Cas.umt.edu2

How the Doctor's Words Affect the Patient's Brain
The online version of this article can be found at: can be found at:
Evaluation & the Health Professions
Additional services and information for
HOW THE DOCTOR’S
WORDS AFFECT THE
PATIENT’S BRAIN
Evaluation & the Health Professions / December 2002 Clinicians have long known that context is important in any medical treatment and that University of Turin Medical School the words and attitudes of doctors and nursescan have great impact on the patient. There isnow experimental evidence indicating that themedical context influences specific neuralsystems. The importance of the context isshown by the lesser effectiveness of hiddenadministrations of analgesics compared withopen ones. Because the placebo effect is acontext effect, its study has been useful in clar-ifying this complex issue. There are now sev-eral lines of evidence that placebo analgesiais mediated by endogenous opioids and pla-cebo motor improvement by endogenousdopamine. Moreover, a placebo treatment iscapable of affecting many brain regions indepressed patients. All these studies, takentogether, lead to a neurobiological under-standing of the events occurring in the brainduring the interaction between the therapistand his or her patient. EVALUATION & THE HEALTH PROFESSIONS, Vol. 25 No. 4, December 2002 369-386DOI: 10.1177/0163278702238051 2002 Sage Publications Evaluation & the Health Professions / December 2002 INTRODUCTION
Any medical treatment is carried out within a context. Although this statement seems quite obvious, what is not so obvious is the roleplayed by the context in the effectiveness of the therapy that is beingadministered or, in other words, what Balint (1955) called the wholeatmosphere around the treatment. The context is made up of anythingthat surrounds the patient under treatment, such as doctors, nurses,hospitals, syringes, pills, machines, and so forth (Di Blasi, Harkness,Ernst, Georgiou, & Kleijnen, 2001). Certainly, doctors and nurses rep-resent a very important component of the context because they cantransmit a lot of information to the patient through their words, atti-tudes, and behavior. Consider, for example, the following two situa-tions. First, a patient in pain lies in a bed with an intravenous line con-nected to an infusion pump. Does it make any difference whether theanalgesic treatment is started automatically by the infusion machineor by the doctor herself at the bedside? Second, a patient in pain goesto the doctor’s office for an analgesic therapy. Does it make any differ-ence whether the doctor gives the patient a painkiller and says “It maywork” or “It does work”? These are very common situations in routinemedical practice, and certainly most of us do not worry very muchabout the subtle differences between a machine-initiated and a doctor-initiated therapy in the postoperative phase or between the words “itmay” and “it does.” Nevertheless, the impact of these different situa-tions on the patient can sometimes be dramatic and can trigger mecha-nisms that have been little understood and partly neglected. Anythingaround the patient can potentially contribute to creating these differ-ences and thus to producing either a positive or a negative impact.
Many clinicians have long known this powerful effect of the con- text and, accordingly, have used the appropriate words and attitudeswith their patients. For example, Thomas (1987) found that both posi-tive and negative consultations in general practice have a tremendousimpact on patients who present with minor illness. Likewise, Kaplan,Greenfield, and Ware (1989) found that blood pressure, blood sugar,functional status, and overall health status were consistently related tospecific aspects of physician-patient communication. Although manyother studies have shown that the doctor-patient relationship plays animportant role in the outcome of illness (Bass et al., 1986; Gracely,Dubner, Deeter, & Wolskee, 1985; Greenfield, Kaplan, & Ware, 1985; Starfield et al., 1981; Stewart, 1995; Stewart, McWhinney, & Buck,1979), the underlying mechanisms are not always clear. For example,a better interaction between the doctor and the patient might lead to abetter compliance with the drug regimens (Inui, Yourtee, & William-son, 1976). However, the symbolic and emotional impact of doctorsand other aspects of medical contexts on the patient certainly has acrucial role (Brody, 1988). This is also shown by the fact that diagnos-tic tests, which have nothing to do with therapy, reduce short-term dis-ability (Sox, Margulies, & Sox, 1981). In addition, it has been shownthat psychosocial treatment has positive effects in advanced malignantdisease (Spiegel, Bloom, Kraemer, & Gottheil, 1989), although con-trasting data exist in this field (Cassileth, Lusk, Miller, Brown, &Miller, 1985).
Whereas all these studies have mainly used a clinical approach, today we are beginning to understand some of the neurobiologicalmechanisms that take part in the complex events linking context andhealth outcomes. In other words, there is now experimental evidencethat the context has an important influence on the outcome of a medi-cal treatment and that it affects the course of some pathological condi-tions, such as pain and motor disorders, through the modulation ofspecific neurochemical systems (for a review, see Benedetti &Amanzio, 1997; Benedetti & Pollo, 2001). The experimentalapproach to this complex issue has been mainly based on the study ofthe placebo effect, both in the clinical and in the laboratory setting.
THE PLACEBO EFFECT
According to Brody (2000a), the placebo effect is a change in the body, or the body-mind unit, that occurs as a result of the symbolic sig-nificance that one attributes to an event or object in the healing envi-ronment. To be more specific, a placebo is an inert substance or, ingeneral, an inert medical treatment and the placebo effect is theresponse to it. However, it is important to point out that the effect is notdue to the inertness of the treatment per se. In fact, an inert medicaltreatment is administered within a context, and it is the context thatplays the crucial role.
