C:iToolsWMSTandF-Journals4521288WorkingFolderNHYP A … · chest pain, orofacial pain, osteoarthritis pain, spinal cord injury, temporomandibular disorders, and other forms of chronic

Intl. Journal of Clinical and Experimental Hypnosis, 62(1): 1–28, 2014Copyright © International Journal of Clinical and Experimental HypnosisISSN: 0020-7144 print / 1744-5183 onlineDOI: 10.1080/00207144.2013.841471

A META-ANALYSIS OF HYPNOSIS FORCHRONIC PAIN PROBLEMS:

A Comparison Between Hypnosis, Standard Care,and Other Psychological Interventions

Tomonori Adachi, Haruo Fujino, Aya Nakae, Takashi Mashimo,and Jun Sasaki1

Osaka University, Japan

Abstract: Hypnosis is regarded as an effective treatment forpsychological and physical ailments. However, its efficacy as a strategyfor managing chronic pain has not been assessed through meta-ana-lytical methods. The objective of the current study was to conduct ameta-analysis to assess the efficacy of hypnosis for managing chronicpain. When compared with standard care, hypnosis provided mod-erate treatment benefit. Hypnosis also showed a moderate superioreffect as compared to other psychological interventions for a non-headache group. The results suggest that hypnosis is efficacious formanaging chronic pain. Given that large heterogeneity among theincluded studies was identified, the nature of hypnosis treatment isfurther discussed.

Many people are suffering from chronic pain worldwide (Verhaak,Kerssens, Dekker, Sorbi, & Bensing, 1998). Chronic pain not only per-sonally impacts those individuals but is also associated with greateconomic cost (i.e., health care costs) inflicted by such conditions (Turk,2002). The United States Congress passed into law a provision declar-ing the 10-year period beginning in January 1, 2001, as the Decadeof Pain Control and Research (Lippe, 2000). This legislation reflectsthe considerable societal drain that can be associated with chronicpain. While chronic pain results in severe external consequences forthe self and society, psychological factors can exacerbate the experi-ence of chronic pain (e.g., catastrophizing, depression, and fear; Cook,Brawer, & Vowles, 2006; Jensen, Turner, & Romano, 2001; Spinhovenet al., 2004; Sullivan et al., 2001, Vlaeyen, Kole-Snijders, Boeren, & vanEek, 1995). Psychotherapy is widely used to help alleviate chronic pain.

Manuscript submitted August 23, 2012; final revision accepted November 20, 2012.1Address correspondence to Tomonori Adachi, Shinri-Kyouiku Soudansitsu,

Graduate School of Human Sciences, Osaka University, 1-2 Yamadaoka, Suita, Osaka,565-0871, Japan. E-mail: [email protected]

1

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

2 TOMONORI ADACHI ET AL.

Cognitive-behavioral and hypnotic approaches are two strategies thatappear to be rather popular and effective (Kisley, Campbell, Skerritt, &Yelland, 2010).

Hypnosis has long been shown to be an effective psychological treat-ment (Melzack & Wall, 1982). Several brain-imaging studies assessinghypnotic analgesia have received a fair amount of research atten-tion (Abrahamsen et al., 2010; Faymonville et al., 2000; Hofbauer,Rainville, Duncan, & Bushnell, 2001; Jensen, 2010; Rainville, Duncan,Price, Carrier, & Bushnell, 1997).

Although some individual studies have shown the efficacy of hyp-nosis for intractable chronic pain (Haanen et al., 1991; Jensen Barber,Romano, Hanley, et al., 2009; Jensen Barber, Romano, Molton, et al.,2009; Muraoka, Komiyama, Hosoi, Mine, & Kubo, 1996), the fieldwould benefit from efforts to summarize results of individual inter-ventions to determine the overall efficacy of hypnotic treatment. Onlya handful of studies have reported the overall efficacy of hypnosisfor chronic pain (Jensen & Patterson, 2006; Montgomery, Duhamel, &Redd, 2000; Patterson & Jensen, 2003). While Jensen and Patterson, andPatterson and Jensen, provide a narrative review of the literature, onlyMontgomery et al. have provided a systematic review. Montgomeryand colleagues’ review is of great utility, because it summarizes theresults of individual interventions for clinical and experimental painusing a meta-analysis.

Montgomery et al. (2000) investigated the effects of hypnosis forclinical pain including burns, coronary disease, cancer, headache, andexperimental pain inductions (i.e., cold pressor, ischemic pain, andfocal pressure). Statistical analysis was conducted on 18 publishedstudies including samples totaling 933 participants. Calculated effectsizes revealed that hypnosis had a large effect (d = 0.80) in manag-ing clinical pain and a moderate-to-large effect (d = 0.70) for managingexperimental pain.

However, Montgomery et al. (2000) included acute and experimen-tal pain within a broader scope. A meta-analytical study reporting theefficacy of hypnosis that narrows the focus to only chronic pain has notbeen previously conducted. Thus, we conducted a systematic review toinvestigate the efficacy of hypnosis for managing chronic pain.

Method

Search StrategyWe searched three electronic databases: Cochrane Central Register

of Controlled Trials (CENTRAL), MEDLINE, and PsycINFO for studiespublished on or before June 29, 2011. Searches were carried out using acombination of the following keywords: pain, hypnosis, and clinical trial.

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

HYPNOSIS FOR CHRONIC PAIN META-ANALYSIS 3

The search strategy for the MEDLINE database is listed in section 1 ofthe Appendix. This approach yielded 130 articles.

For CENTRAL and PsycINFO, we basically used the Medical SubjectHeading and subject search strategies. We also used a keyword searchto complement the former strategies. Hypnosis and pain were used asthe search terms (see the Appendix, sections 2 and 3). This approachyielded 228 articles from CENTRAL and 167 articles from PsycINFO.Moreover, the reference lists of previous review articles (Jensen &Patterson, 2006; Montgomery et al., 2000; Patterson & Jensen, 2003) werescanned to obtain three additional studies.

Inclusion and Exclusion CriteriaOur inclusion criteria were as follows: (a) randomized controlled tri-

als (RCTs) or controlled clinical trials as experimental design; (b) trialstargeting only chronic pain problems; (c) application of hypnosis as themain therapeutic intervention; (d) inclusion of a control group (groupsreceiving both a standard care and other psychological interventions);(e) pain or pain intensity as the main outcome; (f) studies written inEnglish; (g) published journal articles; and (h) studies containing datawhere effect sizes could be calculated, such as means and standarddeviations.

Exclusion criteria were as follows: (a) application of hypnotic-likeinterventions, such as autogenic training (AT) or progressive mus-cle relaxation (PMR) as the main interventions; (b) absence of datathat could be used for calculating effect sizes; (c) studies written in alanguage other than English; (d) use of a case report and/or a narra-tive review as the study method; and (e) unpublished work, doctoraldissertations, published abstracts, books, letters, commentaries, andeditorials.

Assessment of Methodological QualityMethodological quality was assessed with a scale for assessing qual-

ity of psychological trials for treating pain developed by Yates, Morley,Eccleston, and Williams (2005). This scale provides an overall total score(0 to 35) that consists of two subscales. These include a treatment quality(0 to 9) and a design and methods scale (0 to 26). The first and secondauthors (TA and HF) scored all of the studies. TA and HF discussedtheir disagreements in ratings regarding the coding notes of this scale.Each disagreement was discussed until consensus was reached.

Data AnalysisHedges’s g (Hedges & Olkin, 1984) and 95% confidence intervals

were calculated. Q statistics and I2 tests were used to measure statisti-cal heterogeneity; a Q statistic p value of greater than .1 and an I2 value

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


of less than 30% indicated homogeneity (Higgins & Green, 2011). Weused random effects models in all analyses. We examined the effects ofinterventions based on the points of evaluation (i.e., postinterventionor at follow-up) and the type of control groups (standard care or otherpsychological interventions). Calculations were performed using MIX2.0 computer software, and analyses were conducted using SPSS 17.0.

Results

Characteristics of Included StudiesLiterature search. Excluding 53 duplicate articles across three

databases, and two duplicates within the CENTRAL database, 473 rel-evant articles were identified in our initial search. The first author (TA)reviewed these articles and rejected 391 articles on titles and abstracts.TA reviewed full texts of the remaining 82 articles and evaluated themin detail. Finally, only 12 clinical studies met our inclusion criteria.Of these, six were relevant RCTs and six were relevant clinical trials.

