#81: Placebos, nocebos, and the doctor as a placebo

February 5, 2018 | By Molly Heublein

Boost treatment efficacy and amplify the “doctor as a placebo” phenomenon with tips from placebo expert (and ex-cartoonist), Mark W. Green, MD, Director of Headache/Pain Medicine and Professor of Neurology, Anesthesiology, and Rehabilitation Medicine at the Icahn School of Medicine at Mount Sinai. Topics include: What’s that pesky sugar pill actually doing to us? Why do we get a boost from taking medicines that have no active compounds within them? How does placebo work? How can it be leveraged? What does it mean for study design/blinding? And what’s with its pesky dark twin–the nocebo effect? “Enjoy this, our only placebo-controlled episode”  (Thanks, Stuart!).

Written and produced by: Molly Heublein, MD and Nora Taranto MS3
Edited by: Matthew Watto, MD

Join our mailing list and receive a PDF copy of our show notes every Monday. Rate us on iTunes, recommend a guest or topic and give feedback at thecurbsiders@gmail.com.

Case from Kashlak Memorial: Mr. Sugar is a healthy 55-year-old male with knee pain. His knee exam and XRAYs demonstrate moderate knee osteoarthritis. In the past, he had intra-articular steroid injections and reports a positive response, despite mixed evidence for their efficacy in the literature (and your resulting doubt as to whether they’re beneficial). He wants another injection given past efficacy, and is also taking glucosamine/chondroitin, which he finds helpful. He wants your opinion on whether or not he should continue.

Take a Self-Assessment on your placebo knowledge!

 Clinical Pearls about Placebo  

1. Placebo hierarchy: Surgery>>injection>>pills.
2. Pill hierarchy: Gold>>silver>>colors>>white. Bigger>>smaller pills. Brand names>>cheap generics. Several times daily>>once daily dosing.
3. “Doctor as a Placebo”: Placebo is in everything we do and everything we prescribe. A therapy’s effect is amplified, diminished, or neutralized by a practitioner’s demeanor, responses, and interaction with the patient. We shape a patients’ expectations and experiences with a given treatment.
4. Placebo amplification: Response enhanced when providers: 1) are enthusiastic about treatment 2) have good communication, listening skills, and warmth 3) work in a center with a good reputation
5. It is ethical and appropriate to try to enhance the efficacy of anything we prescribe so that patients benefit from it to the maximal extent. One way to maximize treatment efficacy is to establish therapeutic rapport and get as much patient buy-in to the therapy you are advocating as possible.
6. We should read journals with skeptical lens. Ask yourself: What is the placebo response in this particular trial? How are they using placebo arms? Was blinding done effectively? Remember: Trials that are theoretically single or double blinded may in fact not be. Patients, physicians, and a patient’s relatives are quite good at guessing to which arm the patient has been assigned.
7. How should we respond to patients who ask our opinion on using treatments that have evidence showing no benefit over placebo? e.g. Glucosamine, chondroitin, or corticosteroid injections. There’s no absolute answer, but Dr Green suggests:

1) Allow therapies, with soft evidence in their favor.
2) Ensure safety i.e. use of benign, harmless product
3) Use a response that won’t ruin “placebo effect” e.g. “Though there is limited evidence for its active therapeutic efficacy, this treatment may provide some benefit to you.”

Background

1. A treatment’s effect has two components 1) specific and related to the effects of treatment and 2) nonspecific and related to perception of that treatment. When a nonspecific treatment effect is beneficial we call it placebo and when harmful we call it nocebo (Planes S et al Pharmacol Res Persp 2016)
2. Placebo response: A positive response to a treatment without an “active ingredient”. Dr Green notes that sugar induces dopamine release so sugar pills aren’t necessarily inactive.
3. Nocebo response: Sharing information about risks of treatment induces a “nocebo response”, where a patient experiences negative “side effects”. We must educate patients on risks of treatment, BUT also make sure to spend time on the risks of not treatmenting as well.

The Physiology of Placebo Response

1. Dopamine and Reward: Expectation of improvement is at the center of placebo response. Dopamine levels increase based on expectation of improvement or reward. Receptors are deactivated when there is no expectation of improvement. Persistently high dopamine levels correlate with persistence of drug efficacy. We reward ourselves with things that increase dopamine e.g. cigarettes, cocaine, or carbohydrates, which explains the success of Dylan’s Candy Bar during the stock market crash of 2008! And yet, we often use sugar pills as our “placebo” in randomized controlled trials with a placebo arm. Is it actually a neutral, inactive “placebo”?
2. Pain and placebo: When placebo analgesics are given, Mu opioid receptors in brain get activated in a similar manner to active opioids. Patient’s have more self reported pain relief (whether given placebo or active drug) when told that a therapy is an “effective painkiller” before taking it. A given therapy produces superior pain relief if primed with an expectation of relief by the prescriber.
3. Proof of principle with Naloxone trials: (Levine. Lancet. 1978). Naloxone blocks Mu opioid receptor activation, and reverses pain relief from active opioid drugs. Trial: Patient given placebo painkiller. Pain decreases on visual scale. Naloxone surreptitiously infused. Patient’s pain increases on visual scale. This proves that placebo painkiller induced an endogenous opioid response to relieve pain.
4. Dinner party experiment: Real caffeine activates same part of the brain as placebo caffeine. Offer a round of coffee near the end of the night at your next dinner party. Most people will say they want decaf, but give them real coffee (!). The next morning, ask how they slept. If you convinced them it’s decaf coffee they’re far more likely to have a good sleep than if you told them it was regular. In fact, many people will sleep just the same whether you give them fully caffeinated coffee or decaf.

