T H E PLA CEBO - Spanda Foundationspanda.org/SpandaJournal_II,1.pdf · · 2011-02-23quarterly of...

SPANDA JOURNALq u a r t e r l y o f t h e s p a n d a f o u n d a t i o n

O N L I N E E D I T I O N | W W W . S P A N D A . O R G / P U B L I C A T I O N S . H T M L

SPANDA.ORG

II,1/2011

S P A N D A

T H E

P L A C E B OE F F E C T

http://www.spanda.org/publications.html

Come, do not stop at the gate: Life and Death, thetwo guardian angels, will transmute into theGuardians of the Threshold, and let you in. There

lies Beauty, Truth, and Good-ness, the attributes of Love,and the primeval promoter oflife who subtly affects thewhole being. I will be gratefulif you could spread the goodnews: “Come Death, I’m wait-ing for your call.” Realisticmodels should be integrated inthe vision, they are needed aslong as we partake of thisdimension. What needs to bereviewed after all? The inexplic-able that not even the finestdegree of angels could discloseto itself? Be my guest, I requestyou to step in: therein no abyssfrom where to surface, indeednothing is there, neither timenor space, nothing, Nothing atAll. Still the dualistic vision iscreeping in this momentarydimension in which we all live.Placebo.Ut placet in mente dei? Anapophatic utterance uncoversthe world: get rid of all fears andstatements of joy. No groundaround, no glamour, breathedeeply and plunge into the

Unknown: “Hallo, who are you?” You or me?!?Definitely a day to celebrate and rejoice, hallelu-yah– a dynamic shared field of awakened awareness inthe higher collective consciousness is at play.

The body is a device to calculate the astronomy of the spirit.Look through that astrolabe and become oceanic.

RUMI, Mathnawi.

O N O T B E S U R P R I S E D B Y

a pale aestheticalglance, beauty is stillone of the best veils to

look throughout this world.Beauty encompasses all, all andeverything, like Truth andGoodness. Once freed fromtime we wander to the accom-plished deed immune to all ill-ness. Having accomplished itsmortal function, the physicalbody returns to its forming ele-mentals whilst the soul progressesto its further mansion, the houseof misericordiæ endowed withprophethood. Indeed, there isno ‘time’ if there is no ‘be’, andsurely there is no ‘die’ if there isno Thee.Taxi downtown to uncoverwhat really matters: there aremore invisible things than visibleones. It is not what we see, thatis of importance, but what we donot see that is the real substance.There is only one possibleequation valid for all planes ofthe being, where each particle is shaped by thesublime fashioner of the wor[l]d, an exquisitesense of the phrase, a lace vaulting in a vacuum.

S P A N D A J O U R N A L I I , 1 /2 0 1 1 | t h e p l a c e b o e f f e c t

E D I T O R I A L | SSAAHHLLAANN MMOOMMOO

The Perplexity ofthe Unseen 2

FFAABBRRIIZZIIOO BBEENNEEDDEETTTTII

The Many Placebo Effects 5

EEDDZZAARRDD EERRNNSSTT

The Ethical Implications ofUsing Placebo in Clinical Practice 9

CCHHRRIISSTTOOPPEERR JJ.. BBEEEEDDIIEE

All in the Mind? Pain, Placebo, and ErgogenicEffect of Caffeine inSport Performance 1 2

KKUURRTT GGRRAAYY ~~DDAANNIIEELL MM.. WWEEGGNNEERR

The Sting of PotentialPain 2 0

JJOOHHNN NN.. WWIILLLLIIAAMMSS

The Ethics of Placebo-Controlled Trialsin Developing Countries toPrevent Mother-to-ChildTransmission of HIV 2 3

A B S T R A C T S 2 9

g | I N T H I S I S S U ED

S PA N D A J O U R N A L

THE PERPLEXITY

OF THE UNSEEN

SPANDA.ORGSPANDA.ORG

II,1/2011

EDITORIAL

Once the top-down stream of consciousness hasbeen granted, the U conversion of the energy occursand all blessings are gathered into the humanedimension. A new chapter is being chiselled into thesoul: long seated marks are eradicated in a purifica-tion process to get rid of all rights – only the physi-cal death will temporarily procrastinate freedom.The human soul is bound to transit through therealm of quantity to progress further; it needs toexperience its ‘physical’ body as an essential step toreach up anew – only because we dwell on the rela-tive realm of time and space this process is perceivedas bottom-up. A conversion of energies akin to theone taking place at puberty or conversely at old age,

when the life energy inverts its polarity, changesdirection – the superabundance of a force inevitablyproduces its opposite. On a different level, this par-allels the principle of equilibrium in the naturalworld in which any extreme is opposed by the sys-tem to restore balance. “Cold things warm, warmthings cool, wet things dry and parched things getwet,” enantiodromia, an old acquaintance of mineused to say. The ensuing plans of the being areplainly passages from one stage to the next, at timesignited by rites in the human endeavour. In con-sciousness (cit), it is a shift, a leap into a new state.In Reality, in absence of time, the past, the presentand the future are condensed into the ‘now’, thatmeans that the shift already happened, is takingplace and will occur, altogether at the same time,its awareness depending from which state of ourindividual consciousness we are witnessing it.Placebo. Here is not the unrestrained surge of theunconsciousness into consciousness at work, as thevillain maintains, rather their simultaneously merg-ing into oneness. To contemporarily perceive thetwo fluxes is certainly vital, but being aware of theaction while displaying the ‘activity’ without inter-fering in the process is definitely a step ahead.When the performance of an action is one with its

intention (imma), if there is no meaningful iatus ortemporal diastema between intention and execu-tion, that action is effortlessly attuned with theflow of the dharma, is one with it. In the occasion,even the freewill vanishes, as it is one and the samewith the cosmic will. The intention is a focusedattitude toward a well defined goal, a pro-tension toits realization by virtue of its own entelechy toadvance further and overcome all limits; in otherterms, the perennial drive of life pro-tensed towardsits fulfilment. The kernel is that the two energiesand their actions – anode and cathode, or ruh-illofiand ruh al-kuds or whatever other combination ofopposites it may be – are really simultaneous, and

actually the same; if observed through the binarymanner to perceive the world of the thinking mind,they appear to be two, for duality houses in time,but in the dimensionless gathering place of the soulthey are one, just one.In the midst of all this, by sightseeing randomly inthe groove of the path, we gladly report of a fewlandmarks along its landscape, certainly not of theinexplicable goal. We all are mixed blood, we areall bastards, it depends from which stance we arelooking at ourselves. To the body, it may look likehealth, to the spirit, as spiritual health, in between,as a spiritual-material wealth. Placebo: somethingpleases and soothes the wounds and recovers them totheir original state. The way in which the synapsescommunicate in our brain counterparts our socialnetworking. This parallelism of planes may beapplied to the whole of the manifestation, as allentities are connected through their dimensionlesscentre by the axis mundi crossing in every possibledirection. Devoid of time and space, no directionis at bay, everything happens contemporary in thepresent, every and nowhere. The thinking mind isunable to grasp this hierophany, this darshan,unless it affords to be one with its own workingprocess and lifts its hold on it, only then intuition

S P A N D A J O U R N A L I I , 1 /2 0 1 1 | t h e p l a c e b o e f f e c t | 3

emerges as an insight in the heightened conscious-ness. Seen from the realm of quantity, intuitionlooks faster and finer than thought: the manipulatorof matter. Placebo.Harmony is a palintropos in a reflexive tension, likethe bow and the lyre. To be in a reflexive tension, tobe self-reflexive is to be re-flexed in oneself. Theentire matter lies on the capacity to be utterly empty,aloof from the whole lot, yet gladly allow the energyto flow by its own course, according to its own pace,synchronic to the rhythm of the body: the inner andouter are here focused in a spiritual-material alliance.Certainly consciousness is not made up of matter, itis much finer, deeper and higher than matter, is anassembly of relations in a reflexive tension. Indeed,what really matters in all human affairs are the qualityof the relations, not its su[o]bjects (persona), for, theformer – being immaterial – will last; while the latterwill perish. Yesterday I would have liked to give youa rose, not solely a rose but a flowering rose in yourpose, in your improbable chest. But you were notpresent, emptiness around, not even the bristle of aleaf, or the cry of an angel, nor the glance of your eyesI once thought blue. The whole universe couldn’tcompete with that hue: a triumph could not havebeen better, certainly no better than you. How longshould I stand at your dazzling face before joiningwith you? Placebo. Pinocchio identified himself withhis dream and became human. Who for God’s‘shake’ (quiver, spanda) would prefer a placeboinstead of a real shot? “The spiritual life is knowl-edge in the time of trial”: a charming thought inactual fact, which certainly needs a skilled cryptogra-pher to endorse it as a whole. Nothing can existwithout movement, yet the ultimate movementtakes place in consciousness where time has no grip:an endless cycle of expansion and contraction, ofinternalization and externalization of consciousnessitself, relating to the most elevated plane (citananda)of the manifestation.At times it appears as the direction is nowhere tobe found: no longer faiths nor religions to cling to,no more containers of a long lost content, just thereality that inspires and sustains them is here. Nosound came out of their tongues but their hassle toslaughter each other: fundamentalist to their roots,who lost connection with the original life. Beassured, everything passes and changes, they toosoon will be gone. Let’s swing the ladder to theencounter of the two seas. Prospero or Papageno?Even though these bastards deserve a diamond ascrown, for now I can only offer them a leadedbasin and a blade of red iron lore – transmute…transmute… Thought and self-awareness are thetwo parts of the same unfolding process of doing,the non-discursive state of consciousness can hardly,if ever, be conveyed into words. A good intentionis not enough; action must follow. Placebo.Theory, poiesis, praxis… theà horào, yes, we all sawthe goddess, but global thought and local action

are now needed, at least in ourselves to start with.Where has it gone that ability to keep together andjointly strive for a clear direction? Where has Politicsvanished? Still busy with old credentials, withmodels, concepts and visions of old, from ten toone, and then bottom up. A thousand of existenceago ‘ten’ was deprived of its wholeness and became‘one’. If 1 is equal to 0, if uniqueness in wholenessis attained, if one is the whole, then the tensorialmembrane between the two worlds within con-sciousness self-shapes itself according to its owninborn input, and the world is informed by matter.To transmute into its further stage, the soul linksspirit to matter into a whole.The whole manifestation is nothing but how weperceive our own projection on it, the projection ofour particular ‘self ’ in its reflexive mode. A para-dox, a koan, no doubt, contrary to any opinion andbeyond any reasonable doubt. Indeed the paradoxhas always been a powerful device to explore thereality: by unveiling it, the opposites are reconciled:inside and outside become paler and all boundariesare loosened and lost. The ‘apparent’ movementbetween the two polarities makes consciousnessspark and it gives rise to the whole. An itinerariumanimi in which an inclusive reform of the I, and ofthe ability of the soul to manifest the reality is mostin demand. The middle world between spirit andmatter, the mundus imaginalis, links ta physica to tameta, in its ‘imaginal’ geography – not a phantasys –the tensorial membrane – and the soul – receivesthe ‘impressions’ of the spiritual world and reverberatesthem into the world of matter as mundane outcome –let us not forget that to show is not to perform.Not a novel vision for sure, but one former to the dis-section of spirit and matter in time, a compoundwhich still grants to placebo its effect, that, accordingto the followers of the divided reality, shouldn’tindeed be there. Spear me to detail further, wecould disperse a few. Enjoy the issue. ©

a b


a b

We live our lives, for ever taking leave.RAINER MARIA RILKE, Duineser Elegien.

a b

Fabrizio Benedetti, MD is Professor ofPhysiology and Neuroscience at theUniversity of Turin Medical Schooland at the National Institute ofNeuroscience, Italy. He has beennominated member of The Academyof Europe and of the EuropeanDana Alliance for the Brain. He haswritten the books Placebo Effects(Oxford 2008) and The Patient’sBrain (Oxford 2010).Email: [email protected].

I N T R O D U C T I O N

P L A C E B O E F F E C T I S A

psychobiological phe-nomenon occurring inthe patient’s brain fol-lowing the administra-

tion of an inert substance, or ofa sham physical treatment suchas sham surgery, along withverbal suggestions (or any other cue) of clinicalbenefit (Price et al, 2008). Therefore, the effect thatfollows the administration of a placebo cannot beattributable to the inert substance alone, for salinesolutions or sugar pills will never acquire therapeuticproperties. Instead, the effect is due to the psychoso-cial context that surrounds the inert substance andthe patient. We now know that there is not a singleplacebo effect, but many, with different mecha-nisms and in different diseases, systems, and thera-peutic interventions (Benedetti, 2008b; Enck et al,2008). In other words, different processes may be atwork in the patient’s brain in different conditions.Sometimes it is anxiety that is modulated, at someother times reward mechanisms are involved, andin some other circumstances different types oflearning, or even genetic variants, may take place inplacebo responsiveness. In this sense, the placeboeffect is a melting pot of neuroscientific conceptsand ideas, ranging from anxiety and reward mecha-nisms to Pavlovian conditioning and social learn-ing, and from neurogenetics and neurophysiology

to clinical practice and neuroethics. The termsplacebo effect and placebo response are often usedas synonymous, thus both terms will be used here.