Study quality. For the 12 included studies, the mean overall qualityof the studies was 15.00 (SD = 3.13, range: 11 to 21). The mean designquality was 11.00 (SD = 2.30, range: 6 to 15), and the mean treatmentquality was 4.00 (SD = 1.65, range: 2 to 7). When rating design qual-ity, items on an allocation bias and a power calculation were scorelesswithin all included studies.

A Spearman’s rank correlation was calculated to investigate theassociation between the year of publication and treatment qualityscore, design quality score, and total quality score (Eccleston, Palermo,Williams, Lewandowski, & Morley, 2009; Eccleston, Williams, & Morley,2009). This was also calculated between the N at the end of treatmentand three treatment quality parameters. Treatment quality, design qual-ity, and total quality were not associated with the year of publication(treatment, r =.09, ns; design, r =.42, ns; total, r = .33, ns). The N at theend of treatment also was not associated with the three methodologicalqualities (treatment, r = -.09, ns; design, r =-.05, ns; total, r = .05, ns).

Participants. The total number of participants within each studyranged from 22 to 157 (M = 55.75, SD = 37.78). Types of chronicpain included fibromyalgia, headache (i.e., tension-type headache andmigraine), irritable bowel syndrome, multiple sclerosis, noncardiacchest pain, orofacial pain, osteoarthritis pain, spinal cord injury,temporomandibular disorders, and other forms of chronic pain). Onlyfour studies reported participants’ mean pain duration; among these,mean pain duration ranged from 9.5 years to 13.7 years (M = 11.53,SD = 1.76). The mean ages of the study participants were reported in

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


8 of 12 studies. Ages ranged from 36 to 64.7 years (M = 47.15, SD =10.24). The percentage of men in the samples across the studies includedin the review ranged from 0% to 76%. The ratio of females exceededmales within 9 studies. A diagnosis of specific illness, degree of pain,degree of a cognitive impairment, and the presence of a mental disor-der was given as inclusion or exclusion criteria. Characteristics of studyparticipants are summarized in Table 1.

Interventions. The main intervention was hypnosis, because ourmajor interest was to clarify the efficacy of hypnotic treatment. Thenumber of interventions varied from 3 to 12 sessions; the duration ofa single session lasted between 30 to 90 minutes. Eleven studies usedindividual treatment formats, and only one study used a group treat-ment format. Only one study just compared hypnosis with a wait-listcontrol, and eight studies compared hypnosis with other psycholog-ical interventions. Three other studies compared hypnosis with bothtypes of control groups. Standard care included no treatment, treat-ment as usual, and a wait-list control. Other psychological interventionsincluded autogenic training, biofeedback, cognitive-behavior therapy(CBT), guided imagery, progressive muscle relaxation, and supportivepsychotherapy (see Table 2).

Measures. Although pain or pain intensity was the main outcome forinclusion criteria, all included studies measured pain or pain intensityusing a variety of scales. Most studies used the numerical rating scale(NRS) or the visual analogue scale (VAS) to quantify pain. Eight stud-ies calculated pain scores to tally the NRS or used 5- to 11-point Likertscales. Most studies used a questionnaire that assessed psychologicalsymptoms (e.g., the Self-Rating Depression Scale, Zung, 1965; the State-Trait Anxiety Inventory, Spielberger, Gorsuch, & Lushene, 1970; the90-item version of Symptom Check List, Derogatis, Lipman, & Covi,1973). Eight studies measured hypnotizability using several differentscales, and four studies measured treatment expectancy. Other outcomeindicators included pain interference, pain coping strategy, quality oflife, sleep quality, perceived control over pain, and capacity for mentalimagery (see Table 2).

Efficacy of the InterventionsA comparison between hypnosis and a standard care. Table 3

displays results for the postintervention comparisons between hyp-nosis and standard care. We pooled the four studies providingpostintervention data allowing us to conduct this comparison. Theeffect size revealed that hypnosis had a significant and moderate effect,g = .60, 95% CI: 0.03–1.17, p < .05, on treatment efficacy compared tostandard care. However, heterogeneity was quite large, Q = 7.03, p <

.10; I2 = 57.30%.

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e1

Cha

ract

eris

tics

ofSt

udy

Par

tici

pant

s

Stud

yD

iagn

osis

Pain

dur

atio

n(y

ears

)N

Nat

the

end

oftr

eatm

ent

Mea

nag

e(R

ange

)G

end

er%

(M:F

)

Abr

aham

sen

etal

.(2

009)

Tem

poro

man

dib

ular

Dis

ord

ers

11.9

4340

−0

:100

Abr

aham

sen

etal

.(2

008)

Pers

iste

ntId

iopa

thic

Oro

faci

alPa

in9.

541

4156

(–)

15:8

5

Cas

tele

tal.

(200

9)Fi

brom

yalg

ia11

4739

44.2

(–)

5:9

5G

ayet

al.(

2002

)O

steo

arth

riti

sPa

in13

.71

4136

64.7

(–)

8:9

2Je

nsen

,Bar

ber,

Rom

ano,

Han

ley,

etal

.(20

09)

Chr

onic

Pain

inPe

rson

sW

ith

Spin

al-C

ord

Inju

ry−

3728

49.5

(19-

70)

76:2

4

Jens

en,B

arbe

r,R

oman

o,M

olto

n,et

al.(

2009

)

Mul

tipl

eSc

lero

sis

and

Chr

onic

Pain

−22

2251

.7(2

7–75

)27

:73

Jone

set

al.(

2006

)N

onca

rdia

cC

hest

Pain

−28

26−

−Pa

lsso

net

al.(

2002

)Ir

rita

ble

Bow

elSy

ndro

me

−30

2439

.1(–

)38

:62

Spin

hove

net

al.(

1992

)Te

nsio

nH

ead

ache

−56

4636

(20–

62)

39:6

1te

rK

uile

etal

.(19

94)

Chr

onic

Hea

dac

hes

−15

714

6−

−va

nD

yck

etal

.(19

91)

Tens

ion

Hea

dac

he−

7155

36(1

8-60

)49

:51

Zit

man

etal

.(19

92)

Tens

ion

Hea

dac

hes

−96

79−

46:5

4

6

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e2

Cha

ract

eris

tics

ofIn

clud

edSt

udie

s

Stud

yR

CT

Inte

rven

tion

s

Sett

ings

ofH

ypno

tic

Inte

rven

tion

Ass

essm

ent

Poin

tM

easu

res

Pain

Ind

icat

ors

Res

ults

Abr

aham

sen

etal

.(2

009)

Yes

Hyp

nosi

sR

elax

atio

nan

dvi

sual

izat

ion

ofa

com

fort

able

,saf

epl

ace

60m

in.×

4se

ssio

ns;N

od

escr

ipti

ons

abou

tfre

quen

cy;

Ind

ivid

ual;

HW

=lis

teni

ngto

ata

ped

inst

ruct

ion

Pre,

Wee

k3

(Pos

t)N

RS,

MPQ

,CSQ

,SC

L–6

0,PS

QI,

HG

SHS–

A

Pain

dia

ry(t

heav

erag

ed

aily

NR

Spa

insc

ores

ofth

e6

day

sin

each

ofth

efo

urti

me

peri

ods)

∗

Hyp

nosi

s<

Rel

axat

ion

[Pos

t]

Abr

aham

sen

etal

.(2

008)

Yes

Hyp

nosi

sR

elax

atio

nan

dvi

sual

izin

ga

nice

safe

plac

e

5.1

±0.

8(r

ange

3–6)

sess

ions

ofhy

pnos

is;5

.3±

0.9

sess

ions

(ran

ge3–

6)of

cont

rol;

5se

ssio

nsw

ere

plan

ned

;No

des

crip

tion

sab

outf

requ

ency

;H

W=

liste

ning

toa

tape

din

stru

ctio

n;In

div

idua

l

Pre,

Wee

k4

(Pos

t)V

AS

(Pai

nd

iary

),M

PQ,P

SQI,

SCL

–60,

SF–3

6,C

SQ,S

HC

S

MPQ

–Tot

alN

osi

gnifi

cant

dif

fere

nces

betw

een

trea

tmen

tgro

ups

[Pos

t]

(Con

tinu

ed)

7

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e2

(Con

tinu

ed)

Stud

yR

CT

Inte

rven

tion

s

Sett

ings

ofH

ypno

tic

Inte

rven

tion

Ass

essm

ent

Poin

tM

easu

res

Pain

Ind

icat

ors

Res

ults

Cas

tele

tal.