Study design

1. Studies do not always take time address the placebo effect, or even use it as one of the control arms. Randomized, double blind, placebo controlled trials are considered ideal, but even these have some limitations.
2. Non-inferiority trials: Compare new drug, X, to standard therapy, drug Y. Proves drug X no worse than drug Y by a predefined margin, but doesn’t prove a drug’s efficacy versus a placebo control. Drug companies do this to avoid showing inefficacy and also because it is considered unethical to treat life threatening conditions with placebo.

Biases in study design that increase placebo response:

1. Physician buy-in: To truly evaluate efficacy, there should be equipoise about the therapy. But to have physicians participate and enroll patients in clinical trials, there is often some enthusiasm about the therapy.
2. Eligibility creep: “the tendency for patients [in a trial] to have higher measured severity at initial assessment visits when eligibility is determined” (Hick. J Am Acad Dermatol. 2007)
3. Regression to the mean: “phenomenon that if a variable is extreme on its first measurement, it will tend to be closer to the average on its second measurement”. Research studies tend to use people with more severe disease presentations (eligibility creep) and must account for their “regression to the mean”. E.g. If you enrolled someone with severe hypertension, did nothing and followed her over time, her next blood pressure check would likely be lower just due to natural variation.
4. Many studies are short term which limits our ability to predict longer term responses/loss of placebo response.
5. Blinding: Just how good is it?  Many studies have been done. In most clinical trials, the patient, doctor, and family members correctly guess that they’re getting placebo. Even in double-blind RCTs! In one review of 27 studies, 13082 patients, drug allocation guessed correctly by 67% of clinicians, 65% of patients, and 71% of  relatives/staff (Shapiro, et al. J Psychiatr Res. 1979).
6. One alternative study design: Start everyone on placebo, take out responders, then run the clinical trial amongst the non-placebo responders. This is more expensive, and would probably obviate some of the reportable effect magnitude.

Sham surgery: An interesting and ethically ambiguous history of invasive placebo procedures

1. How can we study the efficacy of a procedure vs the risks of performing sham surgery in trials? Important when a surgical technique is accepted as standard of care, but has no data to support its use. Since we know surgery itself induces a strong placebo response, how do we compare a surgery vs a control?
2. Internal mammary artery (IMA) ligation for angina used to be standard of care. It was thought to drive blood into the heart and improve survival in angina. This procedure was quoted as being 90% effective in treating angina. Then, in the 1959, Cobb, a surgeon in Seattle, performed a study: He did a thoracotomy with or without IMA ligation, and demonstrated no difference in outcome with or without the IMA. (Cobb et al. N Engl J Med. 1959)
3. Vertebroplasty for osteoporotic compression fractures was studied using sham surgery in multiple RCTs. Vertebroplasty technique is no better than placebo (Wali et al. Surg Neurol Int. 2017)

Genetic Basis of the Placebo Effect:

1. So far, more than 10 genes have been identified/implicated in the placebo response. It’s all about dopamine, serotonin, and opioids. Hall et al. Genetics and the Placebo Effect: the Placebome. Trends Mol Med. 2015; 21(5):285-294.
2. COMT: Mutations in COMT enzymes affect rates of dopamine metabolism.  Reduced metabolism of dopamine leads to increased placebo response.
3. Genetically, the ability to heal yourself through placebo could provide an evolutionary advantage. Nocebo responders tend to have dopamine deactivation, placebo responders have dopamine activation.

Goal: Listeners will learn the basics of how to talk to patients about therapies that might take advantage of a placebo effect, how physicians can themselves attempt to amplify the power of their therapies through encouragement, and how placebo affects our interpretation of randomized clinical trials and blinding.

Learning objectives:
After listening to this episode listeners will be able to…

1. Define placebo effect, nocebo effect, placebo amplification
2. Understand the role of the “doctor as placebo”
3. Understand the physiology underlying the placebo response
4. Improve patient communication to maximize the efficacy of treatments they prescribe
5. Analyze clinical trial results and their utilization of the placebo arm
6. More carefully interpret the methods and blinding information in RCTs

Disclosures: Both Dr Green and The Curbsiders report no relevant financial disclosures.

