervention or for studying mechanisms of action rather than estimating an intervention’s effectiveness under real world conditions. Thus, these guidelines are more appropriate for trials on the explanatory end of the explanatory-pragmatic continuum. However, each clinical trial is different. Specific solutions will be informed as much by the present guidance as by the uniqueness of the treatment and population under investigation, the research question, and practical considerations. The adoption of all recommendations may not always be feasible or desirable. We encourage researchers to consider each recommendation carefully, assess its relevance and feasibility for their present research project, and justify their decision in the trial protocol, report, or supplement. We also highlight scenarios in which individual recommendations may be particularly important (also see supplementary explanations and elaborations (E&E) document).
Summary points
The design, conduct, and reporting of control interventions are fundamental for interpreting efficacy and mechanistic clinical trials of physical, psychological, and self-management interventions
We developed this guideline using a three-round Delphi study with 64 experts in placebo research and/or clinical trials. This panel was presented with a systematic review of control and blinding methods and a meta-analysis that linked design features to trial endpoints. Consensus items informed interviews with eight patient partners. Finally, 44 Delphi participants and two patient representatives discussed the draft manuscript and results from stakeholder interviews at online consensus meetings
For clinical trials of treatment efficacy or mechanisms, we recommend designing control interventions that are as similar as possible to the tested intervention, apart from the components examined by the study
Structured planning, early stakeholder engagement, feasibility work, and piloting will improve the quality and acceptability of control interventions
When participant blinding is an objective, blinding effectiveness should be routinely assessed and reported
Detailed and transparent reporting will improve the interpretation and repeatability of clinical trials
Methods
We have based our statement on a systematic review of methods,1725 a three-round Delphi study,26 interviews with patient partners, and consensus meetings.27 Methods were adapted from available guidance on best practice for guideline development27 and relevant, related publications,28 notably adding patient involvement and the consultation of placebo research experts to the methodology. A detailed protocol of the consensus process was prospectively registered on the Open Science Framework (https://osf.io/jmyhq/), where a detailed documentation of this project’s methods and results is also available (supplements 1a, 1b, and 2). The study was approved by the Imperial College Institutional Review Board (study No 21IC6668). The scope of this guideline is PPS interventions, excluding surgery, acupuncture, and devices.
The systematic review identified current methodological and reporting practices in relevant trials,17 and our meta-analysis showed that control intervention design influences trial outcomes.25 The first Delphi questionnaire was informed by these insights as well as earlier relevant literature on control design and blinding (such as references 101214293031). A total of 68 experts in placebo research and clinical trials of PPS interventions received the round-one Delphi questionnaire, of whom 48 completed round three (71% retention). During the Delphi study, experts indicated their level of agreement to potential recommendation items for this guideline, provided additional considerations in response to open ended questions, and received feedback about other panellists’ ratings between each round. After the Delphi stage, eight laypeople with present or past experiences of long term pain were interviewed about the proposed recommendations. We recruited these individuals through patient advocacy networks. Draft guidance was then discussed at a series of online meetings with the same Delphi panellists (n=44) and two of the same laypeople living with persistent pain who had volunteered to participate. Apart from being subject experts, several individuals from the panel and author group had experience in guideline development.
The guideline development process is illustrated in figure 1 and described in detail in supplement 1a. The preparatory systematic literature review and meta-analysis have been published.1725
The guidance presented in this article includes items that reached consensus at the Delphi stage (for detailed results per item, see supplement 1b). The CoPPS Checklist provides a summary of essential recommendations, representing quality standards that are required of any control intervention in a PPS trial (table 1). In addition to reaching consensus in the Delphi stage, these items were seen as applicable and essential to all controlled efficacy and mechanistic trials of PPS interventions and were selected during online consensus meetings and manuscript writing. During this project, most discussions and Delphi items concerned the development and implementation of control interventions. However, recommendations for reporting items were also made, which were added to an existing reporting checklist32 to ensure that readers have easy access to all relevant guidance for control interventions in PPS trials.
For readers desiring further detail or wishing to apply the CoPPS Statement in their trial, an E&E document provides additional considerations and practical examples to guide researchers in the decision making process. The E&E document, an editable version of the CoPPS Checklist of essential recommendations, and an editable version of a dedicated reporting checklist are available as “toolbox” supplements, helping researchers to apply the main statement’s recommendations.
Guidance statement
As part of the CoPPS Statement, we specify important terminology, provide general considerations for control interventions in efficacy trials, present fundamental principles for the conceptual development of control interventions, and discuss key aspects of their piloting and their implementation in clinical trials, their evaluation, reporting, and the interpretation of trial results. A checklist of core recommendations for the development and implementation of control interventions in efficacy and mechanistic trials of PPS interventions is presented in table 1 (an editable checklist is available as supplement, which we recommend is used by trialists). The guidance statement is displayed visually in the supplementary infographic.
