Ya Gotta Believe — Expectation and Outcome

Abstract & Commentary

By Russell H. Greenfield, MD, Editor

Synopsis: In a seminal study on the interaction between mind and body, effects of a continuous infusion of a potent opioid on pain sensation were drastically altered as a result of expectation of effectiveness. The study’s results bring to the fore the idea that a patient’s belief in a given drug therapy has a significant impact on its clinical effectiveness.

Source #1: Bingel U, et al. The effect of treatment expectation on drug efficacy: Imaging the analgesic benefit of the opioid remifentanil. Sci Transl Med 2011;3:70ra14.

Practitioners and patients alike find comfort in knowing that pharmaceuticals possess agreed-upon physiologic mechanisms of action, such that therapeutic responses are considered reproducible and essentially linear; but there has long been a nagging question regarding whether an individual’s expectations regarding a given therapy might influence treatment outcome. Prior data suggest that verbal cues that influence patient expectation actually may alter a drug’s therapeutic efficacy.

The authors of this creative, within-subject study investigated the impact of expectation of effectiveness on analgesia provided by the potent m-opioid receptor agonist remifentanil, which has a rapid onset of action and short elimination half-life (about 10 minutes). Pain perception and analgesia were chosen because the neurobiological mechanisms of both are well described.

Healthy volunteers (n = 22; 7 female; all right-handed; mean age, 28 years; range, 21 to 40 years) were recruited with the understanding that the study aimed to investigate the brain mechanisms responsible for differing levels of response to opioids among individuals. They were told that remifentanil relieves pain quickly when infused intravenously, but can worsen pain when the infusion is stopped.

The study comprised two sessions separated by at least 24 hours, one introductory and one main experimental session that included functional magnetic resonance imaging (fMRI). During the introductory session participants underwent what was called an "expectation manipulation-conditioning procedure" to induce positive and negative treatment expectations. They were exposed to painful heat through a probe applied to the right calf in four sequences: two with an IV infusion of saline, during which participants were asked to describe this baseline experience of pain in the absence of treatment; one with IV infusion of remifentanil, during which participants were told they should expect significant pain relief, but during which, and unbeknownst to the subjects, the heat of the painful stimulus was turned down; and one where the IV infusion of remifentanil was stopped, during which subjects were told to expect a worsening of pain, while in fact the temperature of the heat probe was increased, again unbeknownst to the participants.

The main experimental session consisted of four runs of identical thermal stimulation to the right mid-calf, each including 10 thermal pain stimuli lasting approximately 10 minutes (adjusted to produce a pain intensity rating of 70 on a VAS, where 0 corresponds to "no pain" and 100 to "unbearable pain"). After a baseline run performed with a saline infusion only, the remifentanil infusion was started without the subjects being told, so that in the second run the analgesic effect of 30 minutes of a remifentanil infusion could be assessed without any treatment expectation. To distract from potential psychotropic effects experienced with rising CNS concentrations of remifentanil, a 15-minute structural brain scan was performed and participants were told that the study might cause "vibrations that may evoke a sensation of slight disorientation in some participants." To begin the third or "positive expectancy" run, subjects were told that the infusion "would now be started by the anesthetist." When that infusion was completed, participants were told "the infusion will now be stopped to investigate the possible increase in pain after ceasing the opioid infusion." In reality, the infusion was continued throughout the fourth or "negative expectancy" run.

Whole brain fMRI including the brainstem was used to record brain activity and investigate the neural mechanisms by which expectancy might modulate the efficacy of pharmacological treatment. Activity levels within several regions of the brain previously have been reported to consistently correlate with intensity of nociceptive inputs and resultant pain perception, and were taken as surrogate markers of analgesia.

Other study measures included pain intensity rating, anxiety levels pre- and post-treatment, and overall unpleasantness of the painful stimuli. To minimize potential habituation or sensitization during the course of the experiment, the site of thermal stimulation along the right mid-calf was changed slightly after each of the four runs.

The reported results are compelling: the hidden application of remifentanil without treatment expectancy significantly reduced pain intensity ratings from 66 + 2 during baseline saline infusion to 55 + 3 [t(21) = 5.1, P < 0.001]. Positive expectancy significantly enhanced analgesia, as pain ratings further decreased to 39 + 3 [t(21) = 6.4, P < 0.001]. Negative expectancy, when the subjects had been led to believe that the drug was stopped, resulted in a considerable increase in pain intensity from 39 + 3 (positive expectancy run) to 64 + 3 (negative expectancy run) [t(21) = 8.5, P < 0.001]; thus, negative expectancy essentially abolished the analgesic effect of remifentanil, as pain intensity under negative expectancy did not differ from pain intensity during baseline saline infusion [t(21) = 0.68, P = 0.5]. Results for pain unpleasantness ratings showed a similar pattern. The analgesic benefit from positive expectancy was negatively correlated with anxiety ratings obtained at the start of the respective run (r = -0.55, P < 0.01), indicating that participants who were less anxious showed a greater analgesic benefit of positive expectancy.

The reported subjective effects were substantiated by significant changes in the neural activity in brain regions involved with the coding of pain intensity. Positive expectancy effects were associated with activity in the endogenous pain modulatory system, and negative expectancy effects with activity in the hippocampus.

The authors conclude in their model of pain and analgesia that positive treatment expectancies literally double the analgesic benefit of remifentanil, while negative treatment expectation interferes with the analgesic potential of remifentanil to the point of completely negating it. These effects were paralleled by significant changes in neural responses in core brain regions that are involved in the intensity coding of pain. The researchers close by stating that a patient's expectation of a drug's effect critically influences its therapeutic efficacy, and that regulatory brain mechanisms differ as a function of expectancy.

