Is a placebo effect good or bad?

Understanding the influence of placebo and nocebo in clinical practice




Ok, so this one is a little longer than previous posts. If you need a minute to freshen your coffee or grab a snack, I completely understand. This topic requires a little more time to unpack due to its implications. Placebo and nocebo are a component of nearly every patient interaction. Researchers carefully design their experiments to tease out placebo effects (some more carefully and successfully than others). It impacts the interventions we decide to apply, how to apply them, and the success we have with them. You may think I am over-reacting and that is fine. All I ask is that stick with this post to see where I am coming from.

What is the Placebo Effect?

The placebo effect can be broadly defined as a change in a person’s health status that results from the meaning and hopes the person attributes to a procedure or event in a health care setting.[1] We experience placebos every day, not just in the clinic. Take my son for example. Like any toddler, he is prone to the occasional fall on the sidewalk, head bump against a table, finger caught in a door, or dog tail whip to the face. Sometimes I am able to cut off the sadness and subsequent crying with an immediate distraction or chanting “brush it off” or “good catch” but other times the trauma is simply too much to bear. It is at this point that the big guns are brought out…magic kisses. That’s right, a quick kiss on the knee and magically the pain goes away and we are ready to resume our walk and point out the color of literally every car in the neighborhood. As we all know, there is nothing magical about the kiss on the knee. Furthermore, to ensure I don’t have a future uncomfortable conversation with HR managers, please do not start kissing your patients in an effort to reduce their pain.


Getting into the actual interventions and the impact of a placebo, there are two categories to cover: the usage of interventions known to primarily be a placebo and enhancing the effectiveness of your interventions with placebo. This is where the field of modalities takes front and center stage. Most modalities (hot/ice packs, ultrasound, e-stim, diathermy, etc.) show either minimal effects relative to controls/sham interventions or no difference indicating results achieved are likely largely due to the placebo effect. This does not mean they are inherently useless, but they should be approached with caution. If a patient is not responding to more supported and efficacious treatments, or perhaps they have an expectation to receive a certain intervention, and you deem the treatment to be safe based on available literature and the patient has been fully educated on risks, benefits, and alternative options, the modality may be appropriate to utilize.


You must be upfront about the effect of the treatment. Or do you? There is some debate amongst clinicians and researchers on this.  Telling a patient that the ultrasound can reduce their pain is great and can enhance the placebo effect. Telling a patient, the ultrasound can reduce their pain by speeding up the tissue recovery and that it mends your tissue back together during the procedure is another story. We do not have evidence to support the latter and it is unethical to deliberately misinform our patients. We will dive deeper into the ethical considerations at the end of the post, but first, we need more foundational information.


How do we harness the placebo effect?



Contrary to popular opinion, evidence has demonstrated that telling a patient an intervention is likely to be a placebo does not negate future benefits.[2-6] My initial recommendations are as follows: Inform the patient of the reasons why something may work, even if it is likely placebo, and follow that with past experiences of success with previous patients. Do not guarantee anything but highlight that your patients often feel much better following ultrasound and that you think the patient is a great candidate for the intervention. Ensure you emphasize the need for follow-up interventions that yield greater outcomes, and facilitate long-term improvements (i.e. exercise). This is not a license to simply administer any intervention provided you acknowledge placebo is the likely cause. Efficacious treatments should still be prioritized.


There are indirect ways to bolster your treatment intervention. To highlight this, take a study by Krell et al who simply assessed the impact of an effectiveness label on an anti-depressive medication. They found that up to 90% of patients with an expectation of improvement experienced a reduction of depressive symptoms when the medication was labeled as “very effective”, while only 33% experienced an improvement when the medication was labeled as “somewhat effective”.[7] Keep in mind, pharmaceutical trials have far fewer variables to control than physical therapy trials and still, a massive effect was observed. In a previous post on patient satisfaction, I discussed how a provider dresses and interacts with patients can impact the outcome, as well as the overall clinical environment. Another consideration is patients witnessing the delivery and outcome of interventions for other patients in the building. For example, if Jane witnesses John receiving a manipulation and he immediately expresses how much better his back feels, she is more likely to order a manipulation off the menu and have a positive outcome. Never underestimate the power of a positive clinical environment.



Herbert Adler wrote a paper in the Journal of General Internal Medicine which focused on the patient-clinician relationship.[8] He suggested that a placebo can be used as a clinical intervention with three primary features that support the idea.

  1. The engagement of clinical empathy in the context of disease and relief of suffering requires a specific medical interaction.

  2. The use of clinical empathy would require a set of interpersonal skills that depend upon active “emotional labor: by the clinician.