1. Terminology and communication
In non-pharmacological trials, the terms “sham” or “attention control” are commonly used instead of “placebo.” None of these terms is ideal. “Sham” may be associated with deceit and has been thought to undermine trust in the research by consulted patient representatives. “Attention control” is too restrictive, applying only to one component of an experimental treatment, the “attention” from the healthcare system. “Placebo” has a negative connotation, does not acknowledge the potential for direct benefits of control treatments, and has various interpretations among the public. Thus, we encourage the use of simplified terminology and descriptive language, both when reporting research methods within the scientific community and in communication with (potential) trial participants and providers (see box 1). For example:
Glossary of relevant terms
The following definitions have been agreed as part of a consensus-finding process and form the basis of wording in the guidance document. These terms represent a deliberate simplification of language and omit commonly used terms that are considered unhelpful (notably “sham” and “placebo”). We encourage researchers to adopt similar language in their communications with colleagues and members of the public. This guidance focuses on the scope of therapies given by the definition of PPS therapies, as further justified in the text.
Control intervention (in the context of efficacy or mechanistic trials)
Definition: Procedures delivered to trial participants in the control group, specifically designed to test the efficacy or mechanism of the test intervention
These control interventions are distinct from usual care, other recognised treatment, and no-treatment controls. Although no-treatment approaches are also commonly called “controls,” they do not control well for expectancy effects, do not allow for blinding, and are neither specifically designed for a given trial nor common in efficacy or mechanistic trials. Such comparator arms would be better termed a “comparator intervention.”
Note that we are specifically avoiding the terms “sham,” “placebo,” and “attention” control interventions for reasons specified in the text.
Test intervention (or tested intervention/treatment)
Definition: The intervention or therapy investigated by the trial
Synonyms: “Index” or “experimental intervention”
Sometimes the term “active intervention” is used, although the distinction between active and inactive is often not clear-cut in non-pharmacological research, which is why we recommend omitting this term.
Components of interest
Definition: The components of the test intervention that the researchers expect to be responsible for the efficacy of the test intervention under study.
The components of interest are closely aligned to the research question.
These components are sometimes called “specific” (such as Wampold33) or “characteristic” (such as Howick34) components/factors/ingredients. However, these concepts are highly dependent on the treated condition and treatment theory.34 Therefore, we advocate for simplified language in the context of RCTs, describing what treatment components were studied and why. This aspect is explained in further detail in Section 1.
Other components not of interest in the study
Definition: Components of the test intervention that the researchers do not intend to study
These components may be thought to contribute to the placebo effect and/or to other effects that are not of interest in the study.
Placebo and nocebo effect
“The placebo and nocebo effect [are changes in health outcomes that are] specifically attributable to placebo and nocebo mechanisms, [such as] the neurobiological and psychological mechanisms of expectancies [and learning]. These mechanisms are shaped, for example, by verbal instruction, or nonverbal or situational cues that affect treatment expectancies.”35
For details on placebo and nocebo mechanisms, see, for example, references 363738
Placebo and nocebo response
Definition: Changes in health outcomes in the control arm of an efficacy trial
These responses arise from the placebo or nocebo effect as well as other independent phenomena contributing to changes in outcomes in the control arm, such as regression to the mean and natural disease course.35
(Clinical trial) Participants
Definition: People entering a clinical trial for the purpose of receiving or participating in the trial’s interventions.
Treatment provider
Definition: A person providing interventions as part of a trial (can apply to both test and control interventions)
In the case of self-management therapies, “provider” refers to the individual that introduced the trial participant to the self-management therapy. In the case of self-directed therapies using technology, a provider may refer the individual for trial participation.
Researcher
Definition: The individual or group designing, conducting, analysing, and reporting a trial.
Synonym: “Investigator”
Physical, psychological, and self-management (PPS) therapies
Broadly, we are discussing non-surgical, non-pharmacological interventions, including all forms of manual and physical therapy; exercise and rehabilitation therapy; conversation based and psychological therapies; mind-body, spiritual, religious, and other non-material healing practices; and educational interventions.
We exclude any interventions that require the skin to be pierced surgically or with needles or the ingestion or introduction of substances (drugs, supplements, nutritional interventions). Because other opportunities and challenges for sham controls exist, we also exclude therapies based on meridian and Qi concepts (including traditional acupuncture and acupressure), as well as any interventions in which therapists rely on the use of devices (such as ultrasound, transcutaneous electrical nerve stimulation, laser, transcranial stimulation, shockwave, spinal stimulation, splinting and braces, but not exercises using, for example, elastic bands). For the excluded therapies, we refer the reader to existing guidance.283940
“The control is the same as the tested treatment, except that one component has been removed. In this trial, we test the effects of this component.”
“The tested treatment consists of multiple components. The trial’s aim is to study the effect of some of these components. To do so, the test treatment is compared with a control that has all of the original components except those components that the trial aims to study.”
After explaining the concept, one may add that this control intervention is sometimes referred to as “sham” or “placebo control” if this is formally required. Importantly, this approach will enable enhanced communication in all languages and can be adapted to different audiences, ideally guided by stakeholder involvement.