When we talk about the medical context, basically we are talking about the placebo effect. The terms “context effects,” “nonspecific Evaluation & the Health Professions / December 2002 effects,” and “placebo effects” can be used, at least in part, inter-changeably (Di Blasi et al., 2001). Turner, Deyo, Loeser, Von Korff,and Fordyce (1994) emphasized the importance of placebo effects inpain treatment and stressed that the interaction between caregiversand patients can be extremely influential in the therapeutic outcomes.
Likewise, Thomas (1994) pointed out that a placebo effect can also beproduced by a consultation in which no treatment is given. Thus, theclassic concept of the placebo as a phenomenon whereby patients aremade to feel better after receiving an inert treatment is too restrictive.
The broader term “context effect” is advisable to clarify that it is thecontext that influences the specific treatment (Di Blasi et al., 2001).
Although one of the simplest and most controllable contexts, at least from an experimental viewpoint, is represented by words (verbalcontext), there are plenty of contextual factors that contribute to theplacebo response: visual, auditory, olfactory, tactile, and the like—or,in other words, any clue that leads to the knowledge that a medicaltreatment is being performed (Benedetti & Amanzio, 1997; Benedetti& Pollo, 2001). A positive context can produce the reduction of asymptom (placebo effect), and a negative context can produce itsincrease (nocebo effect). Most of these aspects of the placebo effecthave been reviewed in detail recently (Guess, Kleinman, Kusek, &Engel, 2002; Harrington, 1997; White, Tursky, & Schwartz, 1985).
DOCTOR-INITIATED VERSUS
MACHINE-INITIATED THERAPY
The importance of the doctor-patient interaction is shown by the emotional impact that the anesthetist has on his or her patient (Egbert,Battit, Turndorf, & Beecher, 1963; Egbert, Battit, Welch, & Bartlett,1964). Egbert et al. (1964) found a reduction in postoperative pain inthose patients who had been informed about the course of their post-operative pain and encouraged to overcome it. Moreover, the require-ment of narcotics for these patients was much lower as compared to acontrol group. These studies are important because they compare theoutcomes of an analgesic treatment following the anesthetist’s visitwith those following no visit at all, thus emphasizing the importantrole of the doctor-patient interaction in the global experience of pain.
In more recent years, a similar approach has been used in more con- trolled conditions. To eliminate the context around a medical treat-ment, the patient was made completely unaware that a medical therapywas being carried out. To do this, drugs were administered through hid-den infusions by machines (Amanzio, Pollo, Maggi, & Benedetti,2001; Gracely, Dubner, Wolskee, & Deeter, 1983; Levine & Gordon,1984; Levine, Gordon, Smith, & Fields, 1981). It is possible to per-form a hidden infusion of a drug by means of a computer-controlledinfusion pump that is preprogrammed to deliver the drug at the desiredtime. The crucial point here is that the patient does not know that anydrug is being injected. This hidden procedure is relatively easy tocarry out in the postoperative phase, in which the patient recoversfrom surgery and is prepared with several intravenous lines for antibi-otic therapy, blood transfusion, rehydrating infusion, and the like. Thecomputer-controlled infusion pump can deliver a painkiller automati-cally, without any doctor or nurse in the room, leaving the patient com-pletely unaware that an analgesic treatment has been started.
In postoperative pain following the extraction of the third molar, Levine et al. (1981) and Levine and Gordon (1984) found that a hiddeninjection of a 6- to 8-mg dose of morphine corresponds to an openinjection of saline solution in full view of the patient (placebo). Inother words, telling the patient that a painkiller is being injected (whenit is actually a saline solution) is as potent as 6 mg to 8 mg of morphine.
Only by increasing the hidden morphine dose to 12 mg was an analge-sic effect stronger than the placebo observed. These authors con-cluded that an open injection of morphine in full view of the patient,which represents usual medical practice, is more effective than a hid-den one because in the latter, the placebo component is absent.
A careful analysis of the differences between open and hidden injections in the postoperative setting has been recently performed bymy group (Amanzio et al., 2001). We analyzed the effects of fourwidely used painkillers (buprenorphine, tramadol, ketorolac, andmetamizol) that were administered with either open or hidden injec-tions. The open injection was carried out by a doctor at the bedside,who told the patient that the injection was a powerful analgesic andthat the pain was going to subside in a few minutes. By contrast, thehidden injection of the same analgesic dose was performed by an auto-matic infusion machine that started the painkilling infusion withoutany doctor or nurse in the room. Thus, these patients were completely Evaluation & the Health Professions / December 2002 unaware that an analgesic therapy had been started. In one analysis,we found that the analgesic dose needed to reduce the pain by 50%(AD ) was much higher with hidden infusions than with open ones for all four painkillers, indicating that a hidden administration is lesseffective than an open one. In another analysis, we found that the timecourse of postsurgical pain was significantly different between openand hidden injections. In fact, during the first hour after the injection,pain ratings were much higher with a hidden injection than with anopen one.