(200

9)Ye

sH

ypno

sis

+C

ogni

tive

Beh

avio

rT

hera

pyC

ogni

tive

Beh

avio

rT

hera

pyTr

eatm

ent

As

Usu

al

90m

in.×

12se

ssio

ns;N

od

escr

ipti

ons

abou

tfre

quen

cy;

HW

=lis

teni

nga

tape

din

stru

ctio

n;G

roup

Pre,

Aft

er12

ses-

sion

s(P

ost)

NR

S,FI

Q,M

PQ,

HG

SHS–

AU

sual

pain

inte

nsit

y,co

mpo

site

(NR

S)∗

No

sign

ifica

ntd

iffe

renc

esbe

twee

ntr

eatm

entg

roup

s[P

ost]

Gay

etal

.(2

002)

Yes

Hyp

nosi

sJa

cobs

on’s

Prog

ress

ive

Mus

cle

Rel

axat

ion

No

trea

tmen

t

30m

in.×

8se

ssio

ns;

Wee

kly;

No

des

crip

tion

sab

outH

W;

Ind

ivid

ual

Pre,

wee

k4,

wee

k8

(Pos

t),

3m

onth

(Fol

low

-up

),6

mon

th(F

ollo

w-

up)

VA

S,ST

AI,

SDS,

Imag

ery

vivi

dne

ss(d

eriv

edfr

omSh

eeha

n’s

Que

stio

nnai

reof

Men

tal

Imag

ery,

0-4

poin

ts),

SHSS

–C,

apr

iori

belie

fin

the

trea

tmen

tef

ficac

y(3

-poi

nts

scal

e)

Pain

(VA

S)H

ypno

sis

>N

otr

eatm

ent[

Post

]PM

R>

No

trea

tmen

t[Po

st]

Hyp

nosi

s>

No

trea

tmen

t[3

mon

thFU

]

8

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Jens

en,

Bar

ber,

Rom

ano,

Han

ley,

etal

.(2

009)

Yes

Hyp

nosi

sB

iofe

edba

ck10

sess

ions

;Var

iant

freq

uenc

y;H

W=

liste

ning

toa

tape

din

stru

ctio

nat

leas

tonc

ea

day

;Ind

ivid

ual

Pre,

Aft

er10

ses-

sion

s(P

ost)

,3

mon

th(F

ollo

w-

up)

NR

S,T

ES,

BPI

,SH

CS,

SOPA

,C

ES–

D

Dai

lyav

erag

epa

in(N

RS)

∗H

ypno

sis

>

Bio

feed

back

[Pos

t]

Jens

en,

Bar

ber,

Rom

ano,

Mol

ton,

etal

.(2

009)

No

Hyp

nosi

sPr

ogre

ssiv

eM

uscl

eR

elax

atio

n10

sess

ions

;No

des

crip

tion

sab

outf

requ

ency

;In

div

idua

l;H

W=

liste

ning

toa

tape

din

stru

ctio

nat

leas

tonc

ea

day

;Ind

ivid

ual

Pre,

Aft

er10

ses-

sion

s(P

ost)

,3

mon

th(F

ollo

w-

up)

NR

S,T

ES,

BPI

,SH

CS

Dai

lypa

inin

tens

ity

com

posi

te(N

RS)

∗

Hyp

nosi

s>

PMR

[Pos

t],H

ypno

sis

>

PMR

[FU

]

Jone

set

al.

(200

6)Ye

sH

ypno

ther

apy

Supp

orti

veps

ycho

ther

apy

30m

in.×

12se

ssio

ns;

Var

iant

freq

uenc

y;H

W=

liste

ning

toa

tape

din

stru

ctio

n;In

div

idua

l

Bas

elin

e,W

eek

17(P

ost)

agl

obal

asse

ssm

ento

fch

estp

ain,

agl

obal

asse

ssm

ento

fw

ell-

bein

g,M

acN

ewQ

OL

inst

rum

ent,

Lin

ear

anal

ogue

Scal

e(P

ain

seve

rity

0–10

0),H

AD

Pain

seve

rity

(Lin

eran

alog

uesc

ale,

0-10

0)

Hyp

nosi

s>

Supp

orti

veps

ycho

ther

apy

[Pos

t]

(Con

tinu

ed)

9

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e2

(Con

tinu

ed)

Stud

yR

CT

Inte

rven

tion

s

Sett

ings

ofH

ypno

tic

Inte

rven

tion

Ass

essm

ent

Poin

tM

easu

res

Pain

Ind

icat

ors

Res

ults

Pals

son

etal

.(2

002)

No

Hyp

nosi

sW

aiti

nglis

t45

min

.×7

sess

ions

;B

iwee

kly

(ove

r12

wee

ks);

HW

=lis

teni

ngto

ata

ped

inst

ruct

ion

dai

ly;I

ndiv

idua

l

Pre,

Wee

k2

(Pos

t),

4m

onth

(Fol

low

-up

)

the

Sym

ptom

dia

ry(b

owel

mov

emen

t:4-

poin

t,th

ew

orst

epis

ode

ofab

dom

inal

pain

&bl

oati

ng:

5-po

int)

,BD

I,SC

L–9

0R,S

PSI

The

seve

rity

ofth

ew

orst

epis

ode

ofab

dom

inal

pain

(5-p

oint

scal

epe

rd

ay;1

4d

ays

sym

ptom

sum

scor

es)

∗

Hyp

nosi

s>

Wai

ting

list[

Post

]

Spin

hove

net

al.

(199

2)

No

Self

-hyp

nosi

sA

utog

enic

Trai

ning

45m

in.×

4se

ssio

ns;

Biw

eekl

y;H

W=

liste

ning

toa

15-m

in.t

aped

inst

ruct

ion

twic

ed

aily

;Ind

ivid

ual

Pre,

Wee

k16

(Pos

t),

6m

onth

(Fol

low

-up

)

Hea

dac

heIn

dex

,G

loba

lPai

nra

ting

,SC

L–9

0,C

SQ

Pain

inte

nsit

y(H

ead

ache

Ind

ex)∗

No

sign

ifica

ntd

iffe

renc

esbe

twee

ntr

eatm

entg

roup

s[B

oth

Post

&FU

]

10

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

ter

Kui

leet

al.

(199

4)

No

Cog

niti

vese

lf-h

ypno

sis

Aut

ogen

icTr

aini

ngW

aiti

nglis

t

60m

in.×

7se

ssio

ns;

Wee

kly;

HW

=lis

teni

ngto

a15

-min

.tap

edin

stru

ctio

ntw

ice

dai

ly;I

ndiv

idua

l

Pre,

Wee

k8

(Pos

t),

Wee

k35

or6

mon

th(F

ollo

w-

up)

Hea

dac

heIn

dex

(wei

ghed

6-po

intL

iker

tsc

ale)

,SC

L–9

0,SH

CS,

Hea

dac

heC

hara

cter

isti

csQ

uest

ionn

aire

,M

igra

ine

Ind

ex,

Trea

tmen

tE

xpec

tati

on(0

–200

%sc

ale)

Pain

inte

nsit

y(H

ead

ache

Ind

ex)∗

AT

>W

aiti

nglis

t[P

ost]

van

Dyc

ket

al.

(199

1)

No

Futu

re-o

rien

ted

hypn

otic

imag

ery

Aut

ogen

icTr

aini

ng

Tota

l150

min

.;4

sess

ion;

Var

iant

freq

uenc

y;H

W=

liste

ning

tota

ped

inst

ruct

ions

tota

l25

h(1

500

min

.);In

div

idua

l

Pre,

8w

eeks

(Pos

t)H

ead

ache

Dia

ry,

STA

I,SD

S,C

IS,

SHC

S

Pain

inte

nsit

y(H

ead

ache

Ind

ex)∗

No

sign

ifica

ntd

iffe

renc

esbe

twee

ntr

eatm

entg

roup

s[P

ost]

(Con

tinu

ed)

11

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e2

(Con

tinu

ed)

Stud

yR

CT

Inte

rven

tion

s

Sett

ings

ofH

ypno

tic

Inte

rven

tion

Ass

essm

ent

Poin

tM

easu

res

Pain

Ind

icat

ors

Res

ults

Zit

man

etal

.(1

992)

No

Futu

re-o

rien

ted

hypn

otic

imag

ery

Futu

reor

ient

ed-i

mag

ery

Aut

ogen

icTr

aini

ng

Tota

l300

min

.;8

sess

ion;

Var

iant

freq

uenc

y;H

W=

liste

ning

tota

ped

inst

ruct

ions

tota

ling

49h

(294

0m

in.);

Ind

ivid

ual

Pre,

Wee

k8

(Pos

t),

6m

onth

(Fol

low

-up

)

Hea

dac

heD

iary

,ST

AI,

SDS,

VA

S(p

erce

ived

cred

ibili

tyof

the

trea

tmen

t;B

orko

vec

&N

au,1

972)

,the

neur

otic

ism

scor

eof

the

Dut

chPe

rson

alit

yQ

uest

ionn

aire

base

don

the

Cal

ifor

nia

Psyc

holo

gica

lIn

vent

ory

Pain

inte

nsit

y(H

ead

ache

Ind

ex)∗

Hyp

nosi

s>

AT

[FU

]

Not

e.T

heas

teri

sks

ofpa

inin

dic

ator

sre

fer

tost

udie

sca

lcul

atin

gpa

inin

tens

ity.