E X P E C T A T I O N S

Most of the research on placeboshas focused on expectations asthe main factor involved inplacebo responsiveness. In gen-eral, expectation is aimed atpreparing the body to anticipatean event in order to better copewith it, and as such offers a clearevolutionary advantage (Kirsch,1999). There are several mecha-nisms through which expecta-tion of a future event may affectdifferent physiological functions.For example, the expectation ofa negative outcome is aimed atanticipating a possible threat,thus increasing anxiety, whereasthe expectation of a forthcomingpositive outcome may reduceanxiety and/or activate the neu-ronal networks of reward mech-anisms (Price et al, 2008).Indeed, anxiety has been foundto be reduced after placebo

administration in some studies. If one expects adistressing symptom to subside shortly, anxietytends to decrease. In brain imaging studies reducedactivation of anxiety-related areas during a placeboresponse can be observed (Petrovic et al, 2005). Thebest evidence that anxiety takes part in placeboresponses is shown by the nocebo effect, which isopposite to the placebo effect (Benedetti et al,2007). In order to induce a nocebo effect, an inertsubstance is administered along with negative ver-bal suggestions of clinical worsening, e.g. painincrease. Overall, expectations of a negative out-come, such as pain increase, may result in theamplification of pain, and several brain regions,like the anterior cingulate cortex, the prefrontalcortex, the insula, and the hippocampus have beenfound to be activated during the anticipation ofpain in a variety of studies. Nocebo hyperalgesiahas been studied in some detail also from a pharma-cological perspective, in order to identify possibleneurotransmitters that are involved in anxiety-induced pain increase. By using an anxiogenic noceboprocedure, it was shown that cholecystokinin (CCK)


T H E M A N Y P L A C E B O E F F E C T SF A B R I Z I O B E N E D E T T I

“ The miserable

have no other

medicine

but only

hope. ”WW II LL LL II AA MM SS HH AA KK EE SS PP EE AA RR EE

g | O V E R V I E W

A

plays a crucial role. In fact, the pharmacologicalblockade of CCK receptors prevents the occurrenceof nocebo responses.Not only can expectations of future events modu-late anxiety, but they may also induce physiologicalchanges through reward mechanisms. These mech-anisms are mediated by specific neuronal circuitslinking cognitive, emotional, and motor responses,and are traditionally studied in the context of thepursuit of natural (e.g., food), monetary, and drugrewards. There is compelling experimental evidencethat the mesolimbic dopaminergic system, the mainreward network, may be activated in some circum-stances when a subject expects clinical improvement

after placebo administration. This was observed inParkinson’s disease (de la Fuente-Fernandez et al2001), depression (Mayberg et al 2002) and pain(Scott et al 2007). In the case of the placebo response,the reward can be conceptualized as the clinicalimprovement.Several neuropharmacological and neuroimagingstudies investigated neither anxiety nor rewardmechanisms, making it impossible to state specifi-cally whether the observed placebo responses wereattributable to a reduction in anxiety or to the acti-vation of the reward circuitry. In all probability,here too anxiety and/or reward mechanisms play arole, depending on the experimental condition. Inany case, these studies have provided evidence thata complex neural network is involved during theplacebo analgesic and the nocebo hyperalgesicresponses (for a detailed review, see Zubieta andStohler 2009; Tracey 2010).

L E A R N I N G

Learning is another mechanism that is central toplacebo responsiveness. Subjects who suffer from apainful condition, such as headache, and who regu-larly consume aspirin, can associate the shape, colorand taste of the pill to pain decrease. After repeatedassociations, if they are given a sugar pill resemblingaspirin, they will experience pain decrease. Not onlycan shape, color and taste of pills be associated toclinical improvement, but countless other stimuli aswell, such as hospitals, diagnostic and therapeuticequipments, and medical personnel features. Themechanism that underlies this effect is condition-ing, whereby a conditioned (neutral) stimulus, e.g.

the color and shape of a pill, can become effective ininducing the reduction of a symptom if repeatedlyassociated to an unconditioned stimulus, i.e. theactive principle contained in the pill. Conditioned immune responses can be obtainedboth in animals and in humans. For example,Goebel et al (2002) showed that behavioral condition-ing of immunosuppression is possible in humans.Repeated associations between cyclosporine A and aflavored drink induced conditioned immunosup-pression in healthy male volunteers, in which theflavored drink alone produced a suppression of theimmune functions. In the endocrine system, similareffects can be found. The hypoglycemic effects ofinsulin can be conditioned by pairing insulin with aconditioned stimulus in animals and in humans,and several other hormones, such as the growthhormone and cortisol, can be conditioned as well(Benedetti, 2008a, 2008b).


More complex forms of learning can be involvedin placebo responsiveness, whereby different cog-nitive factors play a crucial role. For example,social learning is a form of learning, whereby indi-viduals in a society learn from one another byobservation and imitation. Placebo effects mayinvolve social learning as well. In other words, themere observation of others responding to analgesicsmay lead to robust placebo analgesic responses(Colloca and Benedetti 2009).

G E N E T I C S

A central issue in placebo research is whether anindividual in whom a placebo works possesses oneor more specific characteristics, which can reliablyidentify him a priori as a “placebo responder”,with important implications for both clinical trialsdesign and personalized therapy optimization.Results have so far been rather inconclusive, withdemographic, psychosocial, personality, andbehavioral variables all proposed to play a role, butall inconsistently present across different trials.Recently, however, some genetic variants have beenfound that are particularly responsive to placebotreatment. For example, there is some experimentalevidence that some genetic variants affect placeboresponses in psychiatric disorders, such as social anx-iety (Furmark et al, 2008) and depression (Leuchter etal, 2009). In social anxiety, it was found that thereduced stress-related activity in the amygdalawhich accompanied the placebo response could beobserved only in subjects who were homozygousfor the long allele of the 5-HTTLPR or the G vari-ant of the TPH2 G-703T polymorphism, but not incarriers of short or T alleles. In depression, sub-jects with monoamine oxidase A G/T polymor-phisms (rs6323) coding for the highest activityform of the enzyme (G or G/G) had a significantlylower magnitude of placebo response than thosewith other genotypes.

N O P R E F R O N T A L C O N T R O L ,

N O P L A C E B O R E S P O N S E

Benedetti et al (2006) studied Alzheimer patients atthe initial stage of the disease and after one year, inorder to see whether the placebo component of thetherapy (an ever-present part of the drug effectwhich makes the overall outcome greater than thatproduced purely by the intrinsic principle) wasaffected by the disease. The placebo component ofthe analgesic therapy was correlated with both cog-nitive status, as assessed by means of the FrontalAssessment Battery (FAB) test, and functional con-nectivity among different brain regions, as assessedby means of electroencephalographic connectivityanalysis. It was found that Alzheimer’s patientswith reduced FAB scores showed reduced placebocomponent of an analgesic treatment. In addition,the disruption of the placebo component occurred

just when reduced connectivity of the prefrontallobes with the rest of the brain was present. Theloss of these placebo-related mechanisms reducedthe overall effectiveness of the treatment, andindeed a dose increase was necessary to make upfor this loss in order to produce adequate analgesia.According to this view, the impairment of pre-frontal connectivity would reduce the communica-tion between the prefrontal lobes and the rest ofthe brain, so that no placebo and expectationmechanisms would be triggered. In recent yearsthis notion has been supported by the deactivationof the prefrontal cortex in the experimental setting.On the basis of previous experiments on the block-ade of placebo analgesia by the opioid antagonistnaloxone, Eippert et al (2009) conducted a study toinvestigate the location of naloxone action in thebrain. By combining naloxone administration withfunctional magnetic resonance imaging (fMRI),these authors found that naloxone reduced placeboeffects as well as placebo-induced responses in thedorsolateral prefrontal cortex and the rostral anteriorcingulate cortex. Therefore, as it occurs for prefrontaldegeneration in Alzheimer’s disease, placebo anal-gesic responses are disrupted by the pharmacologicalblockade of prefrontal opioidergic functioning in theexperimental setting.Prefrontal degeneration in Alzheimer’s disease andpharmacological blockade of prefrontal opioidergictransmission are not the only conditions in whichplacebo responses are disrupted. The inactivation ofthe prefrontal cortex, particularly the dorsolateralprefrontal cortex, by means of repetitive transcranialmagnetic stimulation (rTMS) has the same effect(Krummenacher et al, 2010). Therefore, the inacti-vation of prefrontal regions by transcranial magneticstimulation has the same effects as those inducedby pharmacological blockade or prefrontal degener-ation in Alzheimer’s disease. On the basis of allthese studies, a normal functioning of prefrontalareas appears to be critical for placebo responsive-ness. In the presence of a loss of prefrontal control,we also witness a loss of placebo response.

F U T U R E R E S E A R C H D I R E C T I O N S

Despite the recent explosion of neurobiologicalplacebo research and the recent findings that help usbetter understand both human biology and clinicalpractice, several issues need further clarificationsand many questions still remain unanswered. First ofall we need to know where, when, and how placeboswork across different diseases and therapeutic inter-ventions. Second, a better understanding of the con-tribution of different mechanisms, such as expecta-tion, anxiety, reward, learning, genetics, in differenttypes of placebo responses is in order, and thiswould surely help identify the social, psychologicaland neurobiological determinants of the differentplacebo effects. Third, we need to understand why


some subjects respond to placebos, whereas othersubjects do not, a critical point that is likely to beclarified by pursuing further research into bothlearning and genetic mechanisms. ©

R E F E R E N C E S

AMANZIO M., BENEDETTI F. (1999). “Neuropharmaco-logical dissection of placebo analgesia: expectation-activated opioid systems versus conditioning-activatedspecific subsystems” in J Neurosci 19: 484-494.

BENEDETTI F. (2008a). “Mechanisms of placebo andplacebo–related effects across diseases and treat-ments”, in AnnuRev PharmacolToxicol 48: 33-60.

——, (2008b). Placeboeffects: Understand-ing the mechanismsin health and dis-ease (Oxford:Oxford UniversityPress).

——, ARDUINO C.,COSTA S.,VIGHETTI S.,TARENZI L.,RAINERO I et al(2006a). “Loss ofexpectation-relat-ed mechanisms inAlzheimer’s dis-ease makes anal-gesic therapies lesseffective”, in Pain121: 133–144.

——, MAGGI G., LOP-IANO L., COLLOCA

L. (2007). “Whenwords are painful– Unraveling themechanisms of thenocebo effect”, inNeuroscience 147:260-271.

COLLOCA L.,BENEDETTI F.(2009). “Placeboanalgesia inducedby social observa-tional learning”, in Pain 144: 28-34.

DE LA FUENTE-FERNANDEZ R., RUTH TJ., SOSSI V.,SCHULZER M., CALNE DB., STOESSL AJ. (2001).“Expectation and dopamine release: mechanisms ofthe placebo effect in Parkinson’s disease”, in Science293: 1164-1166.

EIPPERT F., BINGEL U., SCHOELL ED., YACUBIAN J.,KLINGER R., LORENZ J., et al (2009). “Activation ofthe opioidergic descending pain control systemunderlies placebo analgesia” in Neuron 63: 533-43.

ENCK P., BENEDETTI F., SCHEDLOWSKI M. (2008). “Newinsights into the placebo and nocebo responses”, inNeuron 59: 195-206.

FURMARK T., APPEL L., HENNINGSSON S., AHS F., FARIA

V., LINNMAN C., et al (2008). “A link between sero-tonin-related gene polymorphisms, amygdala activi-ty, and placebo-induced relief from social anxiety”,in J Neurosci 28: 13066-13074.

GOEBEL MU., TREBST AE., STEINER J., XIE YF., EXTON

MS., FREDE S., et al (2002). “Behavioural condition-ing of immunosuppression is possible in humans”,in FASEB J 16: 1869-1873.

KIRSCH I. (1999). How expectancies shape experience,American Psychological Association.

KRUMMENACHER P., CANDIA V., FOLKERS G., SCHED-LOWSKI M., SCHÖNBÄCHLER G. (2010). “Prefrontalcortex modulates placebo analgesia”, in Pain 148:368-374.

LEUCHTER AF.,MCCRACKEN JT.,HUNTER AM., COOK

IA., ALPERT JE. (2009).“Monoamine oxidase aand catechol-o-methyl-transferase functionalpolymorphisms and theplacebo response inmajor depressive disor-der”, in J Clin Psychophar-macol 29: 372-377.MAYBERG HS., SILVA

JA., BRANNAN SK.,TEKELL JL.. MAHURIN

RK., MCGINNIS S., etal (2002). “The func-tional neuroanatomy ofthe placebo”, in Am JPsychiatry 159: 728-737.PETROVIC P., DIETRICH

T., FRANSSON P.,ANDERSSON J., CARLS-SON K. (2005). “Place-bo in emotional pro-cessing–induced expec-tations of anxiety reliefactivate a generalizedmodulatory network”,in Neuron 46: 957-969.PRICE DD., FINNISS

DG., BENEDETTI F.(2008). “A comprehen-sive review of theplacebo effect: recentadvances and currentthought”, in Annu Rev

Psychol 59: 565-590.SCOTT DJ., STOHLER CS., EGNATUK CM., WANG H.,

KOEPPE RA., ZUBIETA JK. (2007). “Individual differ-ences in reward responding explainplacebo–induced expectations and effects”, in Neu-ron 55: 325-336.

TRACEY I. (2010). “Getting the pain you expect: mecha-nisms of placebo, nocebo and reappraisal effects inhumans”, in Nature Med 16: 1277-1283.

a b


Edzard Ernst (1948, Wiesbaden, Ger-many) a British citizen since 1999, isthe first Professor of ComplementaryMedicine in the world. He was bornand trained in Germany and beganhis medical career at a homeopathichospital in Munich. In 1993, he lefthis chair in Physical Medicine andRehabilitation (PMR) at the Universityof Vienna to set up the department ofComplementary Medicine at theUniversity of Exeter & Plymouth,UK. He became director of Comple-mentary Medicine of the PeninsulaMedical School (PMS) in 2002. He isthe first occupant of the Laing chairin Complementary Medicine.Ernst is the editor-in-chief of twomedical journals, Perfusion andFocus on Alternative and Comple-mentary Therapies. Ernst once con-tributed a regular column to theGuardian newspaper, frequentlyreviewing news stories about comple-mentary medicine from an evidence-based perspective.Since his research began on alternative modalities, Ernsthas become “the scourge of alternative medicine” for pub-lishing critical research that exposes methods that lack docu-mentation of efficacy. Email: [email protected].

1 ~ P R E V A L E N C E O F U S E

E C E N T S U RV E Y D ATA F R O M S E V E R A L C O U N T R I E S

indicate that many clinicians use placeboin their clinical practice 1,2,3,4,5,6,7,8, often on aregular basis. Reasons for doing thisinclude:

~ to follow the wish of the patient;~ to avoid conflict;~ to calm the patient;~ to comply with patients’ demand for receiv-ing some sort of medications;~ to prevent the patient complaining.

According to these data, prescribing placebos isclearly considered to be ethically justifiable by alarge proportion of clinicians.

The subject of placebo is notoriously complex, anddiscussions of this topic are often too abstract toprovide practical guidance. It might therefore behelpful to focus on a concrete example: homeopathy.The best evidence available to date strongly sug-gests that highly dilute homeopathic remedies are

pure placebos e.g.9,10. That is tosay, they are devoid of specifictherapeutic effects. Yet it is unde-niable that many physicians andother healthcare professionalsregularly use homeopathy. Manyof them must be aware of theevidence and employ it as a“benign placebo” e.g.11. The ques-tion thus is whether this preva-lent behaviour can be ethicallyjustified.