In the same study (Amanzio et al., 2001), we also investigated the difference between open and hidden injections in the laboratory set-ting by using the experimental model of ischemic arm pain in healthyvolunteers. Just as in the clinical setting, we found that a hidden injec-tion of the non-opioid painkiller ketorolac was less effective than anopen one. Most interesting, in these controlled experimental condi-tions, we added a 10-mg dose of the opiate antagonist naloxone to anopen injection of ketorolac and found that the effect was as reduced aswith a hidden injection of ketorolac. As will be explained in detailbelow, this indicates that an open injection in full view of the patientactivates the endogenous opioid systems that enhance the effects ofthe injected painkiller. It is fundamental to remember that ketorolac isa non-opioid drug, thus it does not bind to opioid receptors. It is thewords the doctor uses at the bedside or, more generally, the medicalcontext that makes the difference and activates the endogenous opioidsystems.
The importance of these findings is twofold. First, by eliminating the context (i.e., the component that produces the patient’s perceptionof the administration of the agent) by means of a hidden administra-tion of a medical treatment, a reduction of the effectiveness of thetreatment itself occurs. Second, the effects of the context can beblocked either psychologically by means of a hidden administrationor pharmacologically through the opiate antagonist naloxone, thusindicating that the context affects the endogenous opioid systems.
“IT MAY WORK” VERSUS “IT DOES WORK”
Although the previous section shows the importance of the context and the effects of its absence, even more subtle differences exist. For example, can the uncertainty of the doctor’s words and attitudes affectthe outcome of a medical treatment? Does it make any differencewhether we tell the patients “This painkiller may work” or “Restassured, this painkiller does work”? Thomas (1987) conducted either positive or negative general prac- tice consultations in patients with different kinds of pain, cough, gid-diness, nasal congestion, and tiredness. In the positive consultations,the patients were given a firm diagnosis and therapeutic assurance. Ifno prescription was to be given, they were told that they required none,and if a prescription was to be given, they were told that the therapywould certainly make them better. In the negative consultations, nofirm assurance was given. For example, if no prescription was to begiven, the following statement was made: “I cannot be certain whatyour problem is, therefore, I will give you no treatment.” Conversely,if a prescription was to be given, the patients were told, “I am not surethat the treatment I am going to give you will have an effect.” Thetreatment was a placebo (thiamine hydrochloride). Two weeks afterconsultation there was a significant difference in recovery between thepositive and negative groups but not between the treated and untreatedgroups, thus indicating that the words the doctor used were crucial forrecovery.
Another study by Kirsch and Weixel (1988), albeit outside the clin- ical setting, shows that different verbal contexts produce different out-comes. In this study, coffee and decaffeinated coffee were adminis-tered following different verbal instructions. In one case, they weregiven according to the usual double-blind design (i.e., subjects kneweither the active substance or a placebo was being administered), andin the other case, decaffeinated coffee was deceptively presented asreal coffee. Kirsch and Weixel found that the placebo response wasstronger following the deceptive administration than the double-blindparadigm. They concluded that this was due to the fact that the double-blind administration induces less certain expectations about theoutcome.
In light of the importance of these subtle differences in the doctor’s interaction, my colleagues and I conducted a similar study in the clini-cal setting to investigate the differences between the double-blind andthe deceptive paradigm (Pollo et al., 2001). We treated several postop-erative patients with buprenorphine, on request, for 3 consecutive daysand with a basal infusion of saline solution. However, the symbolic Evaluation & the Health Professions / December 2002 meaning of this saline basal infusion varied in three different groupsof patients. The first group was told nothing (natural history or no-treatment group), the second was told that the infusion could be eithera potent analgesic or a placebo (classic double-blind administration),and the third group was told that the infusion was a potent painkiller(deceptive administration). The placebo effect of the saline basal infu-sion was measured by recording the doses of buprenorphine requestedover the 3-day treatment. It is important to stress once again that thedouble-blind group received uncertain verbal instructions (“It can beeither a placebo or a painkiller. Thus, we are not certain that the painwill subside.”), whereas the deceptive administration group receivedcertain instructions (“It is a painkiller. Thus, pain will subside soon.”).
We found a decrease in buprenorphine intake with the double-blindadministration and even more with the deceptive administration of thesaline basal infusion. The reduction of buprenorphine requests in thedouble-blind group was as large as 20.8% compared with the naturalhistory group, and the reduction in the deceptive administration groupwas even larger, reaching 33.8%. It is important to note that the timecourse of pain was the same in the three groups over the 3-day periodof treatment. Thus, the same analgesic effect was obtained with differ-ent doses of buprenorphine.
The above studies teach us that the uncertainty of verbal instruc- tions and attitudes leads to different results. Thus, as Thomas (1987)said, there is a point in being positive. Subtle differences in the verbalcontext around the patient may have a significant impact on the thera-peutic outcome.