BD

I=

Bec

kD

epre

ssio

nIn

vent

ory;

BPI

=B

rief

Pain

Inve

ntor

y;C

ES–

D=

Cen

ter

for

Epi

dem

iolo

gic

Stud

ies

Dep

ress

ion

scal

e;C

IS=

Cre

ativ

eIm

agin

atio

nSc

ale;

CSQ

=C

opin

gSt

rate

gies

Que

stio

nnai

re;F

IQ=

Fibr

omya

lgia

Impa

ctQ

uest

ionn

aire

;FU

=Fo

llow

-Up;

HA

D=

Hos

pita

lanx

iety

and

dep

ress

ion

scal

e;H

GSH

S–A

=H

arva

rdG

roup

Scal

eof

Hyp

noti

cSu

scep

tibi

lity,

Form

A;H

W=

Hom

ewor

k;M

acN

ewQ

OL

inst

rum

ent=

Mac

New

Hea

rtD

isea

seH

ealt

h-R

elat

edQ

ualit

yof

Lif

eQ

uest

ionn

aire

;MPQ

=M

cGill

Pain

Que

stio

nnai

re;N

RS

=N

umer

ical

Rat

ing

Scal

e;PS

QI=

Pitt

sbur

ghSl

eep

Qua

lity

Ind

ex;S

CL

–60

=60

-ite

mve

rsio

nof

the

Sym

ptom

Che

ckL

ist;

SCL

–90

=90

-ite

mve

rsio

nof

the

Sym

ptom

Che

ckL

ist;

SDS

=Se

lf-r

atin

gD

epre

ssio

nSc

ale;

SF–3

6=

Med

ical

Out

com

esSt

udy

36-I

tem

Shor

t-Fo

rmH

ealt

hSu

rvey

;SH

CS

=St

anfo

rdH

ypno

tic

Clin

ical

Scal

e;SH

SS–C

=St

anfo

rdH

ypno

tic

Susc

epti

bilit

ySc

ale,

Form

C;

SOPA

=Su

rvey

ofPa

inA

ttit

udes

;SP

SI=

Stre

ss-r

elat

edPh

ysic

alSy

mpt

oms

Inve

ntor

y;ST

AI=

Stat

e-Tr

aitA

nxie

tyIn

vent

ory;

TE

S=

Trea

tmen

tExp

ecta

ncy

Scal

e;V

AS

=V

isua

lAna

logu

eSc

ale.

12

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


Only Gay, Philippot, and Luminet (2002) allowed for a comparisonbetween hypnosis and standard care at follow-up. Thus, we could notcalculate a difference in intervention efficacy at follow-up.

A comparison between hypnosis and other psychological interven-tions. We next examined the results for a comparison between hyp-nosis and other psychological interventions. Table 4 displays resultsfor the postintervention comparisons between hypnosis and other psy-chological interventions Table 5 also displays results for the follow-upcomparisons We pooled 11 studies that provided postintervention dataand six studies that provided follow-up data. These comparisons werenot significant, and effect sizes did not show a superior effect for hyp-nosis (postintervention, g = .04, 95% CI: –0.22–0.30, ns; follow-up, g =-.05, 95% CI: –0.33–0.23, ns). Heterogeneities were moderate (follow-up,Q = 10.02, ns; I2 = 30.16%) to large (postintervention, Q = 22.21, p < .05;I2 = 50.47%). These results indicate that the efficacy of hypnosis was notdifferent from that of other psychological interventions.

Comparing within a diagnostic group (headache or nonheadache).Furthermore, we explored the efficacy of hypnosis with respect tospecific diagnostic groups (headache group or nonheadache group)and the type of interventions. Jensen Barber, Romano, Hanley, et al.(2009) reported that pain intensity within a hypnosis group was sig-nificantly stronger than within a control group during a pretreatmentphase. We excluded Jensen, Barber, Romano, Hanley, et al. (2009) fromsubsequent analyses in order to more accurately discuss the efficacy ofhypnosis.

Within the headache group, we pooled four studies that pro-vided postintervention data and three studies that provided follow-up data. These comparisons did not produce significant results(postintervention, g = –.21, 95% CI: –0.44–0.03, ns; follow-up,g = –.06, 95% CI: –0.34–0.22, ns). However, the effect sizes revealeda small effect for other psychological interventions being more effica-cious than hypnosis at postintervention. Heterogeneities were relativelysmall (postintervention, Q = 2.28, ns; I2 = 0.00%; follow-up, Q = 2.26,ns; I2 = 0.00%).

Within the nonheadache group, we pooled six studies that providedpostintervention data and two studies that provided follow-up data.Effect sizes indicated a significant moderate effect (postintervention, g =.46, 95% CI: 0.16–0.75, p < .01) and an insignificant small effect (follow-up, g = .27, 95% CI: –0.20–0.74, ns) of hypnosis. Heterogeneities werealso rather small within both comparisons (postintervention, Q = 1.54,ns; I2 = 0.00%; follow-up, Q = 1.30, ns; I2 = 0.00%).

Comparing the type of intervention. We also examined results forcomparisons between hypnosis and three psychological treatments:

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e3

Met

a-A

naly

sis

Com

pari

ngH

ypno

sis

Wit

hSt

anda

rdC

are

atP

osti

nter

vent

ion

Pha

se

Con

trol

Exp

erim

enta

l

Stud

yM

ean

SDN

Mea

nSD

NW

eigh

t(%

)g

ivt,

rand

om,9

5%C

I

Cas

tele

tal.

(200

9)7

1.01

75.

792.

0516

20.9

7%0.

64[−

0.27

,1.5

5]G

ayet

al.(

2002

)4.

231.

1410

1.85

1.65

1319

.71%

1.58

[0.6

1,2.

54]

Pals

son

etal

.(20

02)

16.8

3.6

912

.912

.39

1522

.94%

0.37

[−0.

46,1

.21]

ter

Kui

leet

al.(

1994

)25

.416

5322

.514

.840

36.3

8%0.

19[−

0.23

,.60

]To

tal(

95%

CI)

7984

100.

00%

0.60

[0.0

3,1.

17]

Not

e.iv

t=in

vers

e-va

rian

cew

ith

t2.H

eter

ogen

eity

:Q=

7.03

,df=

3(p

<.1

0):I

2=

57.3

0%;T

estf

orov

eral

leff

ect:

Z=

2.05

(p<

.05)

.

14

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e4

Met

a-A

naly

sis

Com

pari

ngH

ypno

sis

Wit

hO

ther

Psy

chol

ogic

alIn

terv

enti

ons

atP

osti

nter

vent

ion

Pha

se

Con

trol

Exp

erim

enta

l

Stud

yM

ean

SDN

Mea

nSD

NW

eigh

t(%

)g

ivt,

rand

om,9

5%C

I

Abr

aham

sen

etal

.(20

09)

3.9

1.5

202.

92.

420

8.46

%0.

49[−

0.14

,1.1

2]A

brah

amse

net

al.(

2008

)54

.129

.64

1937

.325

.822

8.48

%0.

60[−

0.03

,1.2

2]C

aste

leta

l.(2

009)

6.1

2.52

165.

792.

0516

7.64

%0.

13[−

0.56

,0.8

3]G

ayet

al.(

2002

)2.

371.

6213

1.85

1.65

136.

74%

0.31

[−0.

47,1

.08]

Jens

en,B

arbe

r,R

oman

o,H

anle

y,et

al.(

2009

)3.

361.

2811

5.09

1.92

206.

67%

−0.9

8[−

1.76

,−0.

20]

Jens

en,B

arbe

r,R

oman

o,M

olto

n,et

al.(

2009

)4.