  3. Clinical empathy is a purposeful process attempting to attain specific outcomes within the clinical context that will result in the improvement of health through the engagement of psychobiological mechanisms.

It is important to note that the clinician’s expectations will impact the outcomes of the treatment. Let’ say a patient comes into the clinic with a script for iontophoresis. You have reviewed the literature and deem the treatment unnecessary, but the referring physician and clinician demand you provide the intervention. Even if you do your best to suppress your eye roll, chances are the patient will notice subtle behavior changes that can impact the treatment effectiveness. Even if the studies that demonstrate positive outcomes are likely placebo driven, your actions may negate the potential benefits in the clinic. “Clinical analgesia” depends on both the physiological action of a given intervention and the expectations of both the patient and the clinician.[9]


The opposite side of the coin

In addition to the impact of our body language, phrases that may be traditionally innocuous or even intended to be humorous can negatively impact a patient. I encourage caution when considering using phrases such as “no pain, no gain”, “it hurts so good” and “it’s supposed to hurt, that’s how we know it is working.” Pain is very subjective, variable, and highly influenced by a multitude of personal and environmental factors. This brings us to the other side of the equation, the nocebo effect.


A “nocebo” can be defined as an intervention intended to create negative expectations about health outcomes either intentionally (e.g. disclosure of side effects) or unintentionally (e.g. reading an MRI report).[10] Nocebo effects can be elicited through verbal suggestions, conditioning, and/or social observation, and they can influence pain pathways.[10] Strong predictors of nocebo include higher negative expectancies, clear suggestions of possible symptoms, observing others develop symptoms, and a high perceived dose of exposure. Your initial response may be, “Great. I’ll make sure I never disclose possible risks, make sure all my dramatic and high irritability patients are one on one only so they don’t “infect” my other patients with their negativity, and keep all exercise intensities low. Boom! Everyone wins. In the words of the always entertaining Lee Corso (it looks like we won't get the chance to hear him this fall), not so fast my friend!



In some circumstances, such as disclosing potential risks and explaining post-surgical recovery times, a nocebo is unavoidable. However, the effect does not have to be a net negative. We can both enhance placebos and minimize the magnitude and number of nocebo effects. In fact, there are many nocebo predictors that should be minimized in a clinical setting. Recently, I wrote about the impact of non-verbal behaviors (NVBs) on patient satisfaction and outcomes. These fall into the realm of placebo and nocebo. As a quick refresher, the two most impactful NVBs for nocebo were infrequently looking at your patient and being further away from them in the clinic (pre-social distancing and 6 ft. rule). For this post, I am going to focus more on the words that we use.


Words matter

Once a weekly basis I come across some type of #choosePT or #PTfirst social media post and many of them focus on either immediate referral to PT or minimizing the use of medical interventions. One of the most common targets of attack is the MRI. I am not here to defend the MRI as the evidence is quite clear that the overutilization remains staggering despite nearly every medical guideline stating they should only be ordered if red flags are noted and not for routine care.[11-13] Why are they such as issue? Aside from the significant cost increase and the lack of diagnostic and prognostic utility, they are massive nocebo drivers


We see pretty clearly in the evidence that structural ‘findings’ on an MRI are often meaningless for our patients. Whether it be ‘malalignment’, herniations, arthritis, space narrowing, or tendon fraying/tears, they have little to no impact on the drivers of pain and do not provide valuable information to guide treatment decision making. What they are very good for however is freaking your patient out. Nothing says you are broken and require someone else to “fix” you like ‘malalignment’, herniations, arthritis, space narrowing, or tendon fraying/tears.


Studies demonstrate that patients achieve worse outcomes when they receive an MRI and are informed of the results.[14] Here’s the kicker, aren’t we often doing the same thing following a largely pathoanatomic assessment? What is a patient thinking after we tell them they are weak, their mobility is poor, their reflexes are abnormal, they lack motor control, and they are riddled with trigger points? Not only that, but you are also doomed to future injury, pain, and medical procedures if I don’t fix you immediately! The evidence for those ‘findings’ predicting prognosis and explaining a patient’s pain is quite poor as well.[15] Instead of becoming walking, talking MRIs, we can better serve our patients by avoiding the nocebo and scare tactics to help them “buy into” therapy. We can easily sell the benefits of PT without painting a picture of doom and gloom.



The art of patient education

Creating value, establishing a plan of care, and recognizing the patient’s needs is a multi-stage multi-modal process that involves far more than a pathoanatomic assessment focus. This is where we bring into the equation motivational interviewing, goal setting, building patient rapport and trust, determining both risk and resilience factors, and assessing the psychological impact of our words and actions. Pain is far more complex than nociception and impairments in strength and mobility. We will be more effective in building a therapeutic alliance, establishing value, and achieving a positive outcome if we focus on the aforementioned list. This does not mean everything is rainbows and sunshine and we hide all potential deficits for fear of offending or scaring our patients. Instead, focus on what is important to the patient and the evidence-based methods for achieving those goals. As challenging as it may be, we need to abandon the scare tactics and the treatments and assessments that lack support in the evidence.