THE APPROPRIATE WORDS ACTIVATE
THE ENDOGENOUS OPIOID SYSTEMS
To analyze the effects of the context on the patient, we need to elim- inate the specific action of a medical treatment (e.g., a drug). In otherwords, it is necessary to reproduce a context that is similar in allrespects to that of a real drug administration without, however, thespecific action of the drug itself. To do this, a classic placebo proce-dure is used in which a dummy treatment is given. The patient does notknow that a dummy therapy is being administered. He or she believesthat an effective treatment has been started. In this way, we can study the effects of the context on the patient’s brain. The experimentalinvestigation of these effects is paying off in both the clinical and thelaboratory setting.
An important step in understanding the neurobiological mecha- nisms of the placebo effect was made when Levine, Gordon, andFields (1978) showed that placebo analgesia is mediated by endoge-nous opioids. These pioneering findings have been confirmed by otherstudies (Benedetti, 1996; Grevert, Albert, & Goldstein, 1983; Levine& Gordon, 1984). Today, we know that placebo analgesia has bothopioid and non-opioid components, depending on the procedure usedto induce the placebo response (Amanzio & Benedetti, 1999). In fact,by using the experimental ischemic arm pain model, it was found thatif the placebo response is induced by means of strong expectationcues, it can be blocked by the opioid antagonist naloxone, whereas ifthe expectation cues are eliminated, it proves to be naloxone-insensitive.
The crucial point here is that the expectation cues were induced bymeans of different verbal instructions. In a first experiment, the non-opioid analgesic ketorolac was given for 2 consecutive days andreplaced with saline solution on the third day by telling the subjectsthat it was the same analgesic as the previous days (expectation cues ofanalgesia). In these conditions, naloxone partially blocked placeboanalgesia. In a second experiment, the same procedure was used, butthe subjects were told that the saline solution was an antibiotic (elimi-nation of the expectations of analgesia). In these conditions, naloxonedid not block placebo analgesia. Thus, depending on the verbal con-text, either opioid or non-opioid components can be involved.
Highly specific placebo responses can also be obtained in specific parts of the body (Benedetti, Arduino, & Amanzio, 1999a; Montgom-ery & Kirsch, 1997; Price et al., 1999). For example, it was found thatspecific verbal instructions can be aimed at directing the subject’sattention to specific body parts. In fact, if four noxious stimuli areapplied to the hands and feet and a placebo cream is applied to onehand only, pain is reduced only on the hand where the placebo creamhad been applied. Because this highly specific effect is blocked bynaloxone, these findings suggest that the placebo-activated endoge-nous opioid systems have a precise and somatotopic organization(Benedetti et al., 1999a).
Evaluation & the Health Professions / December 2002 Another line of research suggesting an important role for endoge- nous opioids in placebo analgesia comes from cholecystokinin (CCK)antagonists. In fact, on the basis of the anti-opioid action of CCK(Benedetti, 1997), it was found that CCK antagonists are capable ofpotentiating the placebo analgesic effect (Benedetti, 1996; Benedetti,Amanzio, & Maggi, 1995). Thus, the placebo analgesic responseappears to result from a balance between endogenous opioids andendogenous CCK. It is also worth remembering that the placebo-activated endogenous opioids act not only on pain mechanisms,inducing analgesia, but also on the respiratory centers, inducing respi-ratory depression, a typical side effect of opioids (Benedetti,Amanzio, Baldi, Casadio, & Maggi, 1999b; Benedetti et al., 1998).
Two additional studies confirm the role of endogenous opioids in placebo analgesia. The first study was performed by Lipman et al.
(1990) involving chronic pain patients. These authors found that thosepatients who responded to a placebo administration showed higherconcentrations of peak B endorphin in the cerebrospinal fluid com-pared with those patients who did not respond to the placebo. The sec-ond and very recent study (Petrovic, Kalso, Petersson, & Ingvar,2002) used positron emission tomography to analyze the brain regionsthat are affected by both placebo analgesia and the rapidly actingopioid agonist remifentanil. In both cases, the regional cerebral bloodflow changed in the very same areas in the anterior cingulate cortexand in the brain stem. This anatomical similarity between placeboanalgesia and remifentanil analgesia suggests that placebos activatethe same opioid receptors to which remifentanil binds.
It is important to emphasize that in all these studies on placebo analgesia, the verbal context plays a crucial role. In fact, the typicalplacebo procedure used in the studies described above is characterizedby the appropriate verbal instructions, which make the subjectsbelieve that an analgesic treatment is being performed. Although it isnot known exactly the mechanisms through which the context affectsthe endogenous opioid systems (see below), researchers can nowassert that at least some types of placebo procedures activate theendogenous opioid systems.
THE RIGHT CONTEXT TRIGGERS
THE RELEASE OF ENDOGENOUS DOPAMINE
The release of endogenous substances by placebos is a phenome- non that is not confined to the field of pain but that is also present inmotor disorders such as Parkinson’s disease. The verbal context ishere represented by verbal instructions about motor improvement. Asoccurs with pain, patients are given an inert substance (placebo) andare told that it is an anti-Parkinsonian drug producing an improvementin their motor performance. It has been shown that Parkinsonianpatients respond to placebos quite well (Goetz, Leurgans, Raman, &Stebbins, 2000; Shetty, Friedman, Kieburtz, Marshall, & Oakes,1999). A recent study shows that placebos activate endogenous dopa-mine in the nigrostriatal pathway of Parkinsonian patients—in thevery same circuit that is damaged in Parkinson’s disease (de la Fuente-Fernández et al., 2001). In particular, by using positron emissiontomography to assess the competition between endogenous dopamineand [11C]raclopride for D /D receptors, a method that allows identifi- cation of endogenous dopamine release, this study shows that place-bos trigger the release of dopamine in the striatum, in both the caudatenucleus and the putamen.