2 ~ A R G U M E N T S I N

F A V O U R

Physicians using placebos, suchas homeopathic medicines, inclinical practice might claim doso because they want to helptheir patients via a beneficialplacebo effect4-6. Prescribing ahighly diluted homeopathicremedy is, of course, unlikely to

have direct adverse effects. Many patients expect toreceive a prescription when consulting their doctor,and many believe in homeopathy, i.e. they areunaware of the evidence9,10 and convinced that home-opathic remedies generate specific effects. Issuing aprescription for a homeopathic remedy shouldtherefore be ethically commendable, particularly incases where no drug therapy is indicated 12.

3 ~ A R G U M E N T S A G A I N S T

At first glance this line of argument seems compelling.However, several counter-arguments emerge, oncewe scrutinize the issue in more depth.

3 . 1 ~ M E R E C O N V E N I E N C E

A critical analysis of the most frequent reasons forplacebo-use (see above) is revealing. The desire ofphysicians to help their patients does not seem to bea prominent factor. On the contrary, the reasons pro-vided demonstrate that placebo-use is not altruisticallymotivated but prompted mainly by convenience. It is


“ It takes

a wise

doctor

to know

when

not to

prescribe. ”GG RR AA CC II ÁÁ NN

g | A LT E R N AT I V E S

R

THE ETHICAL IMPLICATIONSO F U S I N G P L A C E B O I N C L I N I C A L P R A C T I C E

E D Z A R D E R N S T

quite simply easier for a doctor to write a prescriptionfor a placebo than to discuss with the patient that nosuch prescription is required or no effective treatmentis available. The hope of avoiding conflict and thedesire to prevent a complaint are, of course, under-standable motives, but they have little to do with thephysician’s duty to help patients. Physicians’ conve-nience, it would seem, is not a reason that is ground-ed in medical ethics.

3 . 2 ~ D E C E I T A N D I N F O R M D C O N S E N T

The prescription of placebos in clinical practice isusually based on what might be called paternalisticdeception. Only a small proportion of clinicians

would tell their patients the true nature of thisprescription 6. Without pretending that the home-opathic placebo is effective beyond placebo, doc-tors cannot expect their patients to experience aclinically relevant placebo-effect. Truthfully tellinga patient that the prescribed remedy contains nopharmacologically active ingredient would notgenerate expectation of benefit. Thus little or noplacebo-response would result from telling thetruth. However, doctors would “violate the princi-ple of respect for patient autonomy and contra-vene the legal and ethical requirement to obtaininformed consent”13, if they decided to deceivepatients, even if it were “for their own good”.

3 . 3 ~ T H E R A P E U T I C R E L A T I O N S H I P

Not telling the truth undermines trust which is anessential ingredient of any therapeutic relationship.A good therapeutic relationship can improve both

the adherence to treatment and the clinical out-come14-17. Placebo-prescriptions cannot be used assubstitutes for acknowledging uncertainty, exploringpatients’ concerns, beliefs and preferences, consider-ing non-pharmacological therapies and invitingshared decision making18,19. These actions would beconsistent with the physician’s ethical responsibili-ties of non-maleficence, beneficence and respect forpatients without, at the same time, depending onthe deception of issuing a placebo.

3 . 4 ~ P O T E N T I A L F O R H A R M

Clinicians using placebos in routine practice shouldconsider at least two important risks of this approach.

Firstly, the placebo might be administered insteadof a treatment that has specific therapeutic effects.Secondly, the placebo might not be free of adverseeffects. Homeopathic remedies, for instance areunlikely to cause adverse effects, but other types ofplacebos clearly do. For example, physicians oftenuse impure placebos, i.e. treatments that areunder-dosed or not indicated for a particular condi-tion. A frequently prescribed impure placebo is anantibiotic for a non-bacterial infection4,6. Suchimpure placebos can cause serious, direct harmthrough their pharmacological actions. Antibioticshave a range of direct adverse effects and, canharm the population at large through the develop-ment of bacterial resistance.

3 . 5 ~ M E D I C A L I Z A T I O N

Prescribing placebos for the types of self-limitingcomplaint which placebos are most commonly

S P A N D A J O U R N A L I I , 1 /2 0 1 1 | t h e p l a c e b o e f f e c t | 1 0

prescribed for will almost inevitably result in themedicalization of common states of reduced well-being. Such conditions are best treated by re-assur-ance and honesty. Prescribing a placebo in suchsituations would simply reinforce the belief thatthere is ‘a pill for every ill’. This would be neitherhonest nor helpful. Braillon therefore stressed that“a placebo is a dangerous tool” because it encour-ages “disease-mongering” 20.

3 . 6 ~ P L A C E B O S A R E N O T N E E D E D

F O R F O R G E N E R A T I N G A

P L A C E B O - R E S P O N S E

In the vast majority of situations, effective treat-ments with proven effects beyond placebo are avail-able – they may only be symptomatic rather thancurative but they are helpful nevertheless. If, insuch cases, we want to increase patient benefitthrough maximising placebo-effects, we shouldrealize that the sympathetic and empathetic admin-istration of any therapy would result in a placebo-effect, in addition to the specific effect of the pre-scribed therapy, particularly if expectancy is maxi-mized 21. Thus prescribing a homeopathic remedyto generate a placebo-effect is usually not onlyunnecessary, it would also deprive the patient ofthe specific effect of that treatment. This wouldclearly not be ethical.

4 ~ C O N C L U S I O N

The prescription of placebos in clinical practicecreates an ethical dilemma. On the one hand, clin-icians might want to ease the suffering of theirpatients. On the other hand, this approach violatesimportant ethical principles. The solution is toanalyse this behaviour (self ) critically. Once we dothis, we are likely to find that much of such place-bo-prescribing serves the convenience of cliniciansrather than the welfare of the patient. Where pos-sible, it should be avoided and replaced with moreethical and professional behaviour. ©

R E F E R E N C E S

1 BERTHELOT JM., MAUGARS Y., ABGRALL M., PROST

A. (2001). “Interindividual variations in beliefsabout the placebo effect: a study in 300 rheumatol-ogy inpatients and 100 nurses”, in Joint Bone Spine68(1): 65-70.

2 NORHEIM AJ., FONNEBO V. (2002). “Attitudes to thecontribution of placebo in acupuncture - A survey”,in Complementar Ther Med; 10(4): 202-209.

3 LIM EC., SEET RC. (2003). “Attitudes of medical stu-dents to placebo therapy”, in Intern Med J 37(3):156-160.

4 HROBJARTSSON A., NORUP M. (2003). “The use ofplacebo interventions in medical practice - a nation-al questionnaire survey of Danish clinicians”, inEvaluation & the Health Professions 26(2): 153-165.

5 NITZAN U., LICHTENBERG P. (2004). “Questionnairesurvey on use of placebo”, in BMJ 329: 944-946.

6 SHERMAN R, HICKNER J. (2008). “Academic physi-cians use placebos in clinical practice and believe inthe mind-body connection”, in J Gen Intern Med23(1): 7-10.

7 TILBURT JC., EMANUEL EJ., KAPTCHUK TJ., CURLIN

FA., MILLER FG., (2008). “Prescribing ‘placebo treat-ments’: results of national survey of US internists andrheumatologists”, in BMJ 337: a1938.

8 FÄSSLER M., GNÄDINGER M., ROSEMANN T., BILLER-ANDORNO N. (2009). “Use of placebo interventionsamong Swiss primary care providers”, in BMC HealthServices Research 9: 144-152.

9 ERNST E. (2010). “Homeopathy: what does the ‘best’evidence tell us?”, in MJA 192: 458-460.

10 SHANG A., HUWILER-MUNTENER K., NARTEY L., JUNI

P., DORIG S., STERNE JA. ET AL., (2005). “Are the clin-ical effects of homoeopathy placebo effects? Compar-ative study of placebo-controlled trials of homoeopa-thy and allopathy”, in Lancet 366: 726-732.

11 DU Y., KNOPF H. (2009). “Paediatric homeopathy inGermany: results of the German Health Interviewand Examination Survey for Children and Adoles-cents (KiGGS)”, in Pharmacoepidemiol and DrugSafety 18: 370-379.

12 JÜTTE R., HOPPE J-D., SCIRBA P. (2010). “Stellung-nahme des Wissenschaftlichen Beirats der Bun-desärztekammer ‘Placebo in der Medizin’”, inDeutsches Ärzteblatt 107(28-29): 1417-1421.

13 MILLER FG., COLLOCA L. (2009). “The legitimacy ofplacebo treatments in clinical practice: evidence andethics”, in Am J Bioethics 9(12): 39-47.

14 STEWART MA. (1995). “Effective physician-patientcommunication and health outcomes: A review”, inCMAJ 152: 1423-1433.

15 ONG LM., DE HAES JC., HOOS AM. (1995). “Doctor-patient communication: a review of the literature”,in Soc Sci Med 40: 903-918.

16 DIMATTEO MR. (2004). “‘Variations in patients’adherence to medical recommendations: a quantita-tive review of 50 years of research”, in Med Care 42:200-209.

17 HASKARD ZOLNIEREK KB., ROBIN DIMATTEO M.(2009). “Physician communication and patient adher-ence to treatment. A meta-analysis”, in Med Care47(8): 826-234.

18 SCHENKER Y., FERNANDEZ A., LO B. (2009). “Place-bo prescriptions are missed opportunities for doc-tor-patient communication”, in Am J Bioethics 9(12):48-54.

19 HROBJARTSSON A. (2008). “Clinical placebo inter-ventions are unethical, unnecessary, and unprofes-sional”, in J Clin Ethics 19: 66-69.

20 BRAILLON A. (2009). “Placebo is far from benign: it isdisease-mongering”, in Am J Bioethics 9(12): 36-38.

21 CROW R., GAGE H., HAMPSON S., HART J., KIMBER

A., THOMAS H. (1999). “The role of expectancies inthe placebo effect and their use in the delivery ofhealth care: A systematic review”, in Health TechnolAssess 3(3): 1-97.

a b


Senior Lecturer/Programme Director,Canterbury Christ Church University,Canterbury, UK. Email: [email protected]


HE ERGOGENIC EFFECTS OF

caffeine on sports per-formance are well doc-umented1. Althoughthese effects have been

observed in a wide variety ofsports, they are generally moreevident in endurance and sus-tained high-intensity closed skillsports than in interactive andstop-go sports 2. The ergogeniceffects of caffeine can be obtainedat doses at or below the dailyintake of normal populations3,and below levels representinghealth risks. Caffeine is not cur-rently proscribed by the World Anti-Doping Agency.Several mechanisms for the ergogenic effects of caf-feine have been proposed, including enhanced fatoxidation, sympathetic nervous system enhance-ment, reduction in central fatigue, attenuation ofneuromuscular conduction block, and potentiationof muscular force output for given input4. Thesemechanisms are not consistently supported byresearch evidence, eg, Graham et al concluded thatthere is very little evidence to support the fat oxida-tion theory1, whilst Tarnopolsky suggested there isno evidence for caffeine-induced changes in periph-eral nerve conduction or neuromuscular transmis-sion4. The lack of consistent support for any onephysiologic mechanism, and the fact that an increas-ing number of studies in sport and elsewhere havedemonstrated significant and substantial placeboeffects of caffeine5,10, suggests that psychologic factorsmight play a significant role in caffeine effects. Perhaps the obvious psychologic factor to consideris pain. The proposal that caffeine ingestionreduces pain in performance is supported by both

experimental findings and the anecdotal reports ofathletes5,11. Furthermore, caffeine-induced painreduction could explain, in part, both biologic andplacebo effects of caffeine on performance. Such apain hypothesis is supported by a number of fac-tors. Firstly, caffeine has well documented hypoal-

gesic or pain-reducing proper-ties12, secondly, pain is a limit-ing factor on performance13,thirdly, pain is highly placeboresponsive14, and lastly, thesports in which ergogeniceffects of caffeine have beenobserved tend to be those inwhich athletes experience sub-stantial levels of sustained painwithout the relief of regulardisruption of activity associatedwith stop-go or interactivesports. Given that pain is alsohighly socially modifiable andcontext-dependent14, a painhypothesis might explain theinterindividual variability incaffeine response observed insport research15. The presentpaper aims to examine thepotential influence of psychoso-cial factors in the ergogeniceffect of caffeine in the context

of caffeine-induced pain reduction. The article willreview selected findings demonstrating placeboeffects of caffeine on sports performance, hypoal-gesic effects of caffeine on sports performance, andplacebo effects on pain in experimental and clinicalmedicine. It will also briefly discuss implicationsfor research.

