Harnessing the placebo response
There may be a few circumstances in psychiatric practice when it makes sense to intentionally prescribe a placebo as treatment, and we discuss those below. But far more frequently, what we know about the elements that contribute to the placebo effect can be applied to enhance the benefits of any treatment. Patients might be best served if deliberate mobilization of the placebo effect was a standard adjunct to conventional clinical care.
Various components of the treatment situation, collectively referred to as placebo, are a powerful antidote for illness, and some of these healing components exert their influence without special activity on the clinician’s part:
- Simply seeking psychiatric care can bring relief by providing some sense of control over distressing symptoms. The standard trappings of the office or clinic and customary office procedures—from the presentation of one’s insurance card to taking a history—offer reassurance and evoke the expectation that improvement or recovery is around the corner.
- The comfort provided by the psychiatrist’s presence is enhanced when patients feel that they are in the hands of a recognized healer. Psychiatrists inspire confidence when they look like a psychiatrist, or more precisely, like the patient’s idea of what a psychiatrist should look like. In our culture, that means a white coat or business attire.
A thorough evaluation is one of the common treatment elements that does the most to reduce distress and inspire confidence. The quality of an evaluation bears a strong relationship to patients’ satisfaction with the medical encounter, and can influence the amount of disability they suffer.3,12-15
Although guidelines for conducting effective psychiatric interviews have been around for almost 100 years, psychiatrists vary considerably in the extent to which they elicit complete and accurate information, build rapport, give patients the sense that they are listened to, and provide a thorough assessment. The degree to which patients feel that the clinician is responsive to their concerns depends as much on the style of the interview as on the amount of time devoted to it. Nonverbal behavior can carry the message that the clinician is paying full attention. Something as simple as not answering the phone during an interview (this seems obvious, but a surprising and troubling number of mental health professionals take phone calls during interviews and treatment sessions) conveys an important message about the importance that the clinician places on the patient’s problems.3
The idea that the treatment situation itself provides reassurance and reduces distress, and in doing so, powers a good bit of the placebo effect, is enshrined in such concepts as the importance of good bedside manner. Many feel that the doctor’s thoughtful attention, positive regard, and optimism—so valued by patients—are justified on humanitarian grounds alone; actual evidence that this caring behavior contributes to healing isn’t required. To many, the healing properties of the treatment situation are self-evident. But as the costs of health care snowball and the demands for efficiency and cost-effectiveness rise, the time that psychiatrists can devote to patients has dwindled. Third-party payors demand evidence, beyond intuition and common sense, that diagnostic procedures and treatments have some usefulness, and rightly so.
Is there any evidence that the common components of the treatment situation provide benefit?3 More specifically, does the quality of the doctor–patient relationship and the patient’s feelings about a therapeutic encounter promote healing? Several studies suggest that the doctor–patient relationship has a demonstrable impact on symptom relief.16 In 1 study, oncologists were randomly assigned to receive a Communication Skills Training (CST) program or not. CST included a 1.5-day face-to-face workshop and 6 hours of monthly videoconferencing that focused on improving communication skills with patients.17 Lessons included building rapport, engaging in appropriate eye contact, and normalizing difficult experiences. One week after initially consulting with their physician, patients who saw an oncologist in the CST group experienced less anxiety and depression than those who saw an oncologist who did not receive CST. The benefit of CST for patient anxiety mostly persisted at a 3-month follow-up.
A recent meta-analysis pooled the results of 47 studies to examine the relationship between how much trust patients have for their doctors and health outcomes. There was a small to medium association: More trust was associated with greater improvement.18 It is possible that a good doctor–patient relationship enhances expectancies. However, it is also likely that a positive therapeutic relationship is inherently soothing and reduces distress or dysfunction independent of expectation. Regardless of the precise mechanism, these studies warrant attention. We all understand that it is important on ethical grounds to treat patients with respect and kindness. Research shows that this type of behavior also promotes recovery.
Patient expectations. The idea that expectation of improvement has a major impact on treatment outcome is firmly grounded in research on the placebo effect. Studies have shown that what people expect to experience as an outcome of treatment has a substantial impact on what they actually experience. In a classic study, a doctor told some patients with symptoms of minor illness that they would feel better soon and another group with the same symptoms that he didn’t know what ailed them.19 Two weeks later, 64% of patients in the “positive expectation” group were improved, compared with only 39% of patients in the “negative” group. In another study, adults were exposed to an allergen that caused a skin reaction.20 Hand lotion (ie, a therapeutically inert substance) was then spread on the skin. Patients were led to believe that the cream would either alleviate or exacerbate the itching. The experimentally-induced wheal-and-flare was measured in both groups a few minutes after the allergen and cream were applied. The wheal-and-flare were worse for participants in the group that expected exacerbation.
Not uncommonly, expectation can have more impact on clinical outcome than a drug’s pharmacologic activity. In a double-blind placebo-controlled study, patients with depression were treated with St. John’s wort, sertraline, or placebo.21 They improved to the same extent with all 3 treatments. But when patients were asked to guess the treatment to which they had been assigned, those who thought they had received placebo showed little improvement, irrespective of which intervention they actually received, and those who guessed they had been given St. John’s wort or sertraline showed uniformly large improvement, irrespective of which intervention they actually received (including placebo). The researchers concluded that “Patient beliefs regarding treatment may have a stronger association with clinical outcome than the actual medication received.”
Psychiatrists who wish to use all the therapeutic tools at their disposal must attend to and manage patient expectations. One part of channeling a patient’s expectation is to thoroughly assess the patient’s beliefs regarding the efficacy of various treatments. If a patient’s uncle said that a certain drug is a miracle cure for anxiety, and the patient believes it to be true, then that expectation must be taken into consideration. Many patients prefer alternative treatments to conventional therapies. As long as there is no reason to think an alternative treatment will cause harm, a compromise might be reasonable. For example, if a patient with schizophrenia wants to treat her symptoms with herbal tea, the psychiatrist could say, “In addition to the tea, I recommend that you also take clozapine. The combination is likely to improve your symptoms.”3 More than anything else, the words a psychiatrist uses when recommending treatment shape the patient’s expectations. “You should be feeling a lot less anxious soon after you start taking this” has a different effect than “Try this. It may help.”