In addition, by assessing the stimulus-response curve of the subthalamic nucleus in Parkinsonian patients by means of intracranialelectrodes, we showed that different verbal contexts (verbal sugges-tions of either bad or good motor performance) alter the motorresponses following the stimulation of the subthalamic nucleus (Polloet al., 2002). All these data, taken together, indicate that a placebo pro-cedure with its verbal context is capable of inducing the release ofendogenous substances in very specific brain regions, such as thebrain stem in analgesia (Petrovic et al., 2002) or the striatum in Parkin-son’s disease (de la Fuente-Fernández et al., 2001).
THE RIGHT WORDS AFFECT BRAIN ACTIVITY
OF DEPRESSED PATIENTS
Very recently, the neurobiology of the placebo effect has also been studied in depression. Depressed patients who received a placebotreatment showed both electrical and metabolic changes in the brain.
Evaluation & the Health Professions / December 2002 In the first case, placebos induced electroencephalographic changes(cordance) in the prefrontal cortex of patients with major depression,particularly in the right hemisphere (Leuchter, Cook, Witte, Morgan,& Abrams, 2002). In the second case, changes in brain glucose metab-olism were measured by using positron emission tomography in sub-jects with unipolar depression. Placebo response was associated withmetabolic increases in the prefrontal, anterior cingulate, premotor,parietal, posterior insula, and posterior cingulate and metabolicdecreases in the subgenual cingulate, para-hippocampus, andthalamus (Mayberg et al., 2002). Interestingly, these regions were alsoaffected by the selective serotonin reuptake inhibitor fluoxetine, thussuggesting a role for serotonin in the placebo response.
THE APPROPRIATE WORDS YIELD A NOCEBO EFFECT
If the context is opposite to that producing the placebo effect, a nocebo response can be elicited. In fact, whereas the verbal instruc-tions to induce a placebo response are represented by a hopeful andtrust-inducing stimulus, the verbal context that elicits a noceboresponse is represented by a fearful and stressful stimulus (Benedetti& Amanzio, 1997; Hahn, 1985, 1997). The nocebo effect has beeninvestigated less than the placebo effect, and very little is known aboutits neurobiological mechanisms, although it is common, distressing,and costly (Barsky, Saintfort, Rogers, & Borus, 2002).
In a study performed by my group (Benedetti, Amanzio, Casadio, Oliaro, & Maggi, 1997), we obtained some results that are not easy tointerpret. In postoperative patients who reported mild pain, weinduced a nocebo effect by injecting an inert substance (saline solu-tion) and telling them that pain was going to increase in a few minutes.
We observed a straightforward nocebo effect that was blocked by add-ing a CCK antagonist, proglumide (0.5 mg to 5 mg), to the saline solu-tion. This indicates that the nocebo hyperalgesia of these patients wasmediated, at least in part, by CCK. However, the effects of proglumidewere not antagonized by naloxone, even at high doses (10 mg), show-ing that endogenous opioids were not involved. Because CCK plays arole in anxiety and the nocebo procedure itself is anxiogenic, we inter-preted these results by suggesting that proglumide acts on a CCK-dependent increase of anxiety during the nocebo procedure.
Therefore, although further research is needed to confirm these findings in the nocebo effect, CCK seems to play a pivotal role in boththe placebo (see above) (Benedetti, 1996; Benedetti et al., 1995) andnocebo effect (Benedetti et al., 1997). The knowledge of these mecha-nisms is particularly important in light of the model suggested byHahn (1985). In fact, in his anthropological analysis on the socio-cultural creation of sickness and healing, Hahn proposes a model ofthe placebo-nocebo phenomenon, in which positive-hopeful beliefsand expectations produce therapeutic effects, whereas negative-fearful beliefs and expectations produce pathological outcomes. Asstated by Hahn (1985), beliefs and expectancies sicken, kill, and heal.
Needless to say, expectations are largely generated from the context(Hahn, 1997).
TOWARD A NEUROBIOLOGY OF
THE DOCTOR-PATIENT INTERACTION
As pointed out by Rowbotham (2001), the placebo analgesic response is a physiological phenomenon and not imagined pain andplain malingering. The interaction between therapists and theirpatients triggers physiological mechanisms that have so far beenneglected. Although we now know that the context can induce therelease of endogenous opioids and dopamine and the modulation ofendogenous CCKergic systems, we have not yet understood how itacts. For example, cognitive factors could be involved, such as expec-tancies (Kirsch, 1999). In this sense, the term “meaning response” inplace of “placebo response” has been emphasized (Moerman, 2002;Moerman & Jonas, 2002). Conversely, a mechanism of classical con-ditioning could play a crucial role without any involvement of expec-tancies, meanings, and symbols (Ader, 1997). Both cognitive and con-ditioning mechanisms are probably important in different situations(Amanzio & Benedetti, 1999), and these mechanisms are not mutu-ally exclusive. Further research is needed to clarify this point.