P L A C E B O E F F E C T S O F C A F F E I N E O N

S P O R T P E R F O R M A N C E

Placebo effects of caffeine on sports performancehave been reported in three recent experimentalstudies. Beedie et al examined the possibility of adose-response relationship to placebos presented as“zero”, “low”, and “high” dose caffeine among sevenwell-trained competitive cyclists5. Measures werepower, heart rate, oxygen uptake, and blood lactate.Following habituation and baseline trials, subjectswere informed that, over three experimental trials,they would receive a placebo, or 4.5 mg/kg or 9.0mg/kg caffeine double-blind and randomlyassigned. However, a biologically inert placebo was


A L L I N T H E M I N D ?PAIN, PLACEBO EFFECT, AND ERGOGENIC EFFECT OF CAFFEINE IN SPORT PERFORMANCE

C H R I S T O P H E R J . B E E D I E

T“ A man

who is

‘of sound mind’

is one who

keeps the

inner madman

under lock

and key. ”PP AA UU LL VV AA LL ÉÉ RR YY

g | C H A L L E N G E S

mailto:[email protected]

administered in all experimental conditions. Postex-perimental baseline trials were also conducted. Adose-response relationship was evident in experimen-tal trials, with subjects producing 1.4% (range, -4.6%-1.9%) less power than baseline when they believedthey had ingested a placebo, 1.3% (-1.4%-4.1%) morepower than at baseline when they believed they hadingested 4.5 mg/kg caffeine, and 3.1% (0.4%-6.7%)more power than at baseline when they believedthey had ingested 9.0 mg/kg caffeine. Of furtherinterest was the fact that no substantial differencesin any measured physiologic variables between base-line and experimental conditions were observed.Follow-up interviews with each subject indicatedthat five subjects believed that they had experienceda placebo effect, proposing mechanisms such as painreduction, fatigue resistance, changes in strategy, andreduced arousal. Pollo et al investigated the effects of a caffeineplacebo on quadriceps muscle performance andperceived fatigue6. Forty-four recreationally activemales were divided into four groups, ie, two con-trol and two placebo (n=11 each). In the first exper-iment, a placebo was deceptively administered,with the suggestion that it was a high dose of caf-feine. This resulted in a significant increase in meanmuscle work (11.8%-16.1%, P< 0.01) but no per-ceived decrease in muscle fatigue (P> 0.05). In thesecond experiment, placebo caffeine administra-tion was accompanied by a conditioning proce-dure whereby the weight to be lifted was surrepti-tiously reduced.The load was then restored to theoriginal weight and placebo caffeine administeredagain. Compared with the first experiment, theplacebo effect was larger, with a significant increasein muscle work (22.1%-23.5%, P< 0.01) and adecrease in perceived muscle fatigue (-7.8-10.1, P<0.01). The authors suggested that their findingsindicated a central mechanism of topdown modu-lation of the global performance of muscles byplacebo and underscore the role of learning in theplacebo response. Foad et al used the balanced placebo design with 14well-trained competitive cyclists in a study examin-ing the effects of caffeine and placebo on 40 km lab-oratory cycling performance7. Subjects performedtwo 40 km time trials in each of four experimentalconditions (informed caffeine/received caffeine,informed no-treatment/received caffeine, informedcaffeine/received placebo, and informed no-treat-ment/received no treatment). Measures were power,oxygen uptake, blood lactate, and heart rate. Theauthors reported a very likely beneficial main effect onmean power of receiving caffeine (3.5% ± 2.0%), and apossibly beneficial main effect of being informed ofcaffeine (0.7% ± 1.4%). A substantial interactionbetween belief and pharmacology (2.6% ± 3.3%)indicated that caffeine exerted a greater effect onperformance when subjects were informed that theyhad not ingested it, while belief exerted a greater

influence on performance in the absence of caffeine.A possibly harmful negative placebo (nocebo) effectrelative to baseline was present when subjects werecorrectly informed that they had ingested no caf-feine (-1.9% ± 2.2%). No substantial changes relativeto baseline were observed in mean heart rate,although clear and substantial increases in bloodlactate were evident following the receipt of caf-feine. The within-subject CV for power in deceptiveconditions at 2.8% was 1.7 times larger than the CVwhen subjects were truthfully informed that theywere receiving caffeine, indicating the possibility ofsome disparity between internal sensations andinstructions amongst some subjects. The authorssuggested that their data supported the ergogenicefficacy of caffeine, but noted that in the absence ofcaffeine, the negative effect on performance of nega-tive expectation was somewhat more substantialthan the positive effect of positive expectation (afinding that could inflate effect sizes in placebo-con-trolled studies). Evidence for placebo effects associated with caf-feine has also been provided elsewhere 8-10. It is evi-dent from these findings that whilst caffeine inges-tion often exerts an influence on behavior, so toocan beliefs about caffeine, and about whether ornot caffeine has been ingested. These findings begseveral questions as to the likely ergogenic mecha-nisms of caffeine and suggest a significant psycho-logic contribution.

H Y P O A L G E S I C E F F E C T S O F C A F F E I N E

I N S P O R T S P E R F O R M A N C E

As suggested above, perhaps the most likely psycho-logic contribution would be pain reduction. Thereis a large body of evidence attesting to the hypoal-gesic properties of caffeine in both medicine16 andphysical activity5,12,17,18. Consistent with this are datademonstrating that subjects’ ratings of perceivedexertion are lower when given caffeine19. Pain is ahighly complex biopsychologic phenomenon, andeven a brief overview of theories and potentialmechanisms is well beyond the scope of this paper.It is reasonable to suggest that whilst pain is certainlynot a necessary condition of sports performance,athletes in endurance and sustained short-durationsports routinely experience pain in competition.Many will experience pain with little respite for asubstantial part, if not for the entirety, of their event(as stated above, it is in such sports that caffeine’sergogenic efficacy has been most widely demonstrat-ed, lending further weight to the pain hypothesis).Gliottoni et al state that there is an expanding bodyof evidence that acute exercise is a natural stimulusthat might transiently, safely, and reliably producemuscle pain12, whilst Cook et al suggest that notonly are descriptors used to describe pain duringexercise similar to those that have been used tocharacterize clinical pain conditions20, but that pain


ratings during exercise as measured by the shortform of the McGill Pain Questionnaire are nearlyone standard deviation above the mean scores asso-ciated with other laboratory methods of inducingpain21. Research has suggested that ingestion of caf-feine significantly reduces the pain response duringperformance12,17,18,22,23. O’Connor et al reported thatcaffeine ingestion has a dose-response effect onreducing leg muscle pain during exercise22. Motl et alreported that leg muscle pain ratings were signifi-cantly and moderately reduced after a high dose (10mg/kg body weight) of caffeine18. Maridakis et alreported that caffeine produced a large and statisti-cally significant hypoalgesia effect during maximalquadriceps contraction23. Gliottoni and Motl reportedthat caffeine administration resulted in a large reduc-tion in leg muscle pain intensity ratings17. Gliottoniet al reported that caffeine ingestion is associatedwith a moderate hypoalgesic effect during high-intensity cycling12. Theoretically, any interventionthat reduces perceived exertion or pain should alsoincrease the perceived headroom for effort, andtherefore has the potential to enhance performance.A hypoalgesic effect driven by caffeine could bedirect, ie, the action of caffeine attenuates the per-ception of pain. Indirectly, caffeine actions couldinfluence other biologic processes that attenuate thepain response. The effect could also result from bothdirect and indirect processes simultaneously.Whichever way, and significantly for the present dis-cussion, the large body of research in medicine andpsychology demonstrating substantial placebo effectson pain could shed some light on the placebo effectsrelated to caffeine and sports performance.

P S Y C H O L O G I C A L M O D U L A T I O N O F P A I N

Pain is highly susceptible to social and psychologicmodulation14, and perhaps because of this, over thelast few years, pain has become one of the mostfruitful areas of research into mind/body interac-tion. In short, evidence from such research demon-strates that expectation of pain relief can modifythe subsequent effectiveness of administered sub-stances, be they active analgesics, such as mor-phine, or inactive placebos. These effects can beboth hypoalgesic and hyperalgesic. Several complexdesigns have been used to elucidate this phenome-non, ranging from covert manipulation of experi-mental pain stimuli, to direct comparison of theeffects of the hidden/deceptive administration ofbiologically active treatments with the overt admin-istration of biologically inactive substances. Studiesusing such designs are informative in demonstrat-ing the degree to which pain can be modified bypsychosocial processes, such as conditioning andexpectation. On that basis, they are of interest tothose investigating or using interventions in whichpain reduction might be a factor in their efficacy.Some illustrative examples of such studies arebriefly described below.

Montgomery and Kirsch exposed subjects to exper-imental pain at baseline, and in subsequent trialssurreptitiously reduced the pain stimulus whilst aplacebo analgesic cream was administered (the lat-ter process designed to lead subjects to believe thatthe cream had reduced the pain)24. Subjects werethen split into two groups, the first was correctlyinformed about the deception, and the second wasnot informed. On re-exposure to the pain at base-line level, subjects who had been correctly informedof the deception experienced no pain relief whenthe placebo analgesia cream was applied, whilstthose in the second group reported substantiallylower pain. The authors also assessed subjects’ expec-tation of analgesia, and reported that this accountedfor 49% of the observed effects. The authors con-cluded that an analgesic response can be condi-tioned, but that the conditioned response might beeither reversed or suppressed by correct informationor negative expectation respectively. Levine et al administered active painkillers covertlyand placebo painkillers openly to two groups ofsubjects following dental surgery25,26. They reportedthat the overt injection of a saline placebo describedas morphine was as effective as a covert injection ofmorphine. Similarly, Benedetti et al compared theopen administration of five different painkillers withthe hidden and automated administration of thesame drugs 27,29. The authors reported that in hiddenadministration conditions the time taken for postop-erative pain to diminish by 50% was greatly increasedfor all drugs compared with open administration.These findings suggest that anticipation of analgesiais a factor in perceived analgesia. Pollo et al treated postoperative patients with apainkiller on request for three consecutive days, aswell as with a saline placebo (patients were given theintravenous saline as a background infusion in addi-tion to the routine analgesic treatment)30. Subjectswere divided into three groups, the first being toldnothing about the saline, the second that they had a50:50 chance of receiving the painkiller or a placebo,and the third that the saline solution was a potentpainkiller (these three conditions forming, respec-tively, natural history, double-blind, and deceptiveadministration groups). The authors reported that,compared with the natural history group, a 20%decrease in requests for analgesia was observed inthe double-blind group, and a 34% decrease inrequests was observed in the deceptive administra-tion group. The authors concluded that instructionsthat induce a certain expectation of analgesia inducegreater placebo analgesia than those that induceuncertain expectation, a finding subsequently sup-ported in a meta-analysis of 14 similar studies31.In an interesting twist on their open-versus-hiddenadministration design above, Benedetti et al used open-versus-hidden interruption of morphine treatment in


postoperative patients28,29. The authors reported thatpatients who were aware that their treatment had beendiscontinued were more likely to request further mor-phine than those who were unaware that their treat-ment had been interrupted. Similarly, Benedetti et aldemonstrated pain increases associated with the admin-istration of a placebo expected to increase pain in bothclinical32 and experimental patients33. The authors con-cluded that, in the same way that positive expectationsof pain relief might induce or enhance actual pain relief,negative expectations might have the opposite effect.

E V I D E N C E F O R B I O L O G I C A L M E C H A N I S M

O F P L A C E B O A N D N O C E B O E F F E C T S

O N P A I N

Whilst early research into the psychosocial modula-tion of pain tended to focus on demonstrating theeffect and speculating as to mechanisms, recentresearch has gone beyond this to investigatingmechanisms in real time using technology such aspositron emission tomography and magnetic reso-nance imaging. A classic experimental model is theadministration of an agent known to antagonizethe pathway that an administered placebo analgesicpurports to mimic. For example, naloxone antago-nizes the action of opiates such as morphine.Therefore, naloxone would also be expected toantagonize placebo analgesia if the same mecha-nisms as for morphine were responsible. In the firststudy of its kind, Levine et al demonstrated thatnaloxone did in fact disrupt placebo analgesia, con-cluding, logically, that the endogenous opioids areinvolved in the placebo response34. These findings

were replicated several times in both experimentaland clini-calsettings. At the University of Turin,Benedetti et al further demonstrated that as well asthe opioids, cholecystokinin, a neurotransmitterwith an antiopioid action, plays a role in bothplacebo analgesia35 and nocebo hyperalgesia32. Sub-sequent research has demonstrated, for example,placebo-induced activation of brain areas involvedin the pain response36, placebo-induced deactiva-tion of brain areas involved in pain processing37,and correlations between the magnitude of placeboanalgesia and dopamine activity38.

In a study of the placebo analgesia mechanismrelated to sports performance, Benedetti et al inves-tigated the placebo analgesic effects of morphineon a pain endurance test39. Subjects had a tourni-quet wrapped around their forearm and wererequired to squeeze a hand spring exerciser repeat-edly until they could no longer continue. Duringprecompetition training, two “teams”, A and B,received no pharmacologic substance whilst teamsC and D were trained with morphine. During com-petition, team A received no treatment while teamsB and C were given placebo morphine one hourbefore competition. Team D also received what theybelieved was morphine, but they actually receivednaloxone (which would be expected to antagonizethe opioid pathways). As hypothesized, the largestplacebo effect on pain tolerance was observed inteam C who received both the morphine precondi-tioning in the “training” trials and also believedthey had ingested morphine in the competition tri-als. In team D, who had received morphine in pre-conditioning trials, naloxone negated the morphine


preconditioning effects. These findings suggestconditioned activation of endogenous opioids afterplacebo administration, although a correlationbetween morphine and placebo suggests the possiblecontribution of nonopioid mechanisms. It is note-worthy that the placebo analgesic responses wereobtained after only two morphine administrationsseparated by as much as a week. These long timeintervals suggest that pharmacologic conditioningprocedures have long-lasting effects (with implica-tions for the use of proscribed drugs in trainingand competition). Collectively, the findings above and many otherssuggest not only that placebo effects have biologicmechanisms, but that these mechanisms may besimilar or identical to those of the drug the placebomimics. Given this, and given the reduced ratingsof perceived exertion/pain observed in both caffeineand placebo caffeine research, it is reasonable tospeculate that pain reduction might be one mecha-nism by which both the biologic and placebo effectsof caffeine on sports performance operate.

I M P L I C A T I O N S

It is proposed above that pain might be a factor inthe observed ergogenic effect of both caffeine andplacebo caffeine on sports performance. This pro-posal was supported by a brief review of researchthat has demonstrated the placebo effects of caffeineon sports performance, the hypoalgesic effects ofcaffeine in exercise, and the psychosocial modula-tion of pain responses. It is suggested that if painreduction is a key factor in the ergogenic effect ofcaffeine, the findings of the latter body of researchwill be instrumental in elucidating mechanisms andexplaining interindividual variability to response tocaffeine. Specifically, these findings suggest that: ananalgesic response can be conditioned and that theconditioned response might be either reversed orsuppressed by correct information or negativeexpectation, respectively; that expectation of analge-sia is a factor in perceived analgesia; that instruc-tions that induce a certain expectation of analgesiainduce greater analgesia than those that induce anuncertain expectation; and that negative expecta-tions of analgesia associated with an analgesicintervention might offset some of the effectivenessof that intervention. If pain is indeed a factor inthe ergogenic effect of caffeine, all of these find-ings are of interest to researchers investigating thephenomenon. Beyond these findings, evidence from over fifty yearsof scientific research indicates that when a personreceives any of a number of biomedical interven-tions, ranging from tablet to surgery, the brainmight play a role in modulating the effectiveness ofthat intervention14,40-42. Whilst this fact is recognizedby sports scientists, and is accounted for in our useof the placebo-controlled experimental design, such

psychologic factors are rarely considered legitimatevariables in experimental research. This is despite thegrowing experimental evidence for placebo effects onsports performance. Whilst this literature is reviewedelsewhere43, it suffices to state that placebo effects onsports performance resulting from the belief that anergogenic substance had been ingested have beenreported in 12 well-controlled studies5-7,39-43-50. Mostof these effects were in the range of 1%-5%. In threestudies, nocebo (or negative placebo) effects wereobserved as the result of subjects either being givennegative information about an intervention44, hav-ing previous negative experience with caffeine5, orfor reasons that were not entirely clear12. Many of thepositive effects were similar in magnitude to theeffects of the substance that the placebo mimicked.In the few studies in which the biologically activesubstance was administered alongside the placebo, asis the case in the traditional placebo-controlled study,results were arguably not as easily interpretable asthose in which placebos only were administered. Thissuggests that the interaction of belief and biology inreal time, or that order effects in which subjectswere able to detect the presence or otherwise of theactive substance, might have created tension betweeninformation (e.g. “Today you have been given aplacebo”) and perception (e.g. “I’m sure this feelslike I have been given caffeine”). This finding initself warrants further investigation. Whilst knowl-edge that placebos might be powerful in the absenceof the biologically active substance they mimic,such an application is rare in the real world whichscientific research aims to inform.