Another consideration we must take into account is the context and perspective of the individuals, the clinical setting, and treatments. This is what makes translation from research to clinical practice difficult. We cannot assume that every type of verbal and non-verbal behavior will be interpreted the same, just as we cannot provide the type and intensity of exercise or manual intervention to each patient, regardless if they have similar diagnoses. Mindsets and responses can be formed and shaped by culture, socialization processes, spirituality and religion, and media exposure to name a few. Each of these factors can directly impact decision-making, behaviors, and attitudes. If accounted for, the strategic use of questioning and motivational interviewing can lend insight into effective and efficient ways to navigate these different aspects of the clinical interaction and setting and allow you to deftly maximize placebo and minimize nocebo.


What are the ethical considerations of using the placebo effect?

To wrap up placebo/nocebo and this series, I want to touch on ethical concerns regarding the use of nocebo and placebo. As was stated previously, a placebo still yields a benefit even if the patient is fully aware it is playing a factor. Clinicians should attempt to maximize the hypoalgesic effect of placebo within (1) ethically accepted parameters (guaranteeing relief would be unethical) and (2) accepted interventions for musculoskeletal pain conditions. Communication about side effects can increase the probability that side effects will be experienced (some would argue disclosing minor side effects is in unnecessary nocebo), however, clinicians have an obligation to obtain informed consent in addition to an obligation to minimize risks to patients from interventions (including the risks associated with information disclosure). In general, interventions should be accompanied by reassuring, empathetic, and supportive communication.


While clinicians may purposefully harness the placebo effect to improve outcomes, some may aggressively convey the benefits of an intervention and guarantee a positive outcome (enhancing patient expectations) rather than through more subtle means such as empathetic communication, professional dress, and designing an inviting clinic environment. When assessing the purposeful use of a placebo, there are two primary categories: pure and impure.


Pure, or inactive, include examples such as sugar pills or injections. Impure would be providing a nutritional supplement in the absence of a nutritional deficiency. On the rehab side, pure would be applying ultrasound with the machine turned off but the patient believes it is on and beneficial (yes this does happen). Impure fall within the majority of rehabilitation placebos (most modalities, kinesiotape, some applications of manual therapy, etc.). While you can conduct a double-blind study to create an ethical pure placebo, in the clinical setting, pure placebos can only be applied with deceiving the patient. This is unnecessary, as stated before, evidence suggests it is possible to elicit clinically meaningful placebo effects even with openly acknowledging to the patient a placebo is likely the primary cause of the outcome.[2-6]


What do the patients want?



An interesting study by Bishop et al1 asked this very question and the results were mixed. Initially, it took some convincing that placebo effects are real, and the mind can overpower traditional cartesian explanations for experiencing symptoms. While the study participants eventually accepted the impact, leaving them thoroughly impressed with the power of the brain and its implications, they became even more accepting when considering children (see “magic kisses” at beginning of the post). However, the primary concern of the paper was whether a practitioner knowingly using a placebo is ethical. For both scenarios, the physician informing the patient of prescribing a placebo (at the beginning, not after the fact) and the physician deceiving the patient, it was a mixed bag of feelings. Here are two conflicting viewpoints:

“But placebos will only work if the doctor is dishonest. The question is - that if the, if it was just a common cold it will be cured in time, the body will cure itself; in that case dishonesty’s justified I think, give out placebo, by all means and then, then make them believe it. I mean you certainly don’t say that it doesn’t contain any active ingredients. You don’t tell them that…I mean I don’t think it’s immoral of the doctor to give a placebo if he’s convinced that the man has got nothing seriously wrong with him, but if he thinks that the man will be happy, then don’t tell him that it’s got no active ingredients, say here’s this, try this, you. The doctor won’t say this will cure you, all a doctor will say is try this you might get better.” “This may make the underlying problem worse cause then the person then becomes totally dependent on the doctor giving him something that isn’t actually, is nothing more than a bit of TLC… I don’t think he really should’ve given him anything. He should tell him that he doesn’t need anything”

There are many ethical considerations to take into account such as financial (did you disclose costs of the placebo and are you deceiving the patient on what they are paying for), potential adverse effects, addressing a single issue such as pain (tunnel vision), hope/false hope, replacing an alternative evidence-based treatment with a placebo-driven treatment lacking evidence, and removing patient autonomy/choice. This is why the “benefit” for treatment is not black and white.