131.

697

3.17

1.75

155.

44%

0.53

[−0.

38,1

.45]

Jone

set

al.(

2006

)47

.31

26.5

513

29.1

325

.31

156.

81%

0.68

[−0.

09,1

.45]

Spin

hove

net

al.(

1992

)2.

51.

423

32.

123

9.15

%−0

.28

[−0.

86,0

.31]

ter

Kui

leet

al.(

1994

)16

.212

.141

22.5

14.8

4011

.37%

−0.4

6[−

0.90

,−0.

02]

van

Dyc

ket

al.(

1991

)48

.630

.628

48.1

48.2

279.

94%

0.01

[−0.

52,0

.54]

Zit

man

etal

.(19

92)–

148

.630

.628

52.9

48.7

249.

68%

−0.1

1[−

0.65

,0.4

4]Z

itm

anet

al.(

1992

)–2

48.1

48.2

2752

.948

.724

9.61

%−0

.10

[−0.

65,0

.45]

Tota

l(95

%C

I)24

625

910

0.00

%0.

04[−

0.22

,0.3

0]

Not

e.iv

t=in

vers

e-va

rian

cew

ith

t2.H

eter

ogen

eity

:Q=

22.2

1,df

=11

(p<

.05)

:I2

=50

.47%

,Tes

tfor

over

alle

ffec

t:Z

=0.

28(n

s).

Hyp

nosi

sis

com

pare

dw

ith

auto

geni

ctr

aini

ngin

“Zit

man

etal

.(19

92)–

1”an

dw

ith

futu

re-o

rien

ted

imag

ery

in“Z

itm

anet

al.(

1992

)–2.

”

15

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

Tabl

e5

Met

a-A

naly

sis

Com

pari

ngH

ypno

sis

Wit

hO

ther

Psy

chol

ogic

alIn

terv

enti

ons

atFo

llow

-Up

Pha

se

Con

trol

Exp

erim

enta

l

Stud

yM

ean

SDN

Mea

nSD

NW

eigh

t(%

)g

ivt,

rand

om,9

5%C

I

Gay

etal

.(20

02)–

12.

751.

9113

1.66

1.49

139.

63%

0.62

[−0.

17,1

.41]

Gay

etal

.(20

02)–

22.

81.

6313

2.38

2.47

1310

.00%

0.19

[−0.

57,0

.97]

Jens

en,B

arbe

r,R

oman

o,H

anle

y,et

al.(

2009

)3.

131.

4211

4.9

2.13

209.

94%

−0.9

0[−

1.67

,−0.

13]

Jens

en,B

arbe

r,R

oman

o,M

olto

n,et

al.(

2009

)3.

351.

927

3.48

2.04

157.

87%

−0.0

6[−

0.96

,0.8

3]Sp

inho

ven

etal

.(19

92)

21.

720

2.5

2.6

2013

.63%

−0.2

2[−

0.85

,0.4

0]te

rK

uile

etal

.(19

94)

15.7

14.7

3719

.614

.235

19.4

7%−0

.27

[−0.

73,0

.20]

Zit

man

etal

.(19

92)–

150

.736

.621

41.6

51.8

2414

.72%

0.20

[−0.

39,0

.78]

Zit

man

etal

.(19

92)–

248

.839

.721

41.6

51.8

2414

.74%

0.15

[−0.

43,0

.74]

Tota

l(95

%C

I)14

316

410

0.00

%−0

.05

[−0.

33,0

.23]

Not

e.iv

t=in

vers

e-va

rian

cew

ith

t2.H

eter

ogen

eity

:Q=

10.0

2,df

=7

(ns)

:I2

=30

.16%

,Tes

tfor

over

alle

ffec

t:Z

=–0

.33

(ns)

.“G

ayet

al.

(200

2)–1

”m

eans

com

pari

son

betw

een

Hyp

nosi

sw

ith

PMR

at3-

mon

thfo

llow

-up

phas

e.“G

ayet

al.

(200

2)–2

”m

eans

com

pari

son

betw

een

Hyp

nosi

sw

ith

PMR

atth

e6-

mon

thfo

llow

-up

phas

e.H

ypno

sis

isco

mpa

red

wit

hau

toge

nic

trai

ning

in“Z

itm

anet

al.(

1992

)–1”

and

wit

hfu

ture

-ori

ente

dim

ager

yin

“Zit

man

etal

.(19

92)–

2.”

16

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


autogenic training (AT), guided imagery, and progressive muscle relax-ation (PMR). For AT, we pooled four studies with postinterventiondata and three studies with follow-up data. Comparisons with thesestudies were not significant (postintervention, g = –.23, 95% CI: –0.49–0.03, ns; follow-up, g = –.12, 95% CI: –0.44–0.19, ns), but AT wasslightly more effective than hypnosis. Heterogeneities were quite small(postintervention, Q = 2.09, ns; I2 = 0.00%; follow-up, Q = 1.61, ns; I2 =0.00%).

For guided imagery, we pooled three studies with postinterventiondata. This comparison with hypnosis was also insignificant(postintervention, g = .31, 95% CI: –0.13–0.74, ns), but the com-parison with guided imagery garnered a small-to-moderate effect infavor of hypnosis. The heterogeneity was moderate (postintervention,Q = 3.20, ns; I2 = 37.56%). Only Zitman, van Dyck, Spinhoven, andLinssen (1992) compared hypnosis with guided imagery at follow-up.Thus, we could not calculate a difference in efficacy between hypnosisand guided imagery at follow-up.

For PMR, we pooled two studies providing postintervention andfollow-up data. Both comparisons were insignificant (postintervention,g = .40, 95% CI: –0.19–0.99, ns; follow-up, g = .27, 95% CI: –0.20–0.74, ns).However, the effect sizes showed that hypnosis produced a small-to-moderate effect in comparison to PMR. Heterogeneities were relativelysmall (postintervention, Q = .14, ns; I2 = 0.00%; follow-up, Q = 1.30, ns;I2 = 0.00%).

Discussion

Main FindingsThe results showed that hypnosis was moderately effective for

managing chronic pain compared to a standard care during apostintervention phase. This result is consistent with those ofMontgomery et al. (2000). Montgomery and colleagues reported thathypnosis was moderately to largely efficacious for treatment of clin-ical pain and for ameliorating experimental pain (e.g., cold pressortest). Next, on the whole, our results revealed no differences in efficacybetween hypnosis and other psychological interventions during bothpostintervention and follow-up phases. According to our diagnosticgroups, hypnosis resulted in a significant moderate effect in compari-son to other psychological interventions within a nonheadache groupduring a postintervention phase. Further analysis suggested that hyp-nosis also resulted in superior efficacy during a follow-up phase. In theheadache group, hypnosis was no more efficacious than other psycho-logical interventions. Conversely, analysis of effect sizes suggested thatother psychological interventions were slightly more efficacious than

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


hypnosis during a postintervention phase. Based on treatment com-parison, effect size analysis indicated that AT had a slightly greatereffect when compared to hypnosis during a postintervention phase.However, hypnosis was slightly or moderately more effective thanguided imagery and PMR. Within these three comparison treatments,no significant differences emerged. Taken together, our results sug-gest that hypnosis is moderately more efficacious than standard careand other psychological interventions within nonheadache groups.Therefore, it appears that hypnosis can be an effective psychotherapyfor chronic pain.

Lunde, Nordhus, and Pallesen (2009) conducted a meta-analysis thatcompared CBT for chronic pain within an older adult sample and useda control group (a wait-list control and pretreatment group). This meta-analysis showed that CBT produced a small-to-moderate effect (d =0.47) during postintervention. In our analysis, hypnosis was moderatelyeffective as compared to standard care during a postintervention phase,and our effect size was greater than that obtained by Lunde et al. (2009).Thus, it is possible that hypnosis is more effective than CBT for manag-ing chronic pain when both hypnosis and CBT are compared to groupsthat do not conduct particular treatments.

