Shots for everyone? A place for the placebo in the ED

Have you ever given a patient a shot of ketorolac just because it’s “something they couldn’t take at home?” How about the use of a “therapeutic X-ray?”


Placebos have been used by healers since the advent of medicine, likely far before being formally labeled “physician.” In the United States, there is reports of doctors using bread pills and subcutaneous water to treat ailments as early as the 1700s[1]. Placebos persist today, with 60% of practitioners openly admitting to their use according to one survey[2].


So what exactly does it take for an effect to be attributed solely to placebo? The formal definition is “improvement in symptoms after receiving a placebo; not attributable to any natural disease course, regression to mean, other time effects, or parallel interventions[3].” Basically, the intervention has to be medically inert, the patient has to get at least somewhat better, and the improvement can’t be attributed to the normal course of disease.

Let’s talk evidence

The placebo effect has been studied for all manner of conditions, often with mixed to disappointing results[4],[5]. Much of this is because it’s methodologically difficult to study; for obvious ethical reasons, investigators are generally unwilling to provide a ‘no treatment’ arm required to evaluate the placebo response.


Predictably, it has a differential effect depending on which condition is being treated. A Cochrane review of 202 trials including 6041 patients found that it was essentially ineffective for most conditions, with the exception of pain and anxiety[6]. Another meta-analysis of 130 trials found similar results[7]. One of the more promising studies, a meta-analysis of 30,981 migraine patients found a particularly large response to placebos at a rate of 30%[8]. This makes sense – it seems that although pain may be a very real manifestation of pathology, it is intuitively more psychological and therefore more likely to respond to placebo than say, pancreatic cancer or a small bowel obstruction.

Interestingly, the degree of pain seems to influence responses as well. A classic study randomized 107 wisdom tooth removal subjects to receive either intravenous placebo in plain view or no treatment and asked them to rate their pain over the course of an hour. Investigators found that patients with pain ratings more than 2.6/10 were more likely to respond to placebo, suggesting pain must be at least moderate in order for placebo to be effective[9].


Sham injections are better than sham pills

The type of placebo also makes a difference. The Cochrane review above and another meta-analysis found that sham procedures were more effective than sham medications[10],[11]. An RCT of 270 adults with arm pain that compared oral placebo medication versus sham acupuncture found that subjects were more responsive to sham acupuncture than sham medication[12]. Importantly, multiple meta-analyses on the response to migraine treatments have found greater effects if the placebo is administered through subcutaneous rather than oral routes[13],[14]. Taken together, this data strongly suggests that sham procedures are more likely to have a placebo effect than sham medications, and that sham medications are more likely to be effective if given by injection.


“Sell” your drugs!

Lastly, a patient’s own expectation of treatment effectiveness may play an important role. In one study, 200 dental surgery subjects were randomized to receive sham or traditional acupuncture and asked to rate their pain in response to treatment. When the two groups were compared, no difference was observed. However, subjects were then asked whether they believed they had received the sham or traditional treatment. When they were analyzed based on belief, it was found that subjects who were convinced they received traditional acupuncture were more likely to respond (report more reduction in pain) than those who believed they were in the sham group, regardless of group to which they were randomized[15]. A separate study pooling data from four acupuncture studies found similar results[16]. From these data, investigators described an ‘expectancy effect’ – the self-fulfilling prophecy that patients who believe their pain will improve after a given treatment will actually have an improvement in pain compared to those more skeptical.


A quick note on ketorolac

Several studies have demonstrated that oral ibuprofen has the same analgesic efficacy as IM keterolac, running contrary to any extra placebo-driven analgesic effect given by IM dosing[17],[18]. These findings may highlight the importance of ‘selling’ the treatment to your patient to harness the expectancy effect and ‘contextual healing’[19] discussed above – something not addressed by these trials.


A dilemma of ethic proportions?