Today, we can rely on modern diagnostic tools and many effective pharmacological and surgical treatments. Thus, it would be a mistaketo believe that placebos and bedside manners heal everything. Arecent meta-analysis study teaches us that the placebo effect is some-times small or even completely absent (Hróbjartsson & Gøtzsche, Evaluation & the Health Professions / December 2002 2001), a concept that we have long known. However, it should be rec-ognized that the humane aspect of care and the importance of the doc-tor at the bedside are often neglected in the treatment of many symp-toms, such as pain and motor disorders, which respond to placebosquite well. The recent neurobiological approach to the placebo effectdescribed throughout this article alerts us to the fact that specificneurochemical mechanisms are triggered by the doctor-patient rela-tionship. Of course, this relationship is a very complex one and cannotbe simplified in a few words. Plenty of factors are involved (Benedetti& Amanzio, 1997; Di Blasi et al., 2001) and these are very difficult toidentify. For example, a hidden injection of a drug (Amanzio et al.,2001) can eliminate only some aspects of the context (e.g., the pres-ence of the doctor), but it cannot eliminate the hospital environment,the beliefs of the patients, the room layout, and so forth. This is onlythe very beginning of a true neurobiology of the therapist-patient rela-tionship, and much research still needs to be done to identify all thecontextual factors that affect the patient’s brain. The understanding ofthese mechanisms may have important clinical and social implica-tions (Brody, 2000b), for example, in medical practice, psychother-apy, and various unconventional medical and nonmedical approaches.
It is interesting to note that the primary reason for lawsuits in the United States is not medical injury itself but the failure of communica-tion between doctors and their patients. Patients sue their physicianwhen they feel that he or she did not care or did not inform them ade-quately (Levinson, 1994). Beckman, Markakis, Suchman, andFrankel (1994) identified four types of communication problems:deserting the patient, devaluing patients’ views, delivering informa-tion poorly, and failing to understand patients’ perspectives. Thus, thesame communication skills that are capable of reducing the risk ofmalpractice lead to what has been seen throughout this article—bettertherapeutic outcomes. Perhaps patients simply need to be betterlooked after and contacted more often for optimum benefit (Wohrl &Hemmer, 2001) so that we can harness these endogenous mecha-nisms. In this sense, a better doctor-patient interaction and a few morewords from the doctor to his patient could enhance the efficacy of dif-ferent methods of pain control as well as other therapeutic interven-tions (Chaput de Saintonge & Herxheimer, 1994; Thomas, 1994;Price, 2001). One must also bear in mind that the word “doctor” can bereplaced with the more general term “healer” to make it clear that the “shaman factor” is always present in any medical treatment (Thong,1995). In other words, the placebo effect and the mechanisms underly-ing the healer-patient relationship act irrespective of whether the ther-apeutic approach comes from conventional or unconventional treat-ments. What counts, at least for some symptoms, is the context and theinteraction of patients with their healers, be they doctors, psycholo-gists, or shamans.
REFERENCES
Ader, R. (1997). The role of conditioning in pharmacotherapy. In A. Harrington (Ed.), The pla- cebo effect: An interdisciplinary exploration (pp. 138-165). Cambridge, MA: Harvard Uni-versity Press.
Amanzio, M., & Benedetti, F. (1999). Neuropharmacological dissection of placebo analgesia: Expectation-activated opioid systems versus conditioning-activated specific subsystems.
Journal of Neuroscience, 19, 484-494.
Amanzio, M., Pollo, A., Maggi, G., & Benedetti, F. (2001). Response variability to analgesics: A role for nonspecific activation of endogenous opioids. Pain, 90, 205-215.
Balint, M. (1955). The doctor, his patient, and the illness. Lancet, 1, 683-688.
Barsky, A. J., Saintfort, R., Rogers, M. P., & Borus, J. F. (2002). Nonspecific medication side effects and the nocebo phenomenon. Journal of the American Medical Association (JAMA),287, 622-627.
Bass, M. J., Buck, C., Turner, L., Dickie, G., Pratt, G., & Robinson, H. C. (1986). The physician’s actions and the outcome of illness in family practice. Journal of Family Practice, 23, 43-47.
Beckman, H. B., Markakis, K. M., Suchman, A. L., & Frankel, R. M. (1994). The doctor-patient relationship and malpractice: Lessons from plaintiff depositions. Archives of Internal Medi-cine, 154, 1365-1370.
Benedetti, F. (1996). The opposite effects of the opiate antagonist naloxone and the cholecystokinin antagonist proglumide on placebo analgesia. Pain, 64, 535-543.
Benedetti, F. (1997). Cholecystokinin type-A and type-B receptors and their modulation of opioid analgesia. News in Physiological Sciences, 12, 263-268.
Benedetti, F., & Amanzio, M. (1997). The neurobiology of placebo: From endogenous opioids to cholecystokinin. Progress in Neurobiology, 52, 109-125.