Our results revealed no differences in efficacy between hypnosis andother psychological interventions for managing overall chronic pain.However, effect size analysis indicated that hypnosis was more effec-tive than other psychological interventions for a nonheadache group.Morley, Eccleston, and Williams (1999) and Eccleston, Williams, et al.(2009) investigated the efficacy of psychotherapies, especially CBT, formanaging chronic pain (excluding headache). These authors suggestedthat psychological treatments for headache are sufficiently differentfrom those used for other types of chronic pain. Morley et al. also saidthat “pain relief is a much more realistic result of treatment than inother chronic pain” (1999, p. 2). Morley and colleagues’ results showedthat CBT had a small, but greater, effect on managing chronic pain thanactive control groups receiving education, physiotherapy, occupationaltherapy, and treatment as usual (Morley et al., 1999, g = .29; Eccleston,Williams, et al., 2009; Standardized Mean Differences = –.14, a negativesign indicates a positive effect of CBT in their study). The effect sizeof our results exceeded those of Morley et al. and Eccleston, Williams,et al. when we compared hypnosis to other psychological interventionsin a nonheadache group during a postintervention. Thus, it is possiblethat hypnosis has superior efficacy to other psychological interventions,including CBT.

Although the differences were not significant, our results revealedthat other psychological interventions were more effective than hypno-sis for those suffering from headaches. During our systematic review,all studies investigating the efficacy of hypnosis for headache groups

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


had substantial overlap with studies comparing the efficacy of hyp-nosis versus AT. Four studies compared hypnosis with AT during apostintervention phase and three studies compared these two treat-ments during a follow-up phase. This suggests that AT is more effica-cious than hypnosis for managing headaches. However, Kanji, White,and Ernst (2006) reported that AT and hypnosis are similarly effica-cious for the treatment of headaches. A meta-analysis of controlledclinical trials reported by Stetter and Kupper (2002) showed that ATfor headaches was weakly inferior to other psychological interventions(postintervention, g = −.25; follow-up, g = −.26). As mentioned above,there were inconsistent results between our results when compared toprevious studies. We need to further investigate differences in efficacybetween hypnosis and other psychological interventions (especially AT)for the treatment of headaches.

Our results indicated that hypnosis was more effective than guidedimagery and PMR for managing chronic pain; however, differencesin the effectiveness of hypnosis were not statistically significant. Ourresults are intriguing given that studies comparing hypnosis with thesetwo interventions through a meta-analysis had not been conductedbefore. There are a few systematic reviews comparing guided imageryand PMR with standard care or a wait-list control (Henschke et al., 2011;Posadzki, Lewandowski, Terry, Ernst, & Stearns, 2012). However, thesereviews did not reveal effect sizes compatible with our results. Furtherexploration will be needed to clarify differences in efficacy betweenhypnosis and guided imagery or PMR.

TA and HF rated methodological quality using procedures describedin Yates et al. (2005). All methodological quality indices revealed lowerscores as compared to previous studies (Eccleston, Palermo, et al., 2009;Eccleston, Williams, et al., 2009). High limits within design quality werelow, and items regarding allocation bias and power calculation scoresequal to 0 were observed in all 12 studies. These scores do not indi-cate that authors of the 12 studies did not perform proper randomallocation and power calculations. These indices only suggest that ade-quate description as to the procedures for random allocation and resultsof the power calculations prior to starting participant recruitment foreach study were not reported in all 12 studies. Decorrelations betweenmethodological quality, the year of publication, and the N at the endof treatment were also recorded. Methodological qualities of clinicalhypnotic intervention studies have remained at low levels, which isinconsistent with the date and the scale of study (i.e., the number ofstudy samples).

In general, people are likely to view hypnosis as a suspect methodand mental state. Clinicians and researchers who use hypnosis arealways required to seek sound evidence for the efficacy of theirtreatment. Some guidelines for improving the methodological quality

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


of intervention studies have been suggested (i.e., the CONSORTguidelines; Begg et al., 1996; Moher et al., 2010; Moher, Schulz, &Altman, 2001). Controlling the researcher’s intent regarding the alloca-tion and power calculations is possible regardless of the vague natureof hypnosis. We suggest that procedures for random allocation and theresults of power analyses should be reported enough in future researcharticles. In future studies, researchers who seek sound evidence of hyp-notic efficacy should aspire to guarantee methodological quality to acertain targeted level in order to heighten the credibility of the results.

ImplicationsImplications for research. Although target diseases and study

designs are diverse, large heterogeneities reported in the current studyare related to the nature of hypnosis. There is ambiguity as to whethera hypnotic intervention from Study X and Study Y can specifically bedefined as “hypnosis.” Both studies might have different therapeuticmechanisms. Whether an altered state of consciousness is evoked by ahypnotic induction procedure is an area of controversy within researchon hypnosis. Using functional brain imaging, Rainville and colleaguesshowed that cerebral physiological changes are evoked by hypnosis(Hofbauer et al., 2001; Rainville, Bao, & Chrétien, 2005; Rainville et al.,1997). Whether an altered state of consciousness is evoked by a hypnoticinduction procedure needs future study. However, studies revealingthe nature of hypnosis from an empirical standpoint are important toextend the credibility of hypnotic techniques.

Although pain or pain intensity was defined as the main outcomein this study, studies have shown that hypnosis works well on boththe affective (Holroyd, 1996) and cognitive dimensions (Jensen et al.,2011) of pain. Perhaps we can better assess the multilateral efficacy ofhypnosis in order to define the unpleasantness of pain or a cognitivedimension of pain as the main outcome. Jensen and Patterson (2006)suggested, “the primary goal is not to alter pain during hypnosis, but tomake hypnotic suggestions and teach skills that will alter pain intensityand its impact throughout the patient’s daily life” (p. 96). Investigationof other outcomes beyond pain should be addressed in future studiesto better determine the efficacy of hypnosis.

Other psychological interventions assessed in our meta-analysisincluded AT, biofeedback, CBT, guided imagery, PMR, and support-ive psychotherapy. Some studies suggest that those interventions sharesimilar therapeutic components with hypnosis (e.g., relaxation andfocused attention; Gay et al., 2002; Jensen Barber, Romano, Hanley, et al.,2009; Jensen Barber, Romano, Molton, et al., 2009). AT is also regardedas a form of self-induced hypnotherapy (Schultz, 1987). Supportive psy-chotherapy reported by Jones, Cooper, Miller, Brooks, and Whorwell(2006) could be distinguished from hypnosis. Since the CBT used in

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


Castel, Salvat, Sala, and Rull (2009) included several interventions, itis difficult to limit one component as the main intervention utilized intheir study. We could argue that other psychological interventions sam-pled in our study are regarded as hypnotic-like interventions, exceptfor the supportive psychotherapy reported in Jones et al. (2006) and theCBT reported in Castel et al. (2009). Given that most psychological inter-ventions in our study are regarded as hypnotic-like interventions, weneed to further examine a comparison of the efficacy between hypnotic-like interventions and less hypnotic-like interventions, such as in vivoexposure (Boersma et al., 2004; Vlaeyen, de Jong, Geilen, Heuts, &van Breukelen, 2002) and operant conditioning (Fordyce, Fowler, &DeLateur, 1968; Fordyce et al., 1973; Lindstrom et al., 1992). Such studieswould help clarify the therapeutic mechanisms of clinical hypnosis.

Implications for practice. Our results have demonstrated the effi-cacy of hypnosis in clinical studies. Below, we describe a few sugges-tions regarding the active clinical use of hypnosis.

The essential aspects of our results in terms of active clinical use ofhypnosis include the following:

• When compared with nonspecific interventions including a wait-listcontrol and a treatment as usual, hypnosis shows good efficacy formanaging overall chronic pain;

• Hypnosis led to larger effect sizes when compared to other psychologicalinterventions, including CBT, for managing nonheadache chronic pain;

• It is unclear whether hypnosis is more effective than other psychologicalinterventions for managing headaches.

From these, hypnosis should be promoted as an effective psycholog-ical intervention for managing chronic pain. We should use hypno-sis actively for intractable chronic pain, such as nonheadache pain.However, we do not have strong evidence for preferentially usinghypnosis to manage headaches.

Although our results showed that the efficacy of hypnosis might besuperior to CBT, the treatments do not have to be mutually exclusive.Kirsch, Montgomery, and Sapirstein (1995) reported that using hypnosisas an adjunct to CBT enhances treatment outcomes for clinical prob-lems, such as chronic pain. Castel, Cascon, Padrol, Sala, and Rull (2012)also conducted an RCT targeting fibromyalgia and reported that CBTwith hypnosis was more effective than just CBT alone. Thus, we suggestthat the combination of hypnosis and CBT could lead to more effectiveclinical treatment.