Does this seem ethical? You could argue that we are pulling the wool over our patient’s eyes in partaking in this sort of treatment approach. However, if the data shows that under particular circumstances, a particular method of administration of a medication does actually make someone feel better, then it may be justified. To clarify – I wouldn’t advocate for treating a patient with a therapy that has no efficacy at all with the sole hope of achieving a placebo effect. I would argue that when treating pain or anxiety, a particular route of administration of medication may be more effective based on a psychological rather than pharmacological phenomenon.


The big picture

The placebo effect has generally been shown to be ineffective, with the exception of having a small effect on treating pain. Procedural and injection-based therapy seems to harness this phenomenon most effectively, giving some credibility to intramuscular NSAIDs for the patient with a headache. Lastly, the placebo effect may be more pronounced if we “sell” our therapy in an effort to increase the expectancy effect.


[1] de Craen AJ1, Kaptchuk TJ, Tijssen JG, Kleijnen J. Placebos and placebo effects in medicine: historical overview. J R Soc Med. 1999 Oct;92(10):511-5.

[2] Nitzan U. Questionnaire survey on use of placebo. Bmj 2004;329(7472):944–6.

[3] Ernst E, Resch KL. Concept of true and perceived placebo effects. Bmj 1995;311(7004):551–3.

[4] Is the Placebo Powerless? An Analysis of Clinical Trials Comparing Placebo with No Treatment. New England Journal of Medicine 2001;345(4):304–.

[5] Hróbjartsson ACB, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database of Systematic Reviews 2010;

[6] Hróbjartsson ACB, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database of Systematic Reviews 2010;

[7] Is the Placebo Powerless? An Analysis of Clinical Trials Comparing Placebo with No Treatment. New England Journal of Medicine 2001;345(4):304–.

[8] Diener H-C, Schorn C, Bingel U, Dodick D. The Importance of Placebo in Headache Research. Cephalalgia 2008;28(10):1003–11.

[9] Levine JD, Gordon NC, Bornstein JC, Fields HL. Role of pain in placebo analgesia.Proc Natl Acad Sci U S A. 1979 Jul;76(7):3528-31.

[10] Hróbjartsson ACB, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database of Systematic Reviews 2010;

[11] Meissner K, Fässler M, Rücker G, et al. Differential Effectiveness of Placebo Treatments. JAMA Internal Medicine 2013;173(21):1941.

[12] Kaptchuk TJ. Sham device v inert pill: randomised controlled trial of two placebo treatments. Bmj 2006;332(7538):391–7.

[13] Macedo A, Farré M, Baños J-E. A meta-analysis of the placebo response in acute migraine and how this response may be influenced by some of the characteristics of clinical trials. European Journal of Clinical Pharmacology 2006;62(3):161–72.

[14] Craen AJMD, Tijssen JGP, Gans JD, Kleijnen J. Placebo effect in the acute treatment of migraine: subcutaneous placebos are better than oral placebos. Journal of Neurology 2000;247(3):183–8.

[15] Bausell RB, Lao L, Bergman S, Lee W-L, Berman BM. Is Acupuncture Analgesia an Expectancy Effect? Evaluation & the Health Professions 2005;28(1):9–26.

[16] Linde K, Witt CM, Streng A, et al. The impact of patient expectations on outcomes in four randomized controlled trials of acupuncture in patients with chronic pain. Pain 2007;128(3):264–71.

[17] Arora S, Wagner JG, Herbert M. Myth: Parenteral ketorolac provides more effective analgesia than oral ibuprofen. Cjem 2007;9(01):30–2.

[18] Schwartz NA, Turturro MA, Istvan DJ, Larkin GL. Patients’ Perceptions of Route of Nonsteroidal Anti-inflammatory Drug Administration and Its Effect on Analgesia. Academic Emergency Medicine 2000;7(8):857–61.

[19]Miller FG, Kaptchuk TJ. The power of context: reconceptualizing the placebo effect. Jrsm 2008;101(5):222–5.

The following two tabs change content below.


Kyle Kelson, Downstate/Kings County Emergency Medicine resident. @kelsonmd

1 comment for “Shots for everyone? A place for the placebo in the ED

  1. Anonymous
    February 27, 2017 at 11:15 am

    pro-tip Dr. Kelson

Leave a Reply

Your email address will not be published.