Benedetti, F., Amanzio, M., Baldi, S., Casadio, C., Cavallo, A., Mancuso, M., et al. (1998). The specific effects of prior opioid exposure on placebo analgesia and placebo respiratorydepression. Pain, 75, 313-319.
Benedetti, F., Amanzio, M., Baldi, S., Casadio, C., & Maggi, G. (1999b). Inducing placebo respiratory depressant responses in humans via opioid receptors. European Journal of Neu-roscience, 11, 625-631.
Benedetti, F., Amanzio, M., Casadio, C., Oliaro, A., & Maggi, G. (1997). Blockade of nocebo hyperalgesia by the cholecystokinin antagonist proglumide. Pain, 71, 135-140.
Benedetti, F., Amanzio, M., & Maggi, G. (1995). Potentiation of placebo analgesia by proglumide. Lancet, 346, 1231.
Benedetti, F., Arduino, C., & Amanzio, M. (1999a). Somatotopic activation of opioid systems by target-directed expectations of analgesia. Journal of Neuroscience, 19, 3639-3648.
Evaluation & the Health Professions / December 2002 Benedetti, F., & Pollo, A. (2001). The pharmacology of placebos. International Journal of Pain Medicine and Palliative Care, 1, 42-48.
Brody, H. (1988). The symbolic power of the modern personal physician: The placebo response under challenge. Journal of Drug Issues, 29, 149-161.
Brody, H. (2000a). The placebo response. New York: HarperCollins.
Brody, H. (2000b). The placebo response: Recent research and implications for family medicine.
Journal of Family Practice, 49, 649-654.
Cassileth, B. R., Lusk, E. J., Miller, D. S., Brown, L. L., & Miller, C. (1985). Psychosocial corre- lates of survival in advanced malignant disease? New England Journal of Medicine, 312,1551-1555.
Chaput de Saintonge, D. M., & Herxheimer, A. (1994). Harnessing placebo effects in health care. Lancet, 344, 995-998.
de la Fuente-Fernández, R., Ruth, T. J., Sossi, V., Schulzer, M., Calne, D. B., & Stoessl, A. J.
(2001). Expectation and dopamine release: Mechanism of the placebo effect in Parkinson’sdisease. Science, 293, 1164-1166.
Di Blasi, Z., Harkness, E., Ernst, E., Georgiou, A., & Kleijnen, J. (2001). Influence of context effects on health outcomes: A systematic review. Lancet, 357, 757-762.
Egbert, L. D., Battit, G. E., Turndorf, H., & Beecher, H. K. (1963). The value of the preoperative visit by an anesthetist. JAMA, 185, 553-555.
Egbert, L. D., Battit, G. E., Welch, C. E., & Bartlett, M. K. (1964). Reduction of postoperative pain by encouragement and instruction of patients. New England Journal of Medicine, 270,825-827.
Goetz, C. G., Leurgans, S., Raman, R., & Stebbins, G. T. (2000). Objective changes in motor function during placebo treatment in Parkinson’s disease. Neurology, 54, 710-714.
Gracely, R. H., Dubner, R., Deeter, W. R., & Wolskee, P. J. (1985). Clinicians’expectations influ- ence placebo analgesia. Lancet, 1, 43.
Gracely, R. H., Dubner, R., Wolskee, P. J., & Deeter, W. R. (1983). Placebo and naloxone can alter postsurgical pain by separate mechanisms. Nature, 306, 264-265.
Greenfield, S., Kaplan, S., & Ware, J. E. (1985). Expanding patient involvement in care. Annals of Internal Medicine, 102, 520-528.
Grevert, P., Albert, L. H., & Goldstein, A. (1983). Partial antagonism of placebo analgesia by naloxone. Pain, 16, 129-143.
Guess, H. A., Kleinman, A., Kusek, J. W., & Engel, L. W. (Eds.). (2002). The science of the pla- cebo: Toward an interdisciplinary research agenda. London: British Medical JournalBooks.
Hahn, R. A. (1985). A sociocultural model of illness and healing. In L. White, B. Tursky, & G. E.
Schwartz (Eds.), Placebo: Theory, research, and mechanisms (pp. 167-195). New York:Guilford.
Hahn, R. A. (1997). The nocebo phenomenon: Scope and foundations. In A. Harrington (Ed.), The placebo effect: An interdisciplinary exploration (pp. 56-76). Cambridge, MA: HarvardUniversity Press.
Harrington, A. (Ed.). (1997). The placebo effect: An interdisciplinary exploration. Cambridge, Hróbjartsson, A., & Gøtzsche, P. C. (2001). Is the placebo powerless? New England Journal of Medicine, 344, 1594-1602.
Inui, T. S., Yourtee, E. L., & Williamson, J. W. (1976). Improved outcomes in hypertension after physician tutorials. Annals of Internal Medicine, 84, 646-651.
Kaplan, S. H., Greenfield, S., & Ware, J. E., Jr. (1989). Assessing the effects of physician-patient interactions on the outcomes of chronic disease. Medical Care, 27(Suppl. 3), S110-S127.
Kirsch, I. (Ed.). (1999). How expectancies shape experience. Washington, DC: American Psy- Kirsch, I., & Weixel, L. J. (1988). Double-blind versus deceptive administration of a placebo.