We need to consider the application of hypnosis within group set-tings. Eleven of our studies used hypnosis in individual settings, whileonly Castel et al. (2009) adopted a group setting. Vinogradov andYalom (1989) described the efficient use of resources and the cost

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


effectiveness of group psychotherapy. Hypnosis within group settingshas been used to treat several diseases, such as allergies (Madrid,Rostel, Pennington, & Murphy, 1995), breast cancer (Butler et al., 2009;Spiegel & Bloom, 1983), depression (Butler et al., 2008), fibromyalgia(Castel et al., 2009, 2012), and posttraumatic stress disorder (Lesmana,Suryani, Jensen, & Tiliopoulos, 2009). These studies also reported theefficacy of hypnotic interventions. Although it is difficult to establish thebest-suited therapy for one patient or client, group interventions mightprovide an effective format to deliver hypnotherapy for chronic pain.

LimitationsFirst, results of our meta-analysis revealed large heterogeneities

between studies. Large heterogeneity distorts the reliability of ourresults. Efficacy of hypnosis might not have been clearly determinedbased on the current results. Second, it is possible that the current studydid not provide a pure comparison of hypnosis with other psychologi-cal interventions by including two studies (Castel et al., 2009; Zitmanet al., 1992). Therefore it is possible that we are overestimating theefficacy of hypnosis in our study.

Conclusions

The objective of this study was to demonstrate the empiricalefficacy of hypnosis. We observed that hypnosis was moderately effec-tive for managing chronic pain compared to a standard care duringa postintervention phase. Hypnosis was also effective when com-pared with other psychological interventions in a nonheadache group.However, the methodological quality of individual studies was ratherlow. Large heterogeneities were also observed. Future studies shouldimprove upon methodological quality and heterogeneity to betterdetermine the efficacy of hypnosis for chronic pain management.

References

∗References marked with an asterisk indicate studies included in the meta-analysis.∗Abrahamsen, R., Baad-Hansen, L., & Svensson, P. (2008). Hypnosis in the management

of persistent idiopathic orofacial pain: Clinical and psychosocial findings. Pain, 136,44–52.

∗Abrahamsen, R., Zachariae, R., & Svensson, P. (2009). Effect of hypnosis on oral functionand psychological factors in temporomandibular disorders patients. Journal of OralRehabilitation, 36, 556–570.

Begg, C., Cho, M., Eastwood, S., Horton, R., Moher, D., Olkin, I., . . . Stroup, D. F. (1996).Improving the quality of reporting of randomized controlled trials: The CONSORTstatement. Journal of the American Medical Association, 276, 637–639.

Boersma, K., Linton, S., Overmeer, T., Jansson, M., Vlaeyen, J., & de Jong, J. (2004).Lowering fear-avoidance and enhancing function through exposure in vivo: A mul-tiple baseline study across six patients with back pain. Pain, 108, 8–16.

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


Borkovec, T. D., & Nau, S. D. (1972). Credibility of analogue therapy rationales. Journal ofBehavior and Experimental Psychiatry, 3, 257–260.

Butler, L. D., Koopman, C., Neri, E., Giese-Davis, J., Palesh, O., Thorne-Yocam, K. A., . . .

Spiegel, D. (2009). Effects of supportive-expressive group therapy on pain in womenwith metastatic breast cancer. Health Psychology, 28, 579–587.

Butler, L. D., Waelde, L. C., Hastings, T. A., Chen, X., Symons, B., Marshall, J., . . . Spiegel,D. (2008). Meditation with yoga, group therapy with hypnosis, and psychoeducationfor long-term depressed mood: A randomized pilot trial. Journal of Clinical Psychology,64, 806–820.

Castel, A., Cascon, R., Padrol, A., Sala, J., & Rull, M. (2012). Multicomponent cognitive-behavioral group therapy with hypnosis for the treatment of fibromyalgia: Long-termoutcome. Journal of Pain, 13, 255–265.

∗Castel, A., Salvat, M., Sala, J., & Rull, M. (2009). Cognitive-behavioural group treatmentwith hypnosis: A randomized pilot trial in fibromyalgia. Contemporary Hypnosis, 26,48–59.

Cook, A. J., Brawer, P. A., & Vowles, K. E. (2006). The fear-avoidance model of chronicpain: Validation and age analysis using structural equation modeling. Pain, 121,195–206.

Derogatis, L. R., Lipman, R. S., & Covi, L. (1973). SCL-90: An outpatient psychiatric ratingscale—preliminary report. Psychopharmacology Bulletin, 9, 13–28.

Eccleston, C., Palermo, T. M., Williams, A. C., Lewandowski, A., & Morley, S. (2009).Psychological therapies for the management of chronic and recurrent pain in childrenand adolescents. Cochrane Database of Systematic Reviews, 2, 1–51.

Eccleston, C., Williams, A. C., & Morley, S. (2009). Psychological therapies for the man-agement of chronic pain (excluding headache) in adults. Cochrane Database of SystematicReviews, 2, 1–101.

Faymonville, M. E., Laureys, S., Degueldre, C., Fiore, G. D., Luxen, A., Franck, G.,. . . Maquet, P. (2000). Neural mechanisms of antinociceptive effects of hypnosis.Anesthesiology, 92, 1257–1267.

Fordyce, W. E., Fowler, R. S., & DeLateur, B. (1968). An application of behavior mod-ification technique to a problem of chronic pain. Behavior Research and Therapy, 6,105–107.

Fordyce, W. E., Fowler, R. S., Lehmann, J. F., DeLateur, B. J., Sand, P. L., & Trieschmann,R. B. (1973). Operant conditioning in the treatment of chronic pain. Archives of PhysicalMedicine and Rehabilitation, 54, 399–408.

∗Gay, M. C., Philippot, P., & Luminet, O. (2002). Differential effectiveness of psychologicalinterventions for reducing osteoarthritis pain: A comparison of Erikson [sic] hypnosisand Jacobson relaxation. European Journal of Pain, 6, 1–16.

Hannen, H. C., Hoenderdos, H. T., van Romunde, L. K., Hop, W. C., Mallee, C., Terwiel,J. P., & Hekster, G. B. (1991). Controlled trial of hypnotherapy in the treatment ofrefractory fibromyalgia. Journal of Rheumatology, 18, 72–75.

Hedges, L. V., & Olkin, I. (1984). Nonparametric estimators of effect size in meta-analysis.Psychological Bulletin, 96, 573–580.

Henschke, N., Ostelo, R. W. J. G., van Tulder, M. W., Vlaeyen, J. W. S., Morley, S.,Assendelft, W. J. J., & Main, C. J. (2011). Behavioural treatment for chronic low-backpain. Cochrane Database of Systematic Reviews, 2, 1–117.

Higgins, J. P. T., & Green, S. (Eds.). (2011). Cochrane handbook for systematic reviews of inter-ventions version 5.1.0 (updated March 2011). Retrieved from http://www.cochrane-handbook.org./http://www.cochrane-handbook.org

Hofbauer, R. K., Rainville, P., Duncan, G. H., & Bushnell, C. (2001). Cortical representationof the sensory dimension of pain. Journal of Neurophysiology, 86, 402–411.

Holroyd, J. (1996). Hypnosis treatment of clinical pain: Understanding why hypnosis isuseful. International Journal of Clinical and Experimental Hypnosis, 44, 33–51.

Jensen, M. P. (2010). A neuropsychological model of pain: Research and clinical implica-tions. Journal of Pain, 11, 2–12.

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


∗Jensen, M. P., Barber, J., Romano, J. M., Hanley, M. A., Raichle, K. A., Molton, I.R.,. . . Patterson, D. R. (2009). Effects of self-hypnosis training and EMG biofeedbackrelaxation training on chronic pain in persons with spinal-cord injury. InternationalJournal of Clinical and Experimental Hypnosis, 57, 239–268.

∗Jensen, M. P., Barber, J. Romano, J. M., Molton, I. R., Raichle, K. A., Osborne, T. L.,. . .

Patterson, D. R. (2009). A comparison of self-hypnosis versus progressive musclerelaxation in patients with multiple sclerosis and chronic pain. International Journalof Clinical and Experimental Hypnosis, 57, 198–221.

Jensen, M. P., Ehde, D. M., Gertz, K. J., Stoelb, B. L., Dillworth, T. M., Hirsh, A. T., . . . Kraft,G. H. (2011). Effects of self-hypnosis training and cognitive restructuring on daily painintensity and catastrophizing in individuals with multiple sclerosis and chronic pain.International Journal of Clinical and Experimental Hypnosis, 59, 45–63.

Jensen, M. P., & Patterson, D. R. (2006). Hypnotic treatment of chronic pain. Journal ofBehavioral Medicine, 29, 95–124.