M O D I F Y I N G R E S E A R C H D E S I G N S T O

U N D E R S T A N D C A F F E I N E E F F E C T S B E T T E R

Burke, in reviewing the caffeine and performanceliterature, argued strongly for more ecologicallyvalid investigations of the phenomenon2. It is fairto argue that, in order to understand any interven-tion fully, it is necessary to determine how it oper-ates in the real world, outside of the laboratory.However, there are many aspects of the ergogeniceffect of caffeine that can be examined in the labo-ratory, one of which is the contribution or other-wise of psychologic factors. The degree to which thebrain is capable of modulating the biologic effect ofan intervention, be that modulation trivial, as islikely the case with a strong poison, or substantial,as has been suggested is the case with several com-plementary medicine treatments, is still littleexplored in sports performance. Biomedical researchis increasingly utilizing sophisticated and elegantdesigns to investigate this area, and in doing so, isproviding a firm biologic basis for what were previ-ously perceived as purely psychologic phenomena.The differences between the laboratory and the fieldand between research and competition are critical fac-tors. Arguably as important, however, are differences


between the subjects’ certainty of having received anintervention or of not having done so, or betweentheir lack of faith in an intervention or extremeconfidence in the same intervention. Burke also high-lighted the interindividual variability in response tocaffeine ingestion in performance research2. Much ofthis variability is likely the result of factors such ashabituation, diet, training status, and availablephysiologic resources. The research described abovesuggests that, if pain reduction is indeed a mecha-nism of the ergogenic effects of caffeine, severalpsychosocial factors might also modulate theergogenic effects of caffeine. It has been suggestedthat some athletes are nonresponsive to caffeine. It

is, however, possible that if caffeine responses arecontext-dependent, an athlete who would notrespond to caffeine in one context, for example, withno explicit expectation of effect, might respond inanother context when given an explicit expectationor when exposed to a conditioning stimulus withfalse-positive performance outcome feedback. Thesefactors are largely related to the subject’s expecta-tions, or conditioned responses, to the caffeineintervention, and these themselves are dependenton either previous experience or currently availableinformation. Therefore, to understand better theeffects of caffeine on performance, these psychoso-cial factors should be treated as variables of interestin caffeine research. Previous experiences of caffeinecould be assessed, and in doing so, several aspects ofthe caffeine experience, including whether the previ-ous intervention had a positive or negative effect onperformance, or whether it was associated with anyside effects, such as nausea or insomnia, could befactored in. It is also useful to evaluate the subjects’expectations of caffeine. As was demonstrated byBeedie et al 5, expectations of a negative effect canresult in a substantial nocebo effect on performance,

even in the absence of caffeine. In this respect it isalso useful to evaluate whether a subject believesthey were given caffeine or placebo in any one trial.Several recent studies have demonstrated that sub-jects who believed themselves to be in placebo con-trol or no-treatment conditions performed belowbaseline, suggesting a potentially powerful noceboeffect. Such an effect is perhaps associated with thehope/anticipation of a positive intervention beingreplaced by the disappointment/anxiety of no inter-vention. It is important to distinguish between con-trolling for current use of, or habituation to, caf-feine and for previous use. Whilst the physiologiceffects of caffeine might be extinct after a few days,

conditioned responses, expectations, or both, mightpersist for several years or even indefinitely. The suggestion that caffeine’s mechanisms mightrelate in part to psychologic factors is intuitivelyappealing, especially when considering the multiplefeedback loops between brain and body during per-formance. However, to date, little research in sporthas examined, or even considered, such biologic andpsychologic interactions in this context. This is notsurprising, given the complex links between percep-tion of effort and pain and the biologic reality at thatexact point in time. However, if sports scientists areto understand fully the mechanisms of interventions,such questions must, at the very least, be asked, ifnot immediately or easily answered. A substantialbody of biomedical research has demonstrated thatbrain and body interact in response to the adminis-tration of a substance about which the subject hassome expectation, be that expectation positive ornegative. This is of course the case when an athleteuses caffeine, ie, the athlete has some expectation ofan effect. Given the failure to definitively supportseveral longstanding theories of caffeine’s ergogenic


effects described by Tarnopolsky above4, and theincreasing database of studies attesting to the poten-tial psychologic contribution to the action ofergogenic aids in sports performance42, it is perhapstimely to investigate the pain hypothesis, and con-sider psychosocial variables that might modulate thepain response, in future caffeine and performanceresearch. ©

R E F E R E N C E S

1 GRAHAM TE., BATTRAM DS., DELA F., EL-SOHEMY

A., THONG FSL. (2008). “Does caffeine alter musclecarbohydrate and fat metabolism during exercise?”,in Appl Physiol Nutr Metab. 33:1311-1318.

2 BURKE LM. (2008). “Caffeine and sport perfor-mance”, in Appl Physiol Nutr Metab. 33: 1319-1334.

3 Bridge CA., Jones MA. (2006). “The effect of caffeineingestion on 8 km run performance in a field set-ting”, in J Sports Sci. 24: 433-439.

4 TARNOPOLSKY MA. (2008). “Effect of caffeine on theneuromuscular system - potential as an ergogenicaid”, in Appl Physiol Nutr Metab. 33: 1284-1289.

5 BEEDIE CJ., STUART EM., COLEMAN DA. et al.(2006). “Placebo effect of caffeine in cycling perfor-mance”, Med Sci Sports Exerc. 38: 2159-2164.

6 POLLO A., CARLINO E., BENEDETTI F. (2008). “Thetop-down influence of ergogenic placebos on musclework and fatigue”, in Eur J Neurosci. 28: 379-388.

7 FOAD AJ., BEEDIE CJ., COLEMAN DA. (2008). “Phar-macological and psychological effects of caffeineingestion in 40 km cycling performance”, in Med SciSports Exerc. 40(Pt 1): 158-165.

8 KIRSCH I., WEIXEL LJ. (1988). “Double-blind versusdeceptive administration of a placebo”, in BehavNeurosci. 102(Pt 22): 319-323.

9 LOTSHAW SC., BRADLEY JR., BROOKS LR. (1996).“Illustrating caffeine’s pharmacological and expectan-cy placebo design”, in J Drug Educ. 26: 13-24.

10 KIRSCH I., ROSADINO MJ. (1993). “Do double-blindstudies with informed consent yield externally validresults? An empirical test”, in Psychopharmacology110: 437-442.

11 BEEDIE CJ. (2007). “The placebo effect in competitivesport: Qualitative data”, in J Sport Sci Med. 6: 21-28.

12 GLIOTTONI RC., MEYERS JR., ARNGRÍMSSON SA.,BROGLIO SP., MOTL RW. (2009). “Effect of caffeineon quadriceps muscle pain during acute cyclingexercise in low versus high caffeine consumers”, inInt J Sport Nutr Exerc Metab. 19: 150-161.

13 MARCORA SM., STAIANO W. (2010). “The limit toexercise tolerance in humans: Mind over muscle?”, inEur J Appl Physiol. Mar 11,. [Epub ahead of print].

14 BENEDETTI F. (2009). Placebo Effects: Understandingthe Mechanisms in Health and Disease (Oxford, UK:Oxford UP).

15 GRAHAM TE., SPRIET LL. (1995). “Metabolic, cate-cholamine, and exercise performance responses to vari-ous doses of caffeine”, in. J Appl Physiol. 78: 867-874.

16 KEOGH E., WITT, G. (2001). “Hypoalgesic effect ofcaffeine in normotensive men and women”, in Psy-chophysiology 38: 886-895.

17 GLIOTTONI RC., MOTL RW. (2008). “Effect of caf-feine on leg muscle pain during intense cycling exer-cise: Possible role of anxiety sensitivity”, in Int JSport Nutr Exerc Metab. 18: 103-115.

18 MOTL RW., O’CONNOR PJ., DISHMAN RK. (2003).“Effect of caffeine on perceptions of leg muscle painduring moderate intensity cycling exercise”, in JPain 4: 316-321.

19 DOHERTY M., SMITH PM. (2005). “Effects of caf-feine ingestion on rating of perceived exertion dur-ing and after exercise: A meta-analysis”, in Scand JMed Sci Sports. 15:69-78.

20 COOK DB. (2006) “Physical activity and pain”, inAcevedo EO., Ekkekakis P. (ed.), Psychobiology ofPhysical Activity (Champaign, IL: Human Kinetics).

21 COOK DB., O’CONNOR PJ., EUBANKS SA., SMITH

JC., LEE M. (1997). “Naturally occurring musclepain during exercise: Assessment and experimentalevidence”, in Med Sci Sports Exerc. 29: 999-1012.

22 O’CONNOR PJ., MOTL RW., BROGLIO SP., ELY MR.(2004) “Dose-dependent effect of caffeine on reducingleg muscle pain during cycling exercise is unrelated tosystolic blood pressure”, in Pain 109: 291-298.

23 MARIDAKIS V., O’CONNOR PJ., DUDLEY GA.,MCCULLY KK. (2007). “Caffeine attenuates delayed-onset muscle pain and force loss following eccentricexercise”, in J Pain 8: 237-243.

24 MONTGOMERY GH., KIRSCH I. (1997). “Classicalconditioning and the placebo effect”, in Pain 72:107-113.

25 LEVINE JD., GORDON NC., SMITH R., FEILDS HL.(1981). “Analgesic responses to morphine and place-bo in individuals with post-operative pain”, in Pain10: 379-389.

26 LEVINE JD., GORDON NC. (1984). “Influence of themethod of drug administration on analgesicresponse”, in Nature. 312: 755-756.

27 AMANZIO M., POLLO A., MAGGI G., BENEDETTI F.(2001). “Response variability to analgesics: A role fornon-specific activation of endogenous opioids”, inPain 90: 205-215.

28 BENEDETTI F., MAGGI G., LOPIANO L. et al. (2003).“Open versus hidden medical treatments: The patient’sknowledge about a therapy affects the therapy outcome”,in Prevent Treat. Available from http://psycnet.apa.org/journals/pre/6/1/1a/ [Accessed May 18, 2010].

29 COLLOCA L., LOPIANO L., LANOTTE M., BENEDETTI

F. (2004). “Overt versus covert treatment for pain,anxiety, and Parkinson’s disease”, in Lancet Neurol.3: 679-684.

30 POLLO A., AMANZIO M., ARSLANIAN A., CASADIO

C., MAGGI G., BENEDETTI F. (2001). “Responseexpectancies in placebo analgesia and their clinicalrelevance”, in Pain 93: 77-84.

31 VASE L., RILEY JL., PRICE DD. (2002). “A comparisonof placebo effects in clinical analgesic trials versusstudies of placebo analgesia”, in Pain 99: 443-452.

32 BENEDETTI F., AMANZIO M., CASADIO C., OLIARO

A., MAGGI G. (1997). “Blockade of nocebo hyperal-gesia by the cholecystokinin antagonist proglu-mide”, in Pain 71: 135-140.

33 BENEDETTI F., ARDUINO C., COSTA S. et al. (2006).“Loss of expectation-related mechanisms inAlzheimer’s disease makes analgesic therapies lesseffective”, in Pain 121: 133-144.


34 LEVINE, JD., GORDON NC., FIELDS HL. (1978).“The mechanism of placebo analgesia”, in Lancet312: 654-657.

35 BENEDETTI F. (1996). “The opposite effects of theopiate antagonist naloxone and the cholecystokininantagonist proglumide on placebo analgesia”, inPain 64: 535-543.

36. PETROVIC, P., KALSO E., PETERSSON KM., INGVAR M.(2002). “Placebo and opioid analgesia - imaging ashared neuronal network”, in Science 295: 1737-1740.

37 WAGER TD., RILLING JK., SMITH EE et al. (2004).“Placebo-induced changes in fMRI in the anticipationand experience of pain”, in Science 303: 1162-1166.

38 SCOTT DJ., STOHLER CS , EGNATUK CM , WANG

H. , KOEPP RA , ZUBIETTA JK. (2007). “Placebo and

nocebo effects are defined by opposite and opioidand dopaminergic responses”, in Arch Gen Psychiatry65: 220-231.

39 BENEDETTI F., POLLO A., COLLOCA L. (2007). “Opi-oid-mediated placebo responses boost pain enduranceand physical performance: Is it doping in sport com-petitions?”, in J Neurosci. 27: 11934-11939.

40 EVANS D. (2003). Placebo: The Belief Effect (London,UK: Harper Collins).

41 KIRSCH I. (2009). The Emperor’s New Drugs: Explodingthe Antidepressant Myth (London, UK: Bodley Head).

42 BEEDIE CJ., FOAD AJ. (2009). “The placebo effect insport. A brief review”, in Sports Med. 39: 313-329.

43 ARIEL G., SAVILLE W. (1972). “Anabolic steroids: Thephysiological effects of placebos”, in Med Sci SportsExerc. 4:124-126.

44 BEEDIE CJ., COLEMAN DA., FOAD AJ. (2007). “Posi-tive and negative placebo effects resulting from thedeceptive administration of an ergogenic aid”, in IntJ Sport Nutr Exerc Metab. 17: 259-269.

45 CLARK VR., HOPKINS WG., HAWLEY JA. et al (2000).“Placebo effect to carbohydrate feeding during a 40-km cycling time trial”, in Med Sci Sports Exerc. 32:1642-1647.

46 FOSTER C., FELKER H., PORCARI JP. et al (2004).“The placebo effect on exercise performance” inMed Sci Sports Exerc. 36 Suppl 5: Abstr S171.

47 KALASOUNTAS V., REED J., FITZPATRICK J. (2007).“The effect of placebo induced changes in expectan-cies on maximal force production in college stu-dents”, in J Appl Sport Psychol. 19(Pt 1): 116-124.

48 MAGANARIS CN., COLLINS D., SHARP M. (2000).“Expectancy effects and strength training: Dosteroids make a difference?”, in Sport Psychologist14(Pt 3): 272-278.

49 MCCLUNG M., COLLINS D. (2007). “‘Because Iknow it will!’: Placebo effects of an ergogenic aid onathletic performance”, in J Sport Exerc Psychol. 29(Pt3): 382-394.