Behavioral Neuroscience, 102, 319-323.
Leuchter, A. F., Cook, I. A., Witte, E. A., Morgan, M., & Abrams, M. (2002). Changes in brain function of depressed subjects during treatment with placebo. American Journal of Psychia-try, 159, 122-129.
Levine, J. D., & Gordon, N. C. (1984). Influence of the method of drug administration on analge- sic response. Nature, 312, 755-756.
Levine, J. D., Gordon, N. C., & Fields, H. L. (1978). The mechanisms of placebo analgesia. Lan- Levine, J. D., Gordon, N. C., Smith, R., & Fields, H. L. (1981). Analgesic responses to morphine and placebo in individuals with postoperative pain. Pain, 10, 379-389.
Levinson, W. (1994). Physician-patient communication: A key to malpractice prevention.
Lipman, J. J., Miller, B. E., Mays, K. S., Miller, M. N. North, W. C., & Byrne, W. L. (1990). Peak B endorphin concentration in cerebrospinal fluid: Reduced in chronic pain patients andincreased during the placebo response. Psychopharmacology, 102, 112-116.
Mayberg, H. S., Silva, A. J., Brannan, S. K., Tekell, J. L., Mahurin, R. K., McGinnis, S., et al.
(2002). The functional neuroanatomy of the placebo effect. American Journal of Psychiatry,159, 728-737.
Moerman, D. E. (2002). Meaningful dimensions of medical care. In H. A. Guess, A. Kleinman, J. W. Kusek, & L. W. Engel (Eds.), The science of the placebo: Toward an interdisciplinaryresearch agenda. London: BMJ Books.
Moerman, D. E., & Jonas, W. B. (2002). Deconstructing the placebo effect and finding the mean- ing response. Annals of Internal Medicine, 136, 471-476.
Montgomery, G. H., & Kirsch, I. (1997). Classical conditioning and the placebo effect. Pain, 72, Petrovic, P., Kalso, E., Petersson, K. M., & Ingvar, M. (2002). Placebo and opioid analgesia— Imaging a shared neuronal network. Science, 295, 1737-1740.
Pollo, A., Amanzio, M., Arslanian, A., Casadio, C., Maggi, G., & Benedetti, F. (2001). Response expectancies in placebo analgesia and their clinical relevance. Pain, 93, 77-84.
Pollo, A., Torre, E., Lopiano, L., Rizzone, M., Lanotte, M., Cavanna, et al. (2002). Expectation modulates the response to subthalamic nucleus stimulation in Parkinsonian patients.
NeuroReport, 13, 1383-1386.
Price, D. D., Milling, L. S., Kirsch, I., Duff, A., Montgomery, G. H., & Nicholls, S. S. (1999). An analysis of factors that contribute to the magnitude of placebo analgesia in an experimentalparadigm. Pain, 83, 147-156.
Price, D. D. (2001). Assessing placebo effects without placebo groups: An untapped possibility? Rowbotham, D. J. (2001). Endogenous opioids, placebo response, and pain. Lancet, 357, 1901- Shetty, N., Friedman, J. H., Kieburtz, K., Marshall, F. J., & Oakes, D. (1999). The placebo response in Parkinson’s disease. Parkinson study group. Clinical Neuropharmacology, 22,207-212.
Sox, H. C., Margulies, I., & Sox, C. H. (1981). Psychologically mediated effects of diagnostic tests. Annals of Internal Medicine, 95, 680-685.
Spiegel, D., Bloom, J. R., Kraemer, H. C., & Gottheil, E. (1989). Effect of psychosocial treat- ment on survival of patients with metastatic breast cancer. Lancet, 2, 888-891.
Starfield, B., Wray, C., Hess, K., Gross, R., Birk, P. S., & D’Lugoff, B. C. (1981). The influence of patient-practitioner agreement on outcome of care. American Journal of Public Health,71, 127-132.
Evaluation & the Health Professions / December 2002 Stewart, M. A. (1995). Effective physician-patient communication and health outcomes: A review. Canadian Medical Association Journal, 152, 1423-1433.
Stewart, M. A., McWhinney, I. R., & Buck, C. W. (1979). The doctor-patient relationship and its effect upon outcome. Journal of Royal College of General Practice, 29, 77-82.
Thomas, K. B. (1987). General practice consultations: Is there any point in being positive? Brit- ish Medical Journal, 294, 1200-1202.
Thomas, K. B. (1994). The placebo in general practice. Lancet, 344, 1066-1067.
Thong, Y. H. (1995). The shaman factor. World Health Forum, 16, 238-240.
Turner, J. A., Deyo, R. A., Loeser, J. D., Von Korff, M., & Fordyce, W. E. (1994). The importance of placebo effects in pain treatment and research. JAMA, 271, 1609-1614.
White, L., Tursky, B., & Schwartz, G. E. (Eds.). (1985). Placebo: Theory, research, and mecha- Wohrl, S., & Hemmer, W. (2001). Placebos and pain management. Lancet, 358, 1101-1102.

Source: http://cas.umt.edu/liberal/documents/BenedettiDoctorWords.pdf