Jensen, M. P., Turner, J. A., & Romano, J. M. (2001). Changes in beliefs, catastrophizing,and coping are associated with improvement in multidisciplinary pain treatment.Journal of Consulting and Clinical Psychology, 69, 655–662.

∗Jones, H., Cooper, P., Miller, V., Brooks, N., & Whorwell, P. J. (2006). Treatment of non-cardiac chest pain: A controlled trial of hypnotherapy. Gut, 55, 1403–1408.

Kanji, N., White, A. R., & Ernst, E. (2006). Autogenic training for tension type headaches:A systematic review of controlled trials. Complementary Therapies in Medicine, 14,144–150.

Kirsch, I., Montgomery, G., & Sapirstein, G. (1995). Hypnosis as an adjunct tocognitive-behavioral psychotherapy: A meta-analysis. Journal of Consulting and ClinicalPsychology, 63, 214–220.

Kisley, S. R., Campbell, L. A., Skerritt, P., & Yelland, M. J. (2010). Psychological interven-tions for symptomatic management of non-specific chest pain in patients with normalcoronary anatomy. Cochrane Database of Systematic Reviews, 1, 1–47.

Lesmana, C. B. J., Suryani, L. K., Jensen, G. D., & Tiliopoulos, N. (2009). A spiritual-hypnosis assisted treatment of children with PTSD after the 2002 Bali terrorist attack.American Journal of Clinical Hypnosis, 52, 23–34.

Lindstrom, I., Ohlund, C., Eek, C., Wallin, L., Peterson, L. E., Fordyce, W. E., &Nachemson, A. L. (1992). The effect of graded activity on patients with subacutelow back pain: A randomized prospective clinical study with an operant-conditioningbehavioral approach. Physical Therapy, 72, 279–290.

Lippe, P. M. (2000). The decade of pain control and research. Pain Medicine, 1, 286.Lunde, L., Nordhus, I. H., & Pallesen, S. (2009). The effectiveness of cognitive and

behavioural treatment of chronic pain in the elderly: A quantitative review. Journalof Clinical Psychology in Medical Settings, 16, 254–262.

Madrid, A., Rostel, G., Pennington, D., & Murphy, D. (1995). Subjective assessmentof allergy relief following group hypnosis and self-hypnosis: A preliminary study.American Journal of Clinical Hypnosis, 38, 80–86.

Melzack, R., & Wall, P. D. (1982). The challenge of pain. London, UK: Penguin.Moher, D., Hopewell, S., Schulz, K., Montori, V., Gøtzsche, P., Devereaux, P., . . . Altman,

D. (2010). Consort 2010 explanation and elaboration: Updated guidelines for reportingparallel group randomised trials. Journal of Clinical Epidemiology, 63, 1–37.

Moher, D., Schulz, K. F., & Altman, D. (2001). The consort statement: Revised recom-mendations for improving the quality of reports of parallel-group randomized trials.Journal of the American Medical Association, 285, 1987–1991.

Montgomery, G. H., Duhamel, K. N., & Redd, W. H. (2000). A meta-analysis of hypnot-ically induced analgesia. International Journal of Clinical and Experimental Hypnosis, 48,138–153.

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


Morley, S., Eccleston, C., & Williams, A. (1999). Systematic review and meta-analysis ofrandomized controlled trials of cognitive behaviour therapy and behaviour therapyfor chronic pain in adults, excluding headache. Pain, 80, 1–13.

Muraoka, M., Komiyama, H., Hosoi, M., Mine, K., & Kubo, C. (1996). Psychosomatic treat-ment of phantom limb pain with posttraumatic stress disorder: A case report. Pain, 66,385–388.

∗Palsson, O. S., Turner, M. J., Johnson, D. A., Burnett, C. K., & Whitehead, W. E. (2002).Hypnosis treatment for severe irritable bowel syndrome: Investigation of mechanismand effects on symptoms. Digestive Diseases and Sciences, 47, 2605–2614.

Patterson, D. R., & Jensen, M. P. (2003). Hypnosis and clinical pain. Psychological Bulletin,129, 495–521.

Posadzki, P., Lewandowski, W., Terry, R., Ernst, E., & Stearns, A. (2012). Guided imageryfor non-musculoskeletal pain: A systematic review of randomized clinical trials.Journal of Pain and Symptom Management, 44, 95–104.

Rainville, P., Bao, Q. V. H., & Chrétien, P. (2005). Pain-related emotions modulateexperimental pain perception and autonomic responses. Pain, 118, 306–318.

Rainville, P., Duncan, G. H., Price, D. D., Carrier, B., & Bushnell, M. C. (1997). Painaffect encoded in human anterior cingulate but not somatosensory cortex. Science, 277,968–971.

Schultz, J. H. (1987). Das autogene training [Autogenic training]. Stuttgart, Germany:Thieme.

Spiegel, D., & Bloom, J. R. (1983). Group therapy and hypnosis reduce metastatic breastcarcinoma pain. Psychosomatic Medicine, 45, 333–339.

Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the State-Trait AnxietyInventory (self-evaluation questionnaire). Palo Alto, CA: Consulting Psychologists.

∗Spinhoven, P., Linssen, A. C., van Dyck, R., & Zitman, F. G. (1992). Autogenic trainingand self-hypnosis in the control of tension headache. General Hospital Psychiatry, 14,408–415.

Spinhoven, P., ter Kuile, M., Kole-Snijders, A. M. J., Mansfeld, M. H., Ouden, D., &Vlaeyen, J. W. S. (2004). Catastrophizing and internal pain control as mediators of out-come in the multidisciplinary treatment of chronic low back pain. European Journal ofPain, 8, 211–219.

Stetter, F., & Kupper, S. (2002). Autogenic training: A meta-analysis of clinical outcomestudies. Applied Psychophysiology and Biofeedback, 27, 45–98.

Sullivan, M. J. L., Thorn, B., Haythornthwaite, J. A., Keefe, F., Martin, M., Bradley, L. A., &Lefebvre, J. C. (2001). Theoretical perspectives on the relation between catastrophizingand pain. Clinical Journal of Pain, 17, 52–64.

∗Ter Kuile, M. M., Spinhoven, P., Linssen, A. C., Zitman, F. G., van Dyck, R., & Rooijmans,H. G. (1994). Autogenic training and cognitive self-hypnosis for the treatment ofrecurrent headaches in three different subject groups. Pain, 58, 331–340.

Turk, D. C. (2002). Clinical effectiveness and cost-effectiveness of treatments for patientswith chronic pain. Clinical Journal of Pain, 18, 355–365.

∗Van Dyck, R., Zitman, F. G., Linssen, A. C., & Spinhoven, P. (1991). Autogenic trainingand future oriented hypnotic imagery in the treatment of tension headache: Outcomeand process. International Journal of Clinical and Experimental Hypnosis, 39, 6–23.

Verhaak, P. F. M., Kerssens, J. J., Dekker, J., Sorbi, M. J., & Bensing, J. M. (1998). Prevalenceof chronic benign pain disorder among adults: A review of the literature. Pain, 77,231–239.

Vinogradov, S., & Yalom, I. (1989). A concise guide to group psychotherapy. Washington, DC:American Psychiatric.

Vlaeyen, J. W. S., de Jong, J., Geilen, M., Heuts, P. H. T. G., & van Breukelen, G. (2002). Thetreatment of fear of movement / (re) injury in chronic low back pain: Further evidenceon the effectiveness of exposure in vivo. Clinical Journal of Pain, 18, 251–261.

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


Vlaeyen, J. W. S., Kole-Snijders, A. M. J., Boeren, R. G. B., & van Eek, H. (1995). Fearof movement/(re) injury in chronic low back pain and its relation to behavioralperformance. Pain, 62, 363–372.

Yates, S. L., Morley, S., Eccleston, C., & Williams, A. C. (2005). A scale for rating the qualityof psychological trials for pain. Pain, 117, 314–325.

∗Zitman, F. G., van Dyck, R., Spinhoven, P., & Linssen, A. C. (1992). Hypnosis andautogenic training in the treatment of tension headaches: A two-phase constructivedesign study with follow-up. Journal of Psychosomatic Research, 36, 219–228.

Zung, W. W. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63–70.

AppendixSearch strategy

1. Medline

1. randomized controlled trial.pt.2. randomized Controlled Trial/3. Randomized Controlled Trial as Topic/4. Controlled clinical trial.pt.5. Clinical trial.pt.6. Clinical Trial/7. Clinical Trial as Topic/8. Random Allocation9. Double