50 PORCARI J., FOSTER C. (2006). “Mind over body:ACE fitness matters”, available from http://www.ace-fitness.org/getfit/PlaceboStudy2006.pdf [AccessedDecember 5, 2006].

a b


a b

Now, my tongue, the mistery telling

Of the glorious Body sing.THOMAS AQUINAS, Pange Lingua Gloriosi.

a b

Kurt Gray is Assistant Professor atDepartment of Psychology, Universityof Maryland, US. Email: [email protected] M. Wegner is a social psychol-ogist, John Lindsley Professor in Mem-ory at William James Department ofPsychology, Harvard University, US,and a fellow of the American Associa-tion for the Advancement of Science.He is known for his work on mentalcontrol (e.g., thought suppression)and conscious will, and for originat-ing the study of transactive memoryand action identification. His book,The Illusion of Conscious Will,tackles the long-debated notion of freewill through the scope of experimen-tal psychology. Wegner's ideas havesparked interest among psychologists,neuroscientists, theologians, andphilosopher interested in the nature ofconsciousness and freedom of action.

H E N S O M E O N E S T E P S O N YO U R TO E O N P U R P O S E,it seems to hurt more than when the per-son does the same thing unintentionally.The physical parameters of the harm maynot differ – your toe is attened in both

cases – but the psychological experience of pain ischanged nonetheless. Intentional harms are pre-meditated by another person and have the specificpurpose of causing pain. In a sense, intendedharms are events initiated by one mind to commu-nicate meaning (malice) to another, and this couldshape the recipient’s experience. This study exam-ined whether self-reported pain is indeed higherwhen the events producing the pain are under-stood as intentionally (as opposed to unintention-ally) caused by another person.Although pain was traditionally conceived to besolely physical in nature (Aydede, 2005), its experi-ence varies substantially with psychological context.The placebo analgesia effect, for example, is the

reduction of pain without a change in physicalstimulation when context, expectations, or sugarpills challenge the interpretation of a sensation aspainful (e.g., Fields, 2008). The nocebo effect, in turn,is the experience of pain without any physical stimu-lation – as when participants report headaches when

told that a (nonexistent) electriccurrent is passing through theirheads (Schweiger & Parducci,1981). These variations in painexperience seem to depend onthe meaning of the stimulus: Asugar pill is meant to decreasepain, whereas electric current ismeant to increase pain. In aninterpersonal context, the mean-ing of an action is derived fromthe perceiver’s perceptions ofthe actor’s intention (Clark,1996), which means that inten-tional harms, unlike accidentalharms, are meant to cause pain. The possibility that the mali-cious intent of other peoplecould be translated into addi-tional physical pain is suggestedby studies demonstrating thatsimilar areas of cortex respondto both physical pain and socialharms (Eisenberger, Lieberman,

& Williams, 2003). Social harms, which are pre-sumably laden with intention, have also been shownto be more painful to relive than simple physicalharms (Chen, Williams, Fitness, & Newton, 2008).So, although a broken toe (or electric shock) mayhurt, an intentionally broken toe (or electric shock)should hurt more.

M E T H O D

Forty-eight participants (68% female, 32% male)participated in a lab study of ‘‘psychophysical per-ception in pairs.’’ Four participants were excludedfor suspicion and one participant was excluded forfailing to follow instructions, leaving a total of 43. On arrival, participants met their study partner – aconfederate – and were escorted to an individualroom. They were then introduced to the psy-chophysical tasks of color matching, number estima-tion, pitch judgment, and discomfort assessment,each of which they completed. Discomfort assess-ment involved being administered an electric shock


W

T H E S T I N GO F P OT E N T I A L PA I N

K U R T G R A Y ~ D A N I E L M . W E G N E R

“ Old age

is the most

unexpected

of all things

that happen

to man. ”LL EE VV TT RR OO TT SS KK YY

g | P R O P O S A L

mailto:[email protected]

and evaluating it on a 7-point scale ranging from notat all uncomfortable to extremely uncomfortable.Shocks of 1-ms duration were delivered to the wristof the dominant hand through a stimulator (BiopacSystems, Goleta, CA), with voltage precalibrated foreach participant to be ‘‘very uncomfortable.’’ Volt-ages ranged from 40 to 75V between subjects. Partici-pants evaluated two blocks of computer-adminis-tered electric shocks initially in an individual prac-tice session as a baseline pain measure. On each experimental trial, participants saw a com-puter screen with two potential tasks before com-pleting one of them. When discomfort assessment

was a potential task, the alternate task was evaluat-ing the relative pitches of tones. On this and othertrials, participants were told that the participant inthe next room (the confederate) would select whichtask the participant would complete.In the intentional condition, the confederate chosethe discomfort-assessment task when it was anoption, and participants received an electric shock.In the unintentional condition, the confederateselected the pitch-judgment task when discomfortassessment was an option. In this condition, however,participants were told that the mapping between theselection and administration of tasks was switched,unbeknownst to the confederate, so they wouldalways receive the task opposite to the one selectedby the confederate. Thus, when pitch judgment wasselected for them, they completed discomfort assess-ment and received an electric shock. On their computer screen, participants saw boththe confederate choice and the actual task to beadministered (in advance), which ensured that par-ticipants were not surprised when they received an

electric shock and also reinforced the intentional orunintentional nature of the shock. Pilot testing con-firmed that shocks were perceived (on a 7-pointscale) as more intentional in the intentional condi-tional (M=5 5.64, SD=1.49) than in the unintentionalcondition (M=2.17, SD=0.83), t(24)=7.13, p<.01,Prep=.99, and that the confederate was seen as moreblameworthy (on a 5-point scale) in the intentionalcondition (M=2.43, SD=1.40) than in the uninten-tional condition (M=1.41, SD=0.67), t(24)=2.29, p<.03,Prep=.91. In both conditions, participants completedthree blocks of experimental trials after the two practice/baseline trials.

R E S U L T S A N D D I S C U S S I O N

Mean pain ratings from shocks in each of the 5blocks (see FIGURE 1) were submitted to a 2 (condi-tion: intentional; unintentional) x 5 (time: blocknumber) between-within analysis of variance, whichrevealed the predicted interaction, F(4, 164)=3.09,p=.02, Prep=.93, π2=.07. A composite of the two prac-tice blocks revealed no significant difference inexperienced pain between conditions (t<1); however,an average of experienced pain in the three experi-mental blocks revealed that intended pain (M=3.62,SD=0.99) was experienced as more painful thanunintended pain (M=3.00, SD=0.78), t(41)=2.21,p=.03, Prep=.91. Additionally, there was a significant decreasing lin-ear trend of experienced pain in the unintentionalcondition, F (1,17)=20.18, p=.001, Prep=.99, suggest-ing that participants in this condition exhibited thestandard pattern of habituation to repeated painfulstimulation (Greffrath, Baumgartner, & Treede,2007). In contrast, there was no linear trend in the


EX

PER

IEN

CE

DPA

IN

BLOCK 1 BLOCK 2 BLOCK 3 BLOCK 4 BLOCK 5

BASELINE/PRACTICE TRIALS EXPERIMENTAL TRIALS

TIME

INTENTIONAL ELECTRIC SHOCK

UNITENTIONAL ELECTRIC SHOCK

FIGURE 1 ~ Experienced pain as a function of whether electric shocks were perceived as intentional or unintentional.

3

4

intentional condition, F=0.08, suggesting that partic-ipants in this condition continued to feel the freshpain of an intentional harm as time went on. This study provides evi-dence that the experienceof pain changes depend-ing upon the psychologi-cal context in which peo-ple are harmed. Specifi-cally, the meaning of aharm – whether it wasintended – influences theamount of pain it causes.Although people canbecome accustomed tothe pain of an uninten-tional harm, the malicebehind an intentionalpain keeps it stinging. ©

A C K N O W L E D G M E N T S

We thank assistant TiffanyNiver, Jane Fang, DanielLiss, and Dan Scaduto. Thiswork was supported by theSocial Science and Humani-ties Research Council ofCanada, the Institute forHumane Studies, andNational Institute of MentalHelalth Grant MH 49127.

R E F E R E N C E S

AYDEDE, M. (2005).“Introduction: A Crit-ical and Quasi-histori-cal Essay on Theoriesof Pain”, in M. Aydede(ed.), Pain: New essayson its nature and themethodology of its study(Cambridge, MA: MITPress): 1-58.

CHEN, Z., WILLIAMS, K.D.,FITNESS, J., & NEW-TON, N. (2008). “Whenhurt will not heal:Exploring the capacityto relive social andphysical pain”, in Psy-chological Science 19,789-795.

CLARK, HH. (1996). Usinglanguage (New York:Cambridge UP).

EISENBERGER, N.I.,LIEBERMAN, M.D., &WILLIAMS, K.D.(2003). “Does rejection hurt? An MRI study of socialexclusion”. in Science 302: 290-292.

FIELDS, H.S. (2008). Pain: Mechanisms and management(2nd ed.) (New York: McGraw-Hill).

GREFFRATH, W., BAUMGARTNER, U., & TREEDE, R-D.(2007). “Peripheral and central components of habit-

uation to heat pain percep-tion and evoked potentialsin humans”, in Pain 132,301-311.SCHWEIGER, A., & PAR-DUCCI, A. (1981). “Noce-bo: The psychologic induc-tion of pain”, in PavlovianJournal of Biological Science16, 140-143.

a b


John N. Williams is an AssociateProfessor in Philosophy at the Schoolof Social Sciences, Singapore Man-agement University. He received hisPhD from Hull University, UK. Hisresearch interests include paradoxes,theory of knowledge, philosophy ofreligion and applied ethics. Hisresearch has been published in ActaAnalytica, American PhilosophicalQuarterly, Analysis, AustralasianJournal of Philosophy, Journal ofPhilosophical Research, Mind,Philosophy East and West, Philo-sophical Studies, Religious Studies,Synthese, and Theoria. He is a co-edi-tor of Moore’s Paradox: New Essayson Belief, Rationality and the FirstPerson, Oxford University Press 2007.Email: [email protected]


L A C E B O - T R I A L S O N H I V - I N F E C T E D P R E G N A N T

women in developing countries like Thai-land and Uganda1,2 have provoked recentcontroversy 3,4. Such experiments aim tofind a treatment that will cut the rate of

vertical transmission more efficiently than existing‘gold standard’ treatments like zidovudine. Is suchan experiment morally justified? I think that theright question to ask is this: “Is it always, never orsometimes, morally justified to experiment on HIV-infected, pregnant women in developing countries(by means of placebo trials) in order to develop anew treatment X which will reduce the rate ofmother-to-child HIV transmission more effectivelythan the existing (‘gold standard’) treatment Y?”

T H R E E E T H I C A L P R I N C I P L E S A N D H O W

T H E Y C O N F L I C T

Put like this, the issue acquires a level of generality.For example, the specific country in which theexperiment is carried out is not the issue. Unless

there are clear differences from country to countrywhich are morally relevant to the question above,our answer to it should be the same whatever thecountry in which such experiments are conducted.Extracting a clear and consistent answer is not easy,given the emotional horror that surrounds even a

suitably generalised question.The consequences of the dis-ease are horrible and threatento multiply through successivegenerations. The horror isaccentuated by the innocenceof the foetus as a recipient ofthose consequences, an inno-cence that remains regardless ofthe academic question ofwhether the foetus is a personor merely a potential person.Add to this the perplexingconundrum of placebo, withits levels of ignorance. Finally,we must contextualise theproblem against the back-ground of the horrors builtinto developing countries, suchas unequal and inadequateresources, lack of educationand poverty.Where is a doctor or a researcherto look for guidance in such a

case? ‘Conscience’, isn’t any substantive answer, sincediffering consciences of doctors are just differinginternalisations of some medical code. However,there is an authoritative code of medical practice,which provides guidance, namely the BelmontReport 5.The Belmont Report provides an excellent sharpen-ing of the principles laid down in the unmodifiedHelsinki Declaration6. In essence, it urges three prin-ciples: the Principle of Utility (there called Benefi-cence), the Principle of Autonomy (there calledRespect for Persons) and the Principle of Justice.The Principle of Utility has a negative and a posi-tive form. Its negative form states that

~ PB neg) It is one’s duty to refrain from doingharm while its positive states that~ PB pos) It is one’s duty to minimise harm andmaximise benefit, to society in general, includingthat of future patients.

The Principle of Autonomy claims that individualsshould be treated as autonomous agents, and that


“ Apart

of the

known

and the

unknown,

what

else is

there? ”HH AA RR OO LL DD PP II NN TT EE RR

g | R E F L E C T I O N S

T H E E T H I C S O F P L AC E B O -C O N T RO L L E DTRIALS IN DEVELOPING COUNTRIES TO PREVENT

MOTHER-TO-CHILD TRANSMISSION OF HIV*

J O H N N I C H O L A S W I L L I A M S

P

persons with diminished autonomy are entitled toprotection. In other words,

~ PA) It is one’s duty to respect autonomouschoices and to protect those with diminishedautonomy.

The Principle of Justice states that ‘research shouldnot unduly involve persons from groups unlikely tobe among the beneficiaries of subsequent applicationsof the research’. The spirit of this principle is that

~ PJ) It is one’s duty to distribute benefits (ofresearch) fairly

or at least, not worsen the imbalance of benefitsand burdens of research among relevant groups.These three voices of utility, autonomy and justicecannot, in themselves, provide clear guidance in allcases, since there are possible scenarios in whichthey give conflicting rulings. It seems reasonable tothink that the negative voice of utility is not a totalprohibition of any form of harm, such as a slightlypainful injection. Rather it should be read as ‘Donot cause more harm than benefit to a single indi-vidual’, but this may still conflict with the positivevoice of utility. Suppose that the general increase ofgood over harm to society in general (by means of areduction of the number of children who will beborn with HIV) can be purchased only at the cost ofexposing the mothers in the experimental group toa risk of substantial harm that is greater than thechance of slight benefit. Since such exposure is itselfa type of harm, the voice of utility is confused.Now consider the cost to the mothers in the controlgroup. Their babies are effectively condemned toHIV infection, the incidence of which could havebeen reduced by giving them the ‘gold standard’standard treatment Y (for example, zidovudine) thatis known to be effective in cutting the rate of verti-cal HIV transmission. In the name of both babiesand their mothers, the negative voice of utility pro-hibits the experiment, while its positive voicedemands it. Suppose further that HIV-infectedmothers can only be recruited for the experiment bycoercing them to participate or by withholdinginformation about the risks involved (the absence ofany effective treatment in the case of the controlgroup and the risks-compared-to-benefits of treat-ment X in the case of the experimental group) rela-tive to the benefits available to others (by means ofthe existing treatment Y). Here the voice of utilitycontradicts that of autonomy, since informed andfree choice to participate has been ruled out. Orsuppose that the treatment X, once available, will berestricted to a minority of HIV-infected pregnantwomen in countries wealthy enough to afford it,excluding those in the developing country in whichthey were originally developed. Even if we could besure that the rate of vertical transmission woulddecrease worldwide, in line with the voice of utility,we would still hear the cry of injustice.