Time Stamps

• 00:00 Disclaimer
• 00:37 Intro banter
• 01:30 Picks of the week
• 08:02 Getting to know our guest
• 14:00 Clinical case of pain
• 15:08 Placebo hierarchy
• 17:19 Non-inferiority studies
• 18:20 Clinical trials, placebo, eligibility creep and regression to the mean
• 22:04 Nocebo response
• 27:05 Physiologic response to placebo
• 29:13 Placebo amplification
• 30:18 Caffeine experient
• Dr Green reports no relevant financial disclosures
• 32:47 How to counsel patients interested in alternative or ineffective therapy
• 34:22 Surgery and placebo; sham surgery
• 37:15 Open-label placebo and genetic response to placebo
• 42:04 Blinding in placebo controlled trials may be ineffective
• 43:53 Script for counseling patients on alternative therapy
• 45:51 Take home points
• 47:56 The Curbsiders recap and lessons learned
• 52:52 Outro

Links from the show:

1. Molly’s Pick of the Week: Painweek.org podcasts https://www.painweek.org/podcasts.html
2. Matt’s Pick of the Week: Predictably Irrational, by Dan Ariely http://danariely.com/tag/predictably-irrational/
3. Stuart’s Pick of the Week: The Princess Bride, by William Goldman (the book, not the movie!) https://en.wikipedia.org/wiki/The_Princess_Bride
4. Dr. Green’s book recommendation: anything by Oliver Sacks, teaching us about humanity https://www.oliversacks.com/
5. Shapiro, et al.  The reliability and validity of a placebo test.  J Psychiatr Res. 1979;15(4):253-90.
6. Physiology of the placebo in the pain response. Naloxone trials. Levine, Gordon, & Fields. The Mechanism of Placebo Analgesia. Lancet. 1978; 2(8091):654-7 http://dx.doi.org/10.1016/S0140-6736(78)92762-9
7. Internal Mammary Artery Ligation and Placebo: Cobb et al. An evaluation of Internal-Mammary-Artery Ligation by a Double Blind Technique. N Engl J Med. 1959; 260:1115-1118.
8. Vertebroplasty and sham surgery: Wali et al. Vertebroplasty for vertebral compression fractures: placebo or effective? Surg Neurol Int. 2017; 8: 81.
9. Genetic basis of placebo: Hall et al. Genetics and the Placebo Effect: the Placebome. Trends Mol Med. 2015; 21(5):285-294. http://dx.doi.org/10.1016/j.molmed.2015.02.009
10. TAILOR trial, Patients prefer their doctors in formal attire. Petrilli et al. Understanding the role of physician attire on patient perceptions; a systematic review of the literature–targeting attire to improve likelihood of rapport (TAILOR) investigators. BMJ Open. 2015; 5:e006578. http://bmjopen.bmj.com/content/5/1/e006578

Recommended Reading:

1. A good overall review : Placebo effects: Biological, Clinical and Ethical Advances. Finniss et al. Lancet. 2010: 375(9715): 686-695. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2832199/
2. Antidepressants and Placebo, and flaws in drug study design: Khan & Brown, Antidepressants vs. Placebo in Major Depression: An Overview. World Psychiatry. 2015; 14(3):294-300. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4592645/
3. Price increases Nocebo effects: Tinnermann et al. Interactions between brain and spinal cord mediate value effects in nocebo hyperalgesia. Science. 2017; 358(6359). http://science.sciencemag.org/content/358/6359/105 (and a good article about this in Science Magazine: http://www.sciencemag.org/news/2017/10/pricier-meds-mean-worse-side-effects-thanks-nocebo-effect)
4. Open-label placebos: a positive effect for non-deceptive placebos. Petkovic et al. Effects of placebos without deception compared with no treatment: a systematic review and meta-analysis. J Evid Based Med. 2017; 10(2):97-10.  http://dx.doi.org/10.1136/bmjopen-2015-009428
5. The Rising placebo effect (From Science, 2015) http://blogs.sciencemag.org/pipeline/archives/2015/10/15/the-rising-placebo-effect
6. The placebo response in painkiller RCTS appears to be increasing in size. http://www.nature.com/news/strong-placebo-response-thwarts-painkiller-trials-1.18511?WT.mc_id=TWT_NatureNews
7. How much are physicians using “placebos”? Tilburt et al. Prescribing “placebo treatments”: results of national survey of US internists and rheumatologists. BMJ. 2008; 337:a1938. http://www.bmj.com/content/337/bmj.a1938
8. The Doctor As Placebo: De Blasi Et al. Influence of context effects on health outcomes: a systematic review.  The Lancet.  Volume 357, Issue 9258, 10 March 2001, Pages 757-762. https://doi.org/10.1016/S0140-6736(00)04169-6

Placebo can influence physiology, in particularly through the endogenous opioid and dopaminergic systems. Scott DJ et al: Placebo and nocebo effects are defined by opposite opioid and dopaminergic responses. Arch Gen Psychiatry 65:220, 2008  https://www.ncbi.nlm.nih.gov/pubmed/18250260