2. General considerations for the design of control interventions
2.1. Objectives of control interventions and the similarity principle of control design
Current placebo research shows that expectation and learning effects can change clinical outcomes.41 Because control interventions aim to control for the placebo effect (apart from other confounders such as symptom regression towards the mean, spontaneous disease remission, etc.), the principal objective of control interventions is to balance the expectations of trial participants. Blinding can achieve this goal, assuming that expectations are more likely to be balanced if trial participants do not know which group they are in. Similarly, most treatment components can influence trial outcomes, either directly or through expectancy or conditioning.254243444546 Because these confounding effects are difficult to predict, replicating most components of the test treatment is another objective of the control intervention. These objectives translate into a design principle for control interventions in efficacy or mechanistic trials of PPS interventions:
Control interventions should replicate as many components of the investigated treatment as possible, apart from the components whose effect the trial aims to study.
In addition to balancing expectancy effects between groups, implementing this principle of similarity will reduce the risk of differential attrition of trial participants25 and may promote acceptability to providers, thus reducing bias in multiple ways.
We recommend that test and control interventions be identical with regard to the features presented in table 2 unless the trial aims to study a particular component of the treatment.
Side effects of a treatment can undermine blinding.47 While we do recommend that the treatment and control “feel” similar (table 2), the replication of side effects and discomfort in a control intervention may be challenging for practical and ethical reasons. Further, the nocebo effect (that is, negative health outcomes through expectancy and learning mechanisms) accounts for some adverse experiences,4849 although more research is required for this field. Complying with the above recommendations would ensure that nocebo effects are balanced between the two groups.
2.2. Further considerations for trial and control intervention design
Additional comparators, such as waiting list or no-treatment groups, may elucidate the magnitude of the placebo effect. Traditional efficacy trials do not usually implement these comparators, although their inclusion can provide insights about the potential real world effectiveness of an intervention50 or help contextualise observed effect sizes between test and control interventions. The latter may be particularly pertinent in the field of PPS: in contrast to pill based placebos in drug trials, most control interventions in PPS trials will not be fully “inert.” Moreover, as discussed below, researchers may not wish or be able to omit all supposedly active treatment components in the control intervention. Thus, it is unclear whether effect sizes that are comparable to drug trials can be expected. This, however, does not negate the need for well informed power calculations (for example, informed by pilot testing or studies with comparable control interventions and considering the clinical meaningfulness of effects). Three-armed designs may further provide an opportunity for the assessment of multiple treatment mechanisms (including different delivery modes or doses). However, decisions for such designs depend on feasibility and the research questions under investigation.
Similarly, a trial’s hypothesis will dictate the choice of outcome measures. We conclude that neither patient-reported nor more “objective” measures are more desirable in the general context of control interventions. This decision depends primarily on the trial’s objectives. Furthermore, the evidence regarding their differential susceptibility to placebo effects is inconclusive.51 If available and appropriate, both patient-reported and more objective outcome measures can be used.5253
2.3. Ethical considerations
One must consider the ethics of trials with a specifically designed control intervention in each case. Such trials are generally considered ethical when no proven treatment exists and when high quality evidence of efficacy is lacking for the tested intervention. This includes situations in which the studied therapy is already commonly used in clinical practice 5455 or when an established treatment is available as alternative comparator. Researchers ought to consider ethical concerns from the perspective of trial participants (see section 3.5.).
3. Control intervention development and testing
3.1. Conceptual development
In designing a new control intervention, one should clearly define the objectives of the control intervention (see section 2.1.).
The development of a control intervention begins as a conceptual process. First, researchers should define the physiological, cognitive, and/or behavioural mechanisms through which the tested intervention is hypothesised or known to exert its effects (also see the “programme theory” section in the 2021 Medical Research Council guidance for complex interventions9). Then, they should describe which components (table 2) of the test treatment are expected to produce clinical effects via these mechanisms. At the same time, researchers must consider all components of the test treatment and its context with regard to their potential to elicit placebo or nocebo effects. We recommend a literature review of comparable control interventions and available blinding methods to guide intervention development. Finally, the control intervention should be designed to replicate all components while omitting the mechanisms of the experimental treatment. As a “safety check,” researchers should consider mechanisms by which the control intervention could produce unanticipated therapeutic benefits and whether these overlap with hypothesised test treatment mechanisms. In other words, the control intervention ought to be as inert as possible for the studied mechanisms.
3.2. Practical development and validation
The development of any control intervention should involve specific feasibility testing or a validation phase, either externally or as part of the trial. Previous validation of a control intervention should not replace the feasibility pre-testing of specific trial procedures or the evaluation of blinding effectiveness in each trial. When a previously validated control is adapted to a new trial intervention, repeated validation testing is required.
In validating a developed control intervention, researchers should ensure that the following quality criteria are met:
The control intervention must be credible as a treatment (believability of the control as an intervention that will provide benefit)5657
If blinding is an objective, the control should successfully blind participants to group allocation (similar proportions of participants in both groups believe they have received the test or control treatment or do not know)58
The control intervention should elicit a similar expectation of benefit as the test intervention.5659
To reduce the risk of attrition and unblinding, researchers should consider how engaging and acceptable the control intervention is for trial participants and providers. Ideally, matching the components in table 2 will produce similarly engaging interventions. However, important intervention components may be removed, and one must consider whether this removal will make the control less engaging. Early consultation of potential trial participants and providers may be helpful to enhance control intervention acceptability (see section 3.5.).