O N E H I E R A R C H Y O F T H E

T H R E E E T H I C A L P R I N C I P L E S

These examples show that the principle of utilitywill contradict those of autonomy and justice incases that are at least possible. This means that wemust decide in advance which voice has authorityover which. As a matter of fact, most of us hear theleast authority in the voice of utility. In examplessuch as those just considered, most will judge thatmaximising utility at the cost of injustice or at thecost of disrespecting the free choices of people (thustreating them as means to the end of increasedhealth world-wide) is morally repugnant. Thus, oneway to achieve consistency among the three princi-ples that will fit the moral intuitions of many is tohold that the voice of utility must be obeyed, but onlyafter the voices of justice and autonomy have beenobeyed. Those who hear things that way will not beswayed in the least by the tinkering with the originalwording of the Helsinki Declaration which was sentto the World Medical Association’s member associa-tions in advance of the World Medical AssociationCouncil Session in Santiago, Chile on April 15, 1999.Working from the background of the questiontowards empirical specifics, the Principle of Justiceis the most effective choice of principles to applyfirst. If the degree of poverty of most pregnant HIV-infected mothers in the developing country (such asUganda) will prevent them from buying theimproved cure, if found, for several generations ofHIV-infected offspring to come, while most of therelatives of the researchers (such as Americans) willbe able to buy it as soon as it hits the market, thensurely this is injustice. As for other scenarios, surelyenlarging the relative size of burdened group versusbenefiting group proportionally worsens the injus-tice. Given that injustice will be done, there is some-thing morally wrong with the experiment, quiteregardless of whatever else is ruled wrong by theother principles. Taking the three principles as aguide to procedure, we must at least postpone suchexperiments until enough economic aid has beengiven to these countries as makes the chances of ben-efit more fairly distributed. Otherwise the researchersshould experiment upon their own extended geo-political-economic kin. Given the commercial hege-mony of the rich drug companies involved, theiropposition to the production of cheaper ‘generic’drugs and the huge disparity between levels of wealthin developing countries and developed countries, thechances of benefit to the burdened group were clearlynot fairly available to them.Let us now suppose (hypothetically but implausibly)that equality has been redressed or is not the issue.What does the voice of autonomy tell us? There aretwo groups for whom it could speak, the foetusesand their mothers. Clearly the foetus is incapable ofmaking choices, informed or not, so whether or notwe say that they are potential persons or already per-sons, there can be no autonomous choices made by


them which are respected or disrespected. In thisrespect, the voice of autonomy is neutral on the per-missibility of the experiment. On the other hand,the principle of autonomy also commands us to‘protect those with diminished autonomy’, whichappears to include the foetus. If so, on balance, thevoice of autonomy prohibits the experiment in thename of the foetus.It might be objected that the class of those entitledto protection excludes non-persons such as thefoetus. This objection clearly has little bite againstthose who hold that the foetus is to some degree aperson at some stage in its development. Nor is itpersuasive against those, including the mothers,who think that whether the foetus is an actual orpotential person, it is still something valuable,which therefore needs protection.How does autonomy speak for the mothers? Itclearly prohibits the experiment unless the motherhas made an autonomous choice to participate.Autonomous choices must be informed choices. Inthe case of a non-placebo experiment in which thenew treatment X is to be tested, this means that themother must understand the risks and probablebenefits to herself and her foetus posed by the newtreatment X, as compared to risks and probable bene-fits conferred by the existing treatment Y. Since theprobability of risks and benefits of the new treat-ment may be precisely what the experiment isdesigned to discover, it may not be possible to giveher precisely this information. In that case, themother must be informed of the degree of uncertaintyof the probability of possible harms and benefits ofthe new treatment X, as compared to the degree ofcertainty already established of the probabilities ofharms and benefits of the existing treatment Y.Anything less would not be full information.

C O M P L I C A T I O N S A R I S I N G

F R O M P L A C E B O C O N T R O L

In the case of a placebo experiment, things are morecomplicated. Placebos aim to separate the causalpowers of X from the causal powers of the belief inthose causal powers. For example, they aim to filterout cases in which a patient feels better simplybecause that patient believes (correctly or incorrectly)that the treatment will work. The belief in questionmay be held by the mothers or the experimenters orboth. Clearly no such belief can be held by the foe-tus. Assume a simple single-blind placebo control inwhich each mother has a fifty-fifty chance of beingselected for the control group to receive sugar asopposed to the experimental group to receive thenew treatment X, such that no mother will knowwhich group she is in. Obviously, this design ofexperiment rules out informing the mothers whichgroup they are in, but this does not mean they can begiven no information at all. Autonomy demands thateach mother understand that she has a half-chance of

ending up in the control group, in which case herfoetus will certainly be born with HIV and a half-chance of ending up in the experimental group,with possible risks and benefits to the foetus whichare uncertain. She must also understand that she facesthese two outcomes in the teeth of the knowledge thattreatment Y already exists, with its more certain risksand benefits to the foetus.It might be objected here that the lack of educationamong such mothers renders them incapable ofunderstanding such information. If so, autonomydemands that education be a more pressing prioritythan medical research. The experiment must be post-poned until the would-be participants are educatedto a level that enables them to understand whatchoices are open to them. Since justice demands thatthe benefits of education be fairly distributedthroughout the developing country in which theexperiment is to take place, this means that theexperiment must be postponed until any would-besub-group of participants has a level of educationthat is representative of the whole population ofthat country.This brings us back to the purpose of the placebo.Suppose that an HIV-infected pregnant womanbelieves that she is receiving a new miracle drug thatcarries an ironclad guarantee of protecting her foe-tus from infection. In fact her belief is mistaken, forthe drug she has been given is just sugar. It seemsunlikely that her belief could have any positive psy-chological effect upon her foetus. In terms of themother, the placebo seems to serve no useful pur-pose. So far, there is no justification for the creationof the control group, in which the foetuses are con-demned to HIV infection. Thus the only other pur-pose that the placebo can serve is to separate thecausal powers of X in reducing vertical transmissionfrom the causal powers of the beliefs of the experi-menters. A double-blind trial, in which the experi-menters do not know which mother is receivingsugar or receiving X, will ensure that they will notbias the development of one group of foetuses overthe other, by giving different levels of care to thetwo groups of mothers. The question to ask now iswhether the increased accuracy of the results of theefficiency of X is worth the price of allowing themothers in the control group to go untreated. Thevoice of autonomy tells us that the price need notbe paid. Since mothers in both groups are personsequally deserving of care and respect, the experi-menters have a duty to provide both groups withthe best possible level of care, regardless of theirbeliefs about whose foetuses are most likely to beprotected from vertical transmission. Once thisduty is discharged, there is no possible justificationleft for the inclusion of the placebo group.Thus there appears to be no need for a placebo groupat all. Surely the ethical procedure would be to givethe control group treatment Y instead. After all, are we


not trying to measure the difference in effectivenessof the new treatment X as compared to the existingtreatment Y? If the participants in both groups werefully informed of the experimental set-up, therewould be no room for objection on grounds ofautonomy, nor any room for objection on grounds ofutility to the treatment of the participants in the con-trol group, since this group is assured of the bestknown treatment available anyway.Given my definition of a placebo as a control whichaims to separate the causal powers of a treatmentfrom the causal powers of belief (or faith) in thattreatment, there is, strictly speaking, a conceptual dif-ference between a placebo group and a ‘no treatment’

control group, which aims to help isolate possibleunwanted side-effects of the treatment. In experi-ments to develop new treatments that are more effec-tive in reducing the vertical transmission of HIV, a ‘notreatment group’ would help to isolate unwantedsideeffects of the new treatment on the foetus. I nowturn to this issue.

‘ N O - T R A T M E N T G R O U P S ’ W I T H I N

C O N T R O L L E D S T U D I E S

Similar moral objections can be made to differentkinds of experiments such as the Rakai project inUganda7, in which the experimenters studied theeffect of other sexually transmitted diseases on therate of heterosexual transmission of HIV and thenatural risk factors that determine the heterosexualtransmission of HIV-1 over periods of unprotectedsex. Such experiments use a ‘no-treatment’ groupwithin a controlled study. Such an experiment isruled unethical on the grounds of autonomy, ifthose who were left untreated were not fullyinformed that this was precisely what would hap-pen to them. By the experimenters’ own admission,

those in the ‘no-treatment’ group were not fullyinformed. Quinn et al 8 reported that 228 HIV-infected couples were left untreated for up to thirtymonths, and the decision to inform the uninfectedpersons that their partner was infected was left upto the infected persons themselves, although theexperimenters regularly saw both. Surely this doesnot count as discharging a duty to give all partiesconcerned the full information. To so discharge it,the experimenters should have told each couplethat one partner was infected as well as telling themthat the infection would go untreated.In the same project, the investigators treated half ofthe villagers for sexually transmitted diseases such

as syphilis while leaving the other half untreated.Their aim was to determine the effect of concur-rent sexually transmitted diseases on the heterosex-ual transmission rate of HIV. Assume for the sake ofargument, that this information was obtained andthat it helped to develop more effective ways ofcutting the HIV transmission rate. In terms of max-imising the utility to society in general, the investi-gators were morally justified. Moreover, anyone whomaintains consistently that the voice of utility alwaysoverrides the voices of autonomy and justice candefend the investigators, but that is not the moralframework I have suggested. Given that utility issubordinate to autonomy and justice, we need to askwhether the investigators infringed the informedchoices of the villagers who were left untreated.Deciding this is not easy. One way to look at it is tosay that the untreated villagers were simply leftalone by the investigators to carry on as before, sono interference took place. On the other hand,there is a clear sense in which the untreated villagerswere selected by the investigators to form one half ofthe experiment. In this sense, they were intentional-ly included in the experiment by default. Therefore


the voice of autonomy demands that the experi-menters give them an informed choice to continueto participate, which in turn means telling themthat they would go untreated. Had the villagersheroically agreed to forgo treatment for the sake offuture generations then there could be no objectionin the name of autonomy. Otherwise, continuingto include them in the experiment would be moral-ly wrong. Against this, it might be objected thatthe investigators were not doing anything extraor-dinary, since the villagers would not have beentreated in the ordinary course of events anyway.However, it is not so clear that this means theinvestigators were not interfering. Given that theinvestigators selected this particular group of vil-lagers for the control arm of the experiment, andthen continued to withhold both treatment andinformation when they could easily have suppliedboth, were they not interfering? In selecting andthen ignoring them, surely the investigators wereactually doing something to them.The scenario of the untreated villagers is no differentin principle from that in which a group of pregnantHIV-infected mothers is left untreated as a control toa second group who are given a new treatment X, inorder to help obtain information on (among otherthings) whether X will have unwanted side-effects onthe foetus. Assume, for the sake of argument, thatthe information is obtained and that it is instrumen-tal in reducing harmful side-effects to babies who arein general, less likely to be born with HIV (thanwhen their mothers were treated with the originaldrug Y). Again, the voice of utility permits, evendemands, the inclusion of the ‘no-treatment group’,but again, given that utility is subordinate to auton-omy and justice, we need to ask whether the investi-gators infringed the informed choices of the motherswho were left untreated. If the investigators firstselected them as a control arm and then withheldtreatment and information (including the informa-tion that they were infected and that treatment wasavailable), then the mothers’ autonomy was violated.

T H E U T I L I T A R I A N ‘ N O L O S S ’ D E F E N S E

Some commentators9 in the debate have arguedthat most HIV-infected pregnant women in devel-oping countries would not be able to afford anysort of treatment against vertical transmission any-way. Thus the women in the original experimentare in a no-loss situation. They have a half-chanceof ending up in the placebo group, in which casethey are no worse off than they would be anywayand a half-chance of ending up in the experimentalgroup, in which case they have a secondary chance(the degree of which is yet unknown) of protectingtheir foetus.This argument suffers from a number of flaws. Arethe women in the placebo group really no worse offthan they would have been had they never partici-pated? If autonomy has been satisfied, then the

women will know that they are in a kind of desperatelottery that holds out a chance for the well-being oftheir unborn. Since the chances of protection forthe foetuses in the experimental group are less thancertain, they should know that overall, the odds areagainst them. Nonetheless, some basis for hopeexists, which may well be the very reason why theyhave agreed to participate. Yet the experimentersknow in advance that for each mother in the placebogroup, all hope will be dashed. This burden of falsehope and its certain disappointment surely repre-sents a significant cost to this group of mothers.The no-loss argument gains its plausibility fromthe claim that the personal utility of the volunteersis not decreased. But it ignores the salient fact thatthe volunteers start from a position of inequality.Were justice done, the proven, if limited benefitsof the existing treatment Y would be distributedfairly throughout the world to those in developingcountries who need them most. Given that theinclusion of the placebo group is justified, theautonomous choice of the mothers would then bethe more heroic one of taking a half-chance of noprotection for their unborn and a half-chance ofthe uncertain degree of protection conferred bytreatment X, in preference to the guarantee of thecertain but limited degree of protection conferredby the existing treatment Y. Given, as I suggestedabove, that the inclusion of the placebo group isnot justified, the autonomous choice of the moth-ers would then be the less heroic but still coura-geous one of trading the certain but limited degreeof protection conferred by treatment Y for the pos-sibly improved but uncertain degree of protectionconferred by treatment X.In the original experiment, even those motherswho end up in the experimental group are doublywronged. Firstly, they are wronged both in thename of justice and in the name of autonomy bybeing unfairly denied the choice of treatment Y.Then the experimenters restrict their choices tothe options of no treatment (by refusing to partici-pate) or the option of a half-chance of the uncer-tain benefits of treatment X (by consenting to par-ticipate). The restriction is genuine, since it isalways within the power of experimenters to offertreatment Y as well. Since this is a perpetuation ofthe original injustice and an erosion of autonomy,this is a further wrong. In this respect, the would-be participant resembles a workman who has beenunfairly denied any payment. His employer offers totoss a coin. If the coin comes down heads, the work-man will receive half his wages; if tails, nothing. Ifhe refuses the bet, again he gets nothing. Theworkman is first wronged by being deprived anypayment. When the bet is offered, he has nothingfurther to lose. Yet surely the offer of the bet is asecond wrong since it perpetuates the originaldeprivation of full payment, one that the employeris in a position to put right.