Duke University researchers studied patients undergoing diagnostic coronary angiography from 1992-1996 found to have clinically significant disease (75% stenosis of 1 coronary artery) in the prospective Mediators of Social Support Study (MOSS). A total of 3737 qualifying patients enrolled, with basic mortality analyses conducted on only 2818 patients (75% of the study population) due to missing information. Coronary artery bypass surgery (CABG) was performed on 1277 of the participants (45.3%) at some point during the follow-up period, with 1156 (41.0%) undergoing percutaneous transluminal coronary angioplasty (PTCA). Of these, both procedures were performed on 396 (14.1% of the sample). The remaining 781 patients (27.7%) were medically treated throughout the course of the study.

The Duke database electronic record was the source of information about comorbidities and relevant health history. Coronary disease severity was controlled in the analyses by including the number of coronary arteries with at least 75% stenosis (1-3), left ventricular ejection fraction, and a 6-level variable indicating the presence and severity of congestive heart failure. These measures were obtained during the baseline angiographic examination. Demographic variables were covaried to control for potential economic and social confounding factors and included education, ethnicity, and marital status.

Study instruments included the Expectations for Coping Scale (ECS) used to determine the patient's expectations regarding future lifestyle and future cardiac prognosis (half of the items were worded so that agreement implied positive expectations, and half were worded in the other direction); the Duke Activity Status Index (DASI) to assess the patient's ability to perform a range of physical activities; the Interpersonal Support Evaluation List (ISEL), designed to measure perceived availability of social support; and the Center for Epidemiologic Studies Depression Scale (CES-D), a measure of the frequency of various depressive symptoms experienced during the previous week.

Follow-up of the patients was conducted at 6 months and 12 months after catheterization and annually thereafter. Telephone interviews were conducted with patients 1 year after hospitalization. DASI scores were obtained for 2392 patients who also had DASI scores at baseline, representing 85% of those in the mortality analyses. Follow-up times for surviving patients averaged 14.6 years and ranged up to 17 years. As of December 2008, 1637 of the 2818 patients had died, with 885 of the deaths classified by an independent committee as being secondary to cardiac causes.

ECS scores indicating positive expectations were associated with reduced mortality risk; unadjusted data showed a mortality rate of 28.8 deaths per 100 patients during the 10 years after baseline for those in the highest quartile of expectations compared with 56.9 deaths per 100 for those in the lowest quartile. For a difference equivalent to an interquartile range of expectations, the hazard ratio (HR) for total mortality was 0.76 (95% confidence interval [CI], 0.71-0.82) and 0.76 (95% CI, 0.69-0.83) for cardiovascular mortality. The HRs were 0.83 (95% CI, 0.76-0.91) and 0.79 (95% CI, 0.70-0.89) with further adjustments for demographic and psychosocial covariates. Similar associations (P < 0.001) were observed for functional status.

The authors conclude that in their large cohort of patients with known coronary artery disease, those patients who had more favorable expectations about their likelihood of recovery and return to a normal lifestyle had better long-term survival, as well as better functional status after their hospitalization.

These two articles are presented jointly because their findings provide the backdrop for broader consideration of the mind–body continuum in clinical care. Add to the current context the findings of Kaptchuk et al (PLoS ONE 2010;5:e15591) recently reviewed in Alternative Medicine Alert (See February 2011) that showed accessing therapeutic benefits from the placebo effect does not require deception. Considered together these results comprise a formidable basis for the concept that health care practitioners must strive to enlist both a patient's trust in their doctor's clinical acumen as well as comfort with recommended care, invasive or otherwise, to optimize the chances of successful treatment outcome.

Belief is a complex variable encompassing a wide variety of factors including at least personal experience, the chronic nature of specific health problems, interpersonal relationships (including those with health care providers), mood, coping skills, and the clinical environment. In light of this, on an Olympic scale one might consider the degree of difficulty enlisting a patient's positive expectations to be high, but each of these issues can be addressed by using something almost all healers have at their disposal – their compassionate selves. This should not be considered some New Age concept — it is, in fact, the basis for the healing relationship between practitioner and provider. What this new scientific evidence brings to light is that practitioners probably do not use the healing relationship inherent in the medical encounter to its greatest effect. According to the results of the aforementioned studies, that could translate into treatment failure even when the biochemical and physiologic rationales behind that treatment are secure.

Time is in short supply in present-day practitioner–patient interactions, but when employed efficiently, that time can be used to describe the mechanism of action of a drug or explain a procedure, and at least to underscore our belief as practitioners in the recommended path forward. This may be construed as manipulation, but if a recommendation is made to a patient regarding care, it is reasonable to assume the practitioner making the recommendation believes it could help — sharing that belief, literally couching the intervention in positive terms, may enhance therapeutic efficacy.

To be sure there are some concerns with the two studies reviewed here, not the least of which being the small sample size in the former, and the missing pieces of data in the second trial; regardless, the methodology is otherwise sound and the conclusions persuasive. Expectations and related beliefs have an impact not only on the effect of therapies, but potentially on long-term survival, too.

Medical treatment is most always considered a matter of physiology, but the role of psychology in any cogent treatment plan may soon share center stage. Recall the children's story "Dumbo," where the elephant could not fly without the "magic feather" until the day he lost the feather, only to find he was able to fly by himself all along. Belief is powerful medicine, an underutilized medicine. As these studies show, to do the best by the people who willingly share their life stories with us as practitioners, we must do more than focus solely on the physical — we must engage them in positive expectation wherever appropriate. Not doing so places our patients at a significant disadvantage.