Researchers should also consider the risk for group contamination, including information exchange between participants in different trial groups and the associated risk of unblinding. Group contamination may occur, for example, when study participants of both groups attend the same clinic, see the same group of providers, or are recruited from the same peer group. If identified, these risks of group contamination should be mitigated.
Explanations to trial participants regarding the interventions should be matched for plausibility, development of expectancies, and the need for adherence. Furthermore, written and verbal information should be presented in a manner consistent with authentic delivery methods. This matching should also be applied when obtaining informed consent within the (control) treatment session after explanations about procedures and possible risks and benefits.
3.3. Provider training: protocol fidelity, blinding, and equipoise
Participant-provider interactions should be standardised to avoid uncontrolled wording, behaviour, treatment provision, and conditioning. We encourage specific provider training that clearly delineates what is in and out of scope for control intervention delivery, so that no off-limit advice or education is provided in the control arm. Providers should receive a clear structure or protocol with operational definitions that guide control intervention encounters, supported by training. The use of suggested language may be useful when conversations cannot be scripted. The use of video recordings to review practice consultations before a trial and fidelity monitoring of conversations and interactions during the trial may also be useful. During trial preparations, staff (therapists and others) should receive training on the importance of maintaining blinding and the need to reduce unblinding risks for any party.
Provider blinding is another challenge in PPS trials. Provider expectations can influence trial results through many mechanisms, including verbal and non-verbal interactions with trial participants and deviations from trial procedures.6061 As such, differential provider expectations may pose a threat to a trial’s internal validity,62 although this aspect is rarely discussed in trial reports.17 Because it is usually impossible to conceal from providers which intervention they deliver, one must instead consider provider equipoise and allegiance. Here, equipoise means a comparable belief in the usefulness of intervention and control, and allegiance refers to the professional or personal commitment to a therapeutic modality.63 While completely balancing these factors may be unrealistic, trial designers should consider how to promote confidence in (control) treatment delivery across providers (box 2). Further, researchers should critically reflect on their own allegiance to and belief in the study interventions.
Possible steps for trial designers to enhance provider confidence in (control) intervention delivery and compliance with trial procedures
These considerations can be applied to any non-blinded staff with trial participant contact.
During the development of control interventions
Consider involving providers in the development of a control intervention
Test the control intervention with providers, assessing for acceptability and provider expectancy
As a part of provider training before a trial
Discuss the concept of equipoise and the need to adhere to trial procedures
Educate providers about the complex nature of the tested intervention, the replication of most of its components in the control arm, and the potential benefit to participants through the placebo effect
Discuss similarity between conditions and the many remaining components of the control (many of which they will recognise as important parts of their everyday care provision)
If required, remind providers of the ethical justifications of control interventions in clinical research and the purpose of RCTs
During a trial
When possible, standardise patient-provider interactions (for example, using a script detailing all procedures and communications)
Evaluate provider expectancy regularly
Monitor fidelity to intervention protocols and conversation scripts
3.4. Blinding of other parties
The blinding of outcome assessors is essential, and blinding of other involved staff should be done along with adequate allocation concealment.64 To avoid performance and confirmation bias, treatment providers must not act as outcome assessors if they are not blinded. Statistical analyses must be blinded.65 Since this can be difficult, pre-registration of a detailed statistical analysis plan is another important method for minimising bias during data analysis.6667
3.5. Patient involvement and patient communication
Discussing the draft guidelines with people living with pain highlighted several potential enablers and barriers for trial participation and for the successful conduct of an efficacy trial of PPS interventions (see table in E&E supplement; for more details, see supplement 2). Consequently, the expected benefit of involving potential participants in the development of a control intervention and planning of a trial is large. Enablers and barriers can be further explored through stakeholder involvement, which must be geared towards the target clinical population and therapeutic modality. Such stakeholder involvement can also be used to optimise communication with trial participants before, during, and after a trial. Examples include wording around the control intervention and education about its purpose.
4. Conducting a controlled trial
4.1. Fidelity monitoring and participant adherence
As described in box 2, researchers should monitor the fidelity of providers to intervention protocols and/or conversation scripts during a trial, and participants’ adherence to and compliance with interventions. These steps are particularly important for trials with prolonged treatment periods, unsupervised self-management components, or complex intervention procedures.
4.2. Measuring participant expectation and blinding effectiveness
Due to the potential implications of expectations for trial outcomes, researchers should evaluate participant expectations of benefit at baseline and after participants have started interventions.48686970 Nonetheless, the influence of expectancies on trial outcomes requires further study, and better methods are needed to adequately assess these influences.71
Although related to participant expectation, blinding effectiveness should be assessed separately because it is a more tangible concept and an important objective of covert control interventions. To assess blinding effectiveness, researchers can ask participants to guess their group allocation. The high importance of successful blinding for most controlled trials requires blinding to be assessed at least once in any trial.
4.3. Attrition
Throughout the trial, researchers should document reasons for participants’ withdrawal from the study (such as adverse events, lack of benefit, symptom improvement, unblinding, etc),7273 while recognising that participants have no duty to disclose reasons or may not disclose the true cause. However, researchers should aim to determine whether differential attrition is linked to unblinding or other aspects of the control intervention, such as a lack of credibility or acceptability that could push participants to seek care elsewhere.
5. Reporting a controlled trial
General reporting guidelines for non-pharmacological clinical trials are available, including Consolidated Standards Of Reporting Trials (CONSORT) extensions.3074 For reporting “placebo” or “sham” controls, a specific guide and reporting checklist exist (TIDieR-Placebo).32 TIDieR-Placebo provides a good basis for improved reporting of control interventions, but not for design of control interventions, nor is TIDieR-Placebo specific to non-pharmacological interventions. We recommend compliance with TIDieR-Placebo, but additional detail is required, particularly for how test and control interventions differ with regards to the features presented in table 2.
Based on our consensus process, we recommend reporting of all items from TIDieR-Placebo 32 and seven additional items. These reporting items are explained in the supplementary E&E document. We provide a TIDieR-Placebo/CoPPS hybrid reporting checklist to specifically improve reporting of control interventions in PPS trials, integrating our recommendations into TIDieR-Placebo (table 3; also provided as editable checklist in the online “toolbox” supplement). Notably, this checklist is for the reporting of control interventions only and not to be confused with our CoPPS Checklist of essential items for control intervention development and conduct (see table 1 and supplement). We recommend that both checklists are submitted and published alongside relevant trial manuscripts.
6. Interpreting efficacy and mechanistic RCTs of PPS interventions
In complex interventions, treatment components may not always interact in an additive manner, but may interact in unpredictable ways.975 Complexity also arises from interactions with the context in which complex interventions are implemented.9 For example, a good therapeutic relationship may reinforce the effects of a particular treatment component, such as by increasing treatment adherence and motivation, but the extent of such an effect is difficult to predict76 and may vary with context. In the control intervention, the composition of the components is altered, which may lead to different interactions. Mediation and moderation analyses may facilitate the interpretation of these effects on trial participants,777879 as well as existing mechanistic studies80 and balanced placebo designs.81 For cases in which multiple components are removed from the control, the mechanistic interpretation becomes even more complex. Importantly, when control interventions are designed according to the principles presented here, studies will reflect the efficacy of the tested component, and only this component, and the involved treatment mechanisms. In contrast to drugs, complex interventions rarely act on a single mechanism. Thus, results usually cannot directly answer questions of real world effectiveness or reflect (lack of) efficacy of the intervention as a whole.82
Similarly, choices made in the design of the control intervention determine the mechanism studied. There is always the possibility that the supposed mechanism of effect was wrongly conceptualised. Thus, conceptual clarity about and transparent reporting of control interventions are paramount to facilitate interpretation and evidence synthesis.
Finally, effect sizes have to be interpreted carefully in efficacy trials with high-similarity control interventions.83 Although such control interventions are important for obtaining a mechanistic understanding of an intervention and for bias control, trials that control for all but one component of a complex intervention cannot be expected to show effect sizes comparable to those of drug trials, where a single pharmacological ingredient may primarily account for the benefit over placebo. However, more research on the directionality, extent, and variability of effect estimates in non-pharmacological efficacy trials is required, also accounting for the unclear influence of blinding.84858687 Effect sizes of efficacy trials may also not reflect the effects that can be obtained in real world practice or the benefits over structurally different “usual care.” However, positive signs from an efficacy trial with a well designed control intervention should increase end users’ confidence in an intervention under real world conditions, even if effect sizes in the efficacy trial are small. Hence, well designed pragmatic trials or additional comparator arms are useful for further elucidating questions of comparative and real world effectiveness.
Discussion
Limitations
We have presented a consensus statement for the development, implementation, and reporting of control interventions in efficacy and mechanistic trials of PPS interventions. A limitation of this statement is that, despite targeted efforts to recruit individuals from low- or medium-income countries, the author group included no experts from these areas (supplement 1a). Additionally, few individuals were recruited from scientific communities in which English is not the dominant language. This may reduce the generalisability of our recommendations to other cultures, healthcare systems, or languages. Second, the consensus process originally focused on pain as the primary outcome and primarily involved individuals with expertise in pain research. However, due to our focus on best-practice principles and interventions rather than conditions, it became clear during the process that this guideline is broadly applicable, especially to trials exploring complex, persistent, or recurring health problems and symptoms that are amenable to expectancy effects. Examples include mental health, but, as described above, the CoPPS Statement must be considered in the context of the given patient population and research question.
Concluding remarks
The successful development and implementation of a control intervention adds considerable expense to a clinical trial and can be burdensome to trial stakeholders. Nonetheless, excessive trade-offs between control design and research burden can undermine the interpretability of trial results. Without trust in the quality and success of the control, the conclusions that can be drawn from a trial are weakened. To support end users, we call for researchers and journal editors to improve reporting practices, including full, transparent reporting of control interventions and blinding effectiveness.
In the management of individuals with chronic conditions, guidelines call for changes in psychosocial and lifestyle factors; thus, research into non-pharmacological, non-invasive alternative treatments abounds. Created for a field with particular challenges for blinding and control intervention design, this guidance provides a robust framework for high-quality efficacy and mechanistic trials of PPS interventions. The guidance acknowledges that this field is distinct from drug research, requiring a contextualised interpretation of trial results and proposing specific solutions for trial designers. Thanks to a rigorous consensus process with experts in placebo research and patient partners, trial designers now have access to unifying principles of best-practice control intervention development and implementation during a trial, amenable to specific research scenarios and potentially generalisable to other non-pharmacological research fields.
Data availability statement
Data from the Delphi study and patient interviews are available in supplements. Anonymised transcripts of consensus meetings and patient interviews are available from the corresponding author.
Acknowledgments
We thank the participants of the patient centred research component, who crucially influenced this work. We also thank Oliver Thomson, Donna Kennedy, and Saskia Jünger, who provided feedback on the methodology of individual aspects of this research; Jules Phalip, who helped with the Delphi analysis; and the Science Editors Network for text editing support at the revision stage (http://www.scienceeditorsnetwork.com/). We acknowledge the contributions of Delphi panellists who did not complete the process.
Footnotes
Contributors: DHS, LV, JV, WS, JDR, and ASCR made substantial contributions to the conception and design of the work and the analysis and interpretation of data. DHS and JDR collected data as part of the Delphi process, interviews with patient partners, and facilitation of consensus workshops. All authors except DHS, LV, JV, WS, JDR, and ASCR participated in at least two Delphi rounds, and all except WS, EC, DH, JV, and MA actively participated in at least one consensus workshop. DHS drafted the manuscript, and all other authors revised it and/or re-drafted sections through three rounds of internal review and as part of consensus workshop discussions. JV had a supervisory role and guided the revision of the manuscript during the publication process. DHS is the guarantor of this work. All authors approved the final manuscript version and agreed to be accountable for all aspects of the work. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.
Funding and sponsors: This work was supported by a PhD studentship from the Alan and Sheila Diamond Charitable Trust awarded to DHS. Neither this funder nor the funders of any of the authors’ positions had any role in the design, conduct, analysis, or dissemination of this study.
Competing interests: All authors have completed the ICMJE uniform disclosure form at https://www.icmje.org/disclosure-of-interest/ and declare:
Support for the submitted work was received by the following authors: DHS (Alan and Sheila Diamond Charitable Trust, PhD stipend), BD (National Institute on Drug Abuse DA053564), JDR (Alan and Sheila Diamond Charitable Trust, payment to affiliated institution to develop a PhD studentship), RSH (National Health and Medical Research Council [NHMRC] Senior Research Fellowship), ASCR (Alan and Sheila Diamond Trust). The following authors declare no support from any organisation for the submitted work: OA, MA, JBa, KB, CB, JBi, FB, EC, FC, AC, DC, LC, PC, RE, LF, VF, NF, SF, HG, DH, WH, CJ, TJ, KJ, RDK, HK, AK, FK, SJL, STM, LAM, DEM, FM, MN, DN, SP, TP, KJP, EMP-Z, AAP, LR, GR, WS, MU, LV, PV, JV, KWa, KWe, CPW, ACDCW.
The following authors declare financial relationships with any organisations that might have an interest in the submitted work in the previous three years: DHS (research grant from The Osteopathic Foundation; royalties from the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials [IMMPACT]; consulting fees from Altern Health Ltd; travel fellowship from Society for Back Pain Research), MA (consulting fees from GLG - Deep Mind, Human Technopole Foundation), FB (John Stuart Colville Fellowship from Arthritis Foundation of South Australia), PC (grants from Canadian Institutes of Health Research, Canadian Chiropractic Research Foundation, Social Sciences and Humanities Research Council of Canada, College of Chiropractors of British Columbia, Canadian Chiropractic Association; payments for expert testimony from Canadian Chiropractic Protective Association and NCMIC), BD (Patient-Centered Outcomes Research Institute [PCORI] #PCS_2021C1_22347, PCORI #1610-3700; stock options in AppliedVR), RE (grants and/or contracts from National Institutes of Health [NIH] and US. Department of Defence), VF (support from Swedish Research Council 437-2014-6767; support to attend meetings and/or travel), NF (support from Innovative Medicines Initiative (IMI) 2 Pain Care - an EU IMI 2 public-private consortium that involves the following companies: Grünenthal, Bayer, Eli Lilly, Esteve, and Teva; consulting fees from Merck, Almirall, NeuroPN, Vertex, Novartis Pharma, Biogen, Merz, and Confo Therapeutics), RSH (NHMRC, Australian Research Council, Medibank), HK (Swiss National Science Foundation, Postdoc.Mobility Fellowship [P400PS_186658/1] and Postdoc.Mobility Return Grant [P5R5PS_203064/1]; project specific research support to affiliated institution by Stiftung für psychosomatische, ganzheitliche Medizin; invited crosstalk on “Chronic pain in children and adolescents: chances and hurdles in clinical practice” at the University of Basel, Faculty of Psychology; invited teaching of a module on “Communication with children and their families in the context of chronic pain,” Bern University of Applied Sciences; travel grant for the 2nd official Society in Placebo Studies conference in Leiden, NL, from the Travel Fund of the University of Basel), LAM (grants or contracts from PAC-12 Student-Athlete Health and Well Being Grant, NIH National Institute of Arthritis and Musculoskeletal and Skin Diseases [NIAMS], American Physical Therapy Association, The Charles D. and Mary A. Bauer Foundation), MN (Department of Health [Australian Government] Grant GO2810 9: Health Professional Pain Management Education and Training Project [2020-2024]; Harper Collins Books royalties for joint authorship of “Manage Your Pain”), FK (support as a co-principal investigator on NIH grant 1U24-NS114416-01 and a co-investigator on a number of NIH grants [1R21-DA052729-01A1, 315-55811085033391-10000116454, 17-112, 5R01-CA193673-05, 5R01-CA202779-05, 5P30-AG064201-02, 2004832463/5UH3-AR077360-03, 239631/1-R01AG064947-01A1, 239631/1-R01AG064947-01A1, 1U24-NS114416-01 (PI), 5R01-CA237892-03, 5R01-CA229425-03, 5R01-NR017801-04, 5UH3-AT009790-05, 5R01-CA249959-03, 577985/U01-DK123813-03, 315-5581108503339/1UG3-NR019196-01, 5113091/5R01-AG058702-04, 359-32491-11000000084/3UG1-CA189824-08S1] and a grant from the American Cancer Society; royalties for a book he co-authored on the self-management of pain; consulting to the Health Innovations Network in London regarding strategies for training physical therapists in teaching coping skills to their patients; honorarium for giving an online lecture on pain to the IMMPACT XXV meeting), TP (NIH research grants; royalties for two books on cognitive-behaviour therapy published with Oxford University Press; consulting fees from Mahana Therapeutics, a digital therapeutics company that made payments for consulting on their cognitive-behaviour therapy product), KJP (Early Career Research Grant from the International Association for the Study of Pain [IASP]; Leiden University Fund research grant), EMP-Z (grants or contracts from the German Research Foundation, the German Federal Ministry of Education and Research, the German Federal Joint Committee, the IMI 2 Joint Undertaking under grant agreement 777500 [the Joint Undertaking receives support from the EU’s Horizon 2020 research and innovation programme and the European Federation of Pharmaceutical Industries and Associations]; payments and honorariums from Grünenthal, Novartis, Medtronic; all money went to affiliated institutions [WWU/UKM]), WS (grants from Medical Research Council and Versus Arthritis MR/W002388/1, National Institute for Health and Care Research PDF2015-08-059), MU has been chief investigator or co-investigator on multiple research projects funded by the NIHR under financial relationships with organisations that might have an interest in the submitted work, PV (grants or contracts from University of Applied Sciences and Arts Western Switzerland, Swiss Osteopathic Science Foundation, Swiss National Fund, CARA; consulting fees from SuisseOsteo for advice to the professional association ethical board; payments and/or honorariums from University of Applied Sciences and Arts Western Switzerland, CIDO, Kookies, SuisseOsteo, SCNJ), CPW (grants from Australian Commonwealth International Research and Training Program; payments or honorariums from The University of Sydney; support for attending meetings and/or travel: World Sleep Conference 2019, Young Investigator Award), JV (consulting fees from Vertex Pharmaceuticals, Embody Orthopaedic, Casquar), ASCR (grants and studentships from UK Research and Innovation [Medical Research Council and Biotechnology and Biological Sciences Research Council], Versus Arthritis, Alan and Sheila Diamond Trust, Royal British Legion, European Commission, Dr Jennie Gwynn Bequests, Royal Society of Medicine; remunerated consultancy work for Imperial College Consultants, including for work for Confo, Pharmnovo, Lateral, Mundipharma, Vertex, Novartis, Orion, Shanghai SIMR Biotech; speaker honorarium for MD Anderson Cancer Center; patents [none being commercialised]: Rice ASC, Vandevoorde S, and Lambert DM. Methods using N-(2-propenyl)hexadecanamide and related amides to relieve pain. WO 2005/079771 and Okuse K et al. Methods of treating pain by inhibition of vgf activity. EP13702262.0/ WO2013 110945; owner of share options in Spinifex Pharmaceuticals from which personal benefit accrued upon the acquisition of Spinifex by Novartis in July 2015 [final payment was made in 2019]). The following authors declare no financial relationships with any organisations that might have an interest in the submitted work in the previous three years: OA, JBa, KB, CB, JBi, EC, FC, AC, DC, LC, JDR, LF, SF, HG, DH, WH, CJ, TJ, KJ, RDK, AK, SJL, STM, DEM, FM, DN, SP, AAP, LR, GR, LV, KWa, KWe, ACDCW.
The following authors declare other relationships or activities that could appear to have influenced the submitted work: OA (unpaid leadership or fiduciary roles for Gail Sickle Initiative, Pan Africa Sickle Cell Federation, Rare Disease South Africa, and Global Alliance of Partners for Pain Advocacy Task Force), CB (Swiss Pain Society, board member; European Pain Federation [EFIC], councillor; Swiss Society for Psychosomatic Medicine, scientific advisory board), FB (speaker fees for talks related to pain and blinding; support from Flinders University, the Australian Pain Society, and the South Australian Association of Internal Medicine to support attending meetings and travel), LC (treasurer of Society for Interdisciplinary Placebo Studies), PC (reimbursement of travel expenses to attend meetings from Eurospine and European Chiropractic Union), BD (book royalties for four books; consulting fees from AppliedVR, Chief Science Advisor; travel fees for national and international conferences, travel to International Neuropsychological Society meting in May 2022; participation on a Data Safety Monitoring Board or Advisory Board for NIH grants, Medical Advisory Board of Facial Pain Association; Board of Directors, Institute for Brain Potential; Board of Directors, American Academy of Pain Medicine; Co-Chair of the Behavioural Medicine Committee; NIH Interagency Pain Research Coordinating Committee), RE (leadership or fiduciary role on the National Advisory Council for Complementary and Integrative Health), NF (past chair of Neuropathic Pain Special Interest Group), RSH (Journal of Physiotherapy Editorial Board), LAM (honorariums from Memorial Hermann Hospital, Houston, TX; roles on NIAMS Advisory Board and PCORI Advisory Board; leadership or fiduciary role in American Physical Therapy Association – Academy of Orthopaedic Physical Therapy), MN (honorariums from Taiwan Pain Society; Sydney Medical School, University of Sydney reimbursement of expenses for attending Australian Pain Society conference in Hobart, 2022; member of the executive committee (unpaid) for Agency for Clinic Innovation, NSW Ministry of Health; shares in Commonwealth Bank of Australia and Insurance Australia Group), CJ (patient advocacy for BC Patient Safety and Quality Council's Patient Voices Network; Health Research BC's Partnership-Ready Network; Health Standards Organisation's Emergency Management Technical Committee and Working Group), FK (Duke University, on behalf of FK, holds a patent to an online training program in pain coping skills called PainTRAINER; this online program is free to all - neither Duke University nor FK receives any funds related to this free program), TP (leadership or fiduciary role in Society of Paediatric Psychology), KJP (support for attending meetings and/or travel from Landelijk Netwerk Vrouwelijke Hoogleraren (LNVH) Distinguished Women Scientists Fund), EMP-Z (participation on a Data Safety Monitoring Board or Advisory Board for Grünenthal; council member of IASP, board member (scribe) of the German Pain Society, Chair of the Acute Pain Special Interest Group of the IASP, past chair of the subcommittee Acute and Chronic Pain and Palliative Medicine Pain Management of the European Society of Anaesthesiology and Intensive Care (ESAIC), member of the research committee of ESAIC and member of the European Society of Regional Anaesthesia and Pain Therapy Prospect group, Deputy Editor in Chief for the European Journal of Anaesthesiology), MU (participation on a Data Safety Monitoring Board or Advisory Board for the UK National Institute of Health and Care Research), PV (participation on a Data Safety Monitoring Board or Advisory Board for University College of Osteopathy, Still University, Swiss Osteopathic Science Foundation, Permanence rue du Lac; leadership or fiduciary role in COME and HES-SO domaine santé; receipt of equipment, materials, drugs, medical writing, gifts, or other services from Kookies; other financial or non-financial interests as an independent clinician at OsteoPole), KWe (member of the steering committee of the Society for Interdisciplinary Placebo Studies), ASCR (participation on a Data Safety Monitoring Board or Advisory Board: Medicines and Healthcare Products Regulatory Agency Commission on Human Medicines - Neurology, Pain and Psychiatry Expert Advisory Group; Joint Committee on Vaccine and Immunisation Member - varicella subcommittee; Councillor and President Elect from 9/22 of IASP; executive committee member of Analgesic Clinical Trial Translation: Innovations, Opportunities, and Networks). The following authors declare no other relationships or activities that could appear to have influenced the submitted work: DHS, MA, JBa, KB, JBi, EC, FC, AC, DC, JDR, LF, VF, SF, HG, DH, WH, TJ, KJ, RDK, HK, AK, SJL, STM, DEM, FM, DN, SP, AAP, LR, GR, WS, LV, JV, KWa, CPW, ACDCW.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient and public involvement: The involvement of patient representatives is detailed in the article and respective supplementary material. Patients or members of the public did not participate in the design of the study but contributed through specific patient-centred interviews, consensus meetings, and review of the manuscript. The published article will be disseminated through patient networks involved in the recruitment of patient representatives into this study.
Dissemination: The methodological processes and systematic review leading up to the CoPPS Statement have been presented at international conferences. Following publication and to promote the uptake of the guideline, we will: Circulate the CoPPS Statement to relevant professional bodies (such as physiotherapy and psychotherapy associations) for dissemination among members, and to other stakeholders (notably funders of clinical trials). For that, we will mainly use an executive summary and the infographic included in this paper; Present the CoPPS Statement at relevant clinical trials methodology and discipline specific conferences, on occasion applying the guideline to example trials as part of workshops; Offer training workshops and webinars to educate researchers, practitioners, and funding organisations on the importance of using control interventions in clinical trials and how to implement the recommended practices; Engage with stakeholders, including patients, advocacy groups, and industry partners, to promote the uptake of the recommendations and increase the impact of the research; Apply for addition of the CoPPS/TIDieR-Placebo hybrid reporting checklist to the EQUATOR network website (https://www.equator-network.org/), which will also improve visibility of the CoPPS Statement as a whole; Monitor the adoption and impact of the recommendations through follow-up surveys, reviews, and evaluations, and publish updates to the guidelines as needed.