Typically, there are two primary motivators for clinicians when prescribing an intervention and designing a plan of care: achieve a positive outcome and keep the patient. This can lead to justification of placebos (“all that matters is the patient got better”) and nocebos (“If you don’t come into therapy you will need surgery in the future”). But we don’t need these approaches to be successful, in fact, they will hinder our long-term impact. There is an abundance of strategies to achieve great outcomes through evidence-based, patient-informed treatments that are built on trust. If we display an attitude of caring, respect, and patient empowerment through meaningful patient-centered discussions, we can develop a fundamental basis for engaging the placebo effect and prevent the nocebo effect in any treatment that is decided upon.[9]



I ended with the ethical considerations rather than a wrap up of the series intentionally. I understand that many of the information in the posts can be both challenging and frustrating. Furthermore, they can lead to some uncomfortable decisions and recognition of current treatment decision making. It is vital that we put the patient first and ensure we are providing them with best practice. While that can be uncomfortable as it may result in abandoning treatments and assessments that we have had success with in the past, physical therapy (and healthcare as a whole for that matter) can only progress if we follow the evidence. It is easy to point to other professions and cry foul when they utilize outdated or completed unsupported treatment interventions but are we holding ourselves to the same standard?

References

  1. Bishop FL, Aizlewood L, Adams AE. When and why placebo-prescribing is acceptable and unacceptable: a focus group study of patients' views. PLoS One. 2014;9(7):e101822.

  2. Kaptchuk TJ, Friedlander E, Kelley JM, et al. Placebos without deception: a randomized controlled trial in irritable bowel syndrome. PLoS One. 2010;5(12):e15591.

  3. Ballou S, Kaptchuk TJ, Hirsch W, et al. Open-label versus double-blind placebo treatment in irritable bowel syndrome: study protocol for a randomized controlled trial. Trials. 2017;18(1):234.

  4. Carvalho C, Caetano JM, Cunha L, Rebouta P, Kaptchuk TJ, Kirsch I. Open-label placebo treatment in chronic low back pain: a randomized controlled trial. Pain. 2016;157(12):2766-2772.

  5. Colloca L, Pine DS, Ernst M, Miller FG, Grillon C. Vasopressin Boosts Placebo Analgesic Effects in Women: A Randomized Trial. Biol Psychiatry. 2016;79(10):794-802.

  6. Locher C, Frey Nascimento A, Kirsch I, Kossowsky J, Meyer A, Gaab J. Is the rationale more important than deception? A randomized controlled trial of open-label placebo analgesia. Pain. 2017;158(12):2320-2328.

  7. Krell HV, Leuchter AF, Morgan M, Cook IA, Abrams M. Subject expectations of treatment effectiveness and outcome of treatment with an experimental antidepressant. J Clin Psychiatry. 2004;65(9):1174-1179.

  8. Adler HM. Toward a biopsychosocial understanding of the patient-physician relationship: an emerging dialogue. J Gen Intern Med. 2007;22(2):280-285.

  9. Blasini M, Peiris N, Wright T, Colloca L. The Role of Patient-Practitioner Relationships in Placebo and Nocebo Phenomena. Int Rev Neurobiol. 2018;139:211-231.

  10. Blasini M, Corsi N, Klinger R, Colloca L. Nocebo and pain: An overview of the psychoneurobiological mechanisms. Pain Rep. 2017;2(2).

  11. Svanbergsson G, Ingvarsson T, Arnardottir RH. [MRI for diagnosis of low back pain: Usability, association with symptoms and influence on treatment]. Laeknabladid. 2017;103(1):17-22.

  12. Fritz JM, Brennan GP, Hunter SJ. Physical Therapy or Advanced Imaging as First Management Strategy Following a New Consultation for Low Back Pain in Primary Care: Associations with Future Health Care Utilization and Charges. Health Serv Res. 2015;50(6):1927-1940.

  13. Delitto A, George SZ, Van Dillen L, et al. Low back pain. J Orthop Sports Phys Ther. 2012;42(4):A1-57.

  14. Webster BS, Bauer AZ, Choi Y, Cifuentes M, Pransky GS. Iatrogenic consequences of early magnetic resonance imaging in acute, work-related, disabling low back pain. Spine (Phila Pa 1976). 2013;38(22):1939-1946.

  15. Clifton DR, Grooms DR, Hertel J, Onate JA. Predicting Injury: Challenges in Prospective Injury Risk Factor Identification. J Athl Train. 2016;51(8):658-661.


This post can be found on the PT Solutions Live Clinically Blog

Want to see a specific topic covered? Let me know!