E F F E C T S O F M O D I F Y I N G T H E

D E C L A C A R A T I O N O F H E L S I N K I

Whereas the current Declaration assures researchparticipants of ‘the best proven diagnostic and ther-apeutic method,’ the corresponding section in themodified Declaration adds the phrase ‘that wouldotherwise be available to him or her.’ This appearsto vindicate the ‘no loss’ argument, since there is notreatment available to the would-be participants inthe experiment should they not participate, giventhat they are too poor to afford any treatment,including treatment Y. But this vindication is anillusion. For one thing, there is an ambiguity in thephrase ‘available’. On one reading, treatment Y isnot available to the group in the sense that theywould not have been able to obtain it, had theynever come in contact with the experimenters. However,in that sense, since the best treatment available oth-erwise is none, this amendment is just another wayof saying that the group in question is assured of notreatment. We should therefore reject this amend-ment to the Declaration, in the name of both jus-tice and utility as originally voiced by it. Utilitydemands that the benefit to society be maximisedand justice demands that this benefit be fairly dis-tributed. In other words, the group in questionmust be given a fair chance of treatment Y. Theother, more sensible, reading of ‘available’ is thattreatment Y is not available to the group as would-beparticipants who are now in contact with the experi-menters. However, in this sense, treatment Y is avail-able to them even if they now choose not to partici-pate, since it is fully within the experimenters’power to provide it. Thus even the modified decla-ration tells us that the experiment is morally wrong.

C O N C L U S I O N

Given plausible assumptions about the level ofpoverty and education in the developing countrytargeted, the placebo-controlled trials of the type dis-cussed are unethical violations of both justice andautonomy. In any case, no moral justification can befound for the inclusion of a placebo group. By con-trast, the inclusion of a ‘no control’ group may bejustified, but only when the experimenters have notinterfered with the autonomy of its members. Exper-iments such as the Rakai project in Uganda are suchunethical violations of autonomy. The developmentof third-world countries, in the form of economicdevelopment and education, must be priorities thatcome before such experiments. ©

——————

* This paper was first submitted when the author was aFellow in Philosophy in the Department of Philosophy,National University of Singapore and subsequently revisedat the Singapore Management University).

R E F E R E N C E S

1 CONNOR E. M., SPERLING R. S., GELBER R., KISELEV

P., SCOTT G., O’SULLIVAN MJ. et al (1994). “Reduc-tion of maternal-infant transmission of humanimmunodeficiency virus type 1 with zidovudinetreatment”, in N Engl J Med 331: 1173-80.

2 SHAFFER N., CHUACHOOWONG R., MOCK P.A.,BHADRAKOM C., SIRIWASIN W., YOUNG NL., et al(1999). “Short-course zidovudine for perinatal HIV-1transmission in Bangkok, Thailand: A RandomisedControlled Trial”, in Lancet 353: 773-80.

3 LURIE P., WOLFE S. M. (1997). “Unethical trials of inter-ventions to reduce perinatal transmission of thehuman immunodeficiency virus in developing coun-tries”, in N Engl J Med 337: 853-6.

4 ANGELL M. (1998). “The Ethics of Clinical Researchin the Third World”, in N Engl J Med 337: 847-9[see also correspondence in N Engl J Med 1998; 338:836-44].

5 NATIONAL COMMISSION FOR THE PROTECTION OF

HUMAN SUBJECTS OF BIOMEDICAL AND BEHAVIORAL

RESEARCH. The Belmont Report. April 18, 1979.6 DECLARATION OF HELSINKI IV, 41st World Medical

Assembly, Hong Kong, September 1989. In: AnnasGJ., Grodin MA. (ed.) The Nazi Doctors and theNuremberg Code: Human Rights in Human Experi-mentation (New York: Oxford UP, 1992: 339-42.

7 WAWER MJ., SEWANKAMBO NK., SERWADDA D.,QUINN TC., PAXTON LA., KIWANUKA N. et al(1999). “Control of sexually transmitted diseases forAIDS prevention in Uganda: a randomised commu-nity trial” in Lancet 353: 525-35.

8 QUINN TC., WAWER MJ., SEWANKAMBO N., SERWAD-DA D., LI C., WABWIRE-MANGEN F. et al (2000).“Viral load and heterosexual transmission of humanimmunodeficiency virus type 1”, in N Engl J Med342: 921-9.

9 PHANUPHAK P. (1998). “Ethical issues in studies inThailand of the vertical transmission of HIV” in NEngl J Med 338: 834-5.

a b


a b

The limits of my languagemean the limits of my world.

LU DW I G W I T TG E N S T E I N , Tractactus Logico-Philosophicus.

a b

F A B R I Z I O B E N E D E T T I

T H E M A N Y P L A C E B O E F F E C T S

Due to the involvement of many mechanisms, the studyof the placebo effect can be viewed as a melting pot ofconcepts and ideas for neuroscience. Indeed, there existnot a single placebo effect, but many, with differentmechanisms and in different medical conditions andtherapeutic interventions. In fact, expectation, anxietyand reward are all involved, as well as a variety of learn-ing phenomena. There is also some experimental evi-dence of different genetic variants in placebo responsive-ness. Pain and Parkinson’s disease are today the most pro-ductive models to better understand the neurobiology ofthe placebo effect, and the neural networks that areinvolved have been identified, such as an opioidergic-cholecystokinergic-dopaminergic modulatory network inpain and part of the basal ganglia circuitry in Parkinson’sdisease. Important implications emerge from these recentadvances in placebo research, such as the impact of thepsychosocial context on the patient’s brain.

E D Z A R D E R N S T

THE ETHICAL IMPLICATIONS OF USING PLACEBOIN CLINICAL PRACTICE

By using the concrete example of homeopathy, this paperconsiders whether placebo use can be ethically justified.On one hand, clinicians may want to help patientsthrough a beneficial placebo effect, and highly dilutedhomeopathic remedies are unlikely to have adverseeffects. Also, many patients expect to receive a prescrip-tion when visiting the doctor and believe in the effective-ness of homeopathy. However, the author presents sever-al reasons the use of placebos may not be considered eth-ical. Recent survey data shows that the most frequentreasons for prescribing placebos are mainly for the physi-cians’ convenience rather than for altruistic motives. Toobtain the placebo effect, physicians cannot tell thepatient they are receiving a placebo, which goes againstthe ethical requirement for a doctor to obtain informedconsent from a patient. This deception can underminetrust, which is necessary for good therapeutic relation-ships. Placebos, especially impure ones, also present therisk of producing adverse effects and may cause people tobelieve there is a medical cure for every state of reducedwell-being. This paper suggests that an effective treat-ment that treats a patient’s symptom(s) will be more use-ful than a homeopathic remedy because the patient willget relief from his/her symptom(s) while also receivingthe placebo effect. The author recommends that clini-cians analyse their own behaviour to determine whetherplacebo use is in the best interest of the patient or fortheir own convenience.

C H R I S T O P H E R J . B E E D I E

A L L I N T H E M I N D ? P A I N , P L A C E B OE F F E C T , A N D E R G O G E N I C E F F E C T O FC A F F E I N E I N S P O R T S S P E R F O R M A N C E

The ergogenic effects of caffeine on performance arewell documented. These effects are more evident in

endurance and short-duration, sustained-effort eventsthan in interactive or stop-go sports. Experimentally-induced placebo effects of caffeine on sports perfor-mance have also been observed in a number of recentstudies. In the present paper it is argued that, given thenature of the sports in which caffeine effects are observed,the well documented hypoalgesic effects of caffeine, andthe fact that pain is highly placebo-responsive, a reduc-tion in perceived pain might be the common factor inboth the biologic and placebo ergogenic effects of caf-feine on sports performance. This idea is supported byevidence from medicine that suggests placebo effects areoften associated with mechanisms similar or identical tothose of the substance the subject believes they haveingested. Research findings from both biomedicine andsports medicine that attest to the interaction of biologicand psychologic factors in caffeine and pain responsesare briefly reviewed. In conclusion, it is recommendedthat researchers investigate the pain hypothesis. Further-more, researchers should consider psychosocial factorsthat might modulate the pain response as variables ofinterest in future caffeine and performance research.

K U R T G R A Y ~ D A N I E L M . W E G N E R

T H E S T I N G O F I N T E N T I O N A L P A I N

This study tests whether the psychological context ofpain affects the amount of pain an individual feels,namely whether intentional harm causes more perceivedpain than unintentional harm. The researchers used anexperiment in which the subjects were instructed to per-form a series of tasks with a partner in a separate room,who was actually an accomplice. One task, discomfortassessment, involved the subject receiving an electricshock and rating his/her level of discomfort. One groupwas led to believe their partner intentionally chose forthem to receive the shock rather than choosing for themto complete another task, while the other group was toldthe partner unintentionally chose for them to get theelectric shock. The results show that intended pain wasexperienced as more painful than unintended pain. Inaddition, intended pain appeared to be felt fresh everytime, rather than lessening as the subject became habitu-ated to repeated painful stimulation.

J O H N N I C H O L A S W I L L I A M S

THE ETHICS OF PLACEBO-CONTROLLEDTRIALS IN DEVELOPING COUNTRIES TO PREVENT

MOTHER-TOCHILD TRANSMISSION OF HIV

Placebo-trials on HIV-infected pregnant women in devel-oping countries like Thailand and Uganda have pro-voked controversy. Such experiments aim to find a treat-ment that will cut the rate of vertical transmission moreefficiently than existing treatments like zidovudine. Thisscenario is first stated as generally as possible, beforethree ethical principles found in the Belmont Report,itself a sharpening of the Helsinki Declaration, are stated.These three principles are the Principle of Utility, thePrinciple of Autonomy and the Principle of Justice.These are taken as voices of moral imperative. Butalthough each has intuitive appeal, it can be shown that


::AB

STRA

CTS

::SU

MMAR

IES

::AB

STRA

CTS

::SU

MMAR

IES

::

there are possible scenarios in which they give conflictingprescriptions. To achieve consistency, one must be subor-dinate to the others. The voice of utility is taken as sub-ordinate to those of justice and autonomy and it isshown that given plausible assumptions about the levelof poverty and education in the developing country tar-geted, the experiment is ruled morally wrong in thename of both justice and autonomy.Moreover, it is argued that no justification can be foundfor the inclusion of a placebo group, when strictlydefined. By contrast, a ‘notreatment’ control arm mightbe justified, but only when the demands of autonomyare satisfied, demands that are more stringent than theymight appear. A utilitarian defense of the experiment isexamined, namely that the would-be participants are in ano-loss situation, and it is shown that this defense is seri-ously flawed. Finally, it is concluded that there is no jus-tification for amending the Declaration of Helsinki. ©

a b

S P A N D A J O U R N A L I I , 1 /2 0 1 1 | t h e p l a c e b o e f f e c t

EDITOR SAHLAN MOMO

EDITORIAL BOARD REBECCA BENTLEY,ALESSANDRO COLOMBO, CLAUDE LORRAINE,SAHLAN MOMO, LIAM OSBOURNE,BETHANY POLLER, JASPER SCHELLEKENS.EDITORIAL AND EXECUTIVE OFFICE

LAAN VAN MEERDERVOORT 70

NL~2517AN THE HAGUE | THE NETHERLANDS

T +31 (0) 70 362 6523 | F +31 (0) 70 362 9848

[email protected] | WWW.SPANDA.ORG

PUBLISHER SPANDA PUBLISHING

© MMXI SPANDA FOUNDATION

The Spanda Journal, the newsletter Spanda Monitor,and the Spanda website are the official publications ofthe Foundation. Spanda does not accept responsibility forthe views expressed in any writing, signed or unsigned,which appears in its pages: what it does accept is theresponsibility for giving them a chance to appear here. Spanda Journal makes every possible effort to ensurethat the information in articles is accurate. To reporterrors requiring correction or clarification, please [email protected] reasonable attempt has been made to identify theowners of pictures’ copyrights. Errors or omissions willbe corrected in subsequent issues.Neither the Spanda Foundation nor any person acting onher behalf is responsible for the way in which informationcontained in this publication may be used.

�

CALL FOR CONTRIBUTIONS You are warmly encouragedto contribute articles to the Spanda Journal. Help usmake this Journal with your pictures, views and stories.Guidelines for contributors GO!

Submissions to Spanda Journal may be edited for a varietyof reasons including the need to shorten them or improveexpression.QUESTIONS AND COMMENTS? GO!

SEND THIS PUBLICATION TO A FRIEND GO!

TO SUBSCRIBE [email protected] GO!

NEXT ISSUE «Human Subjects Research».

N | C O L O P H O N

year II, no. 1 ~ janu ary/april 2011

ISSN 2210-2175

s p a n d a j o u r n a lB A C k I S S U E S

Available at www.spanda.org/publications.html GO!

�

VOL. I, 1. 2007 ~ Multicultural Youth.VOL. I, 2. 2007 ~ Water Wise.VOL. I, 3. 2007 ~ Gender.VOL. I, 4. 2007 ~ Death Penalty.VOL. II, 1. 2008 ~ Hunger. A Clim(c)tic Perspective. VOL. II, 2. 2008 ~ Education & Development. Africa. VOL. II, 3. 2008 ~ Indegeneous Knowledge (IK) &

Environment.VOL. II, 4. 2008 ~ Consciousness & Development.VOL. III, 1. 2009 ~ Human Rights & Security.

N E W S E R I E S

VOL. I, 1. 2010 ~ Energy & Development.VOL. I, 2. 2010 ~ Microfinance.VOL. II, 1. 2011 ~ The Placebo Effect.

ISSN 2210-2175

9772210217004

03

a b

Some circumstantial evidence is very strong,as when you find a trout in the milk.

HENRY DAVID THOUREAU, Journal, 11 November 1850.

a b

http://www.spanda.org

http://www.spanda.org/SpandaNews_Guidelines.pdf

mailto:[email protected]?subject=Questions

mailto:?subject=Take a look at the Spanda Journal

mailto:[email protected]?subject=Subscribe to Spanda Journal

T H E PLA CEBO - Spanda Foundationspanda.org/SpandaJournal_II,1.pdf · · 2011-02-23quarterly of...

Documents

Transcript of T H E PLA CEBO - Spanda Foundationspanda.org/SpandaJournal_II,1.pdf · · 2011-02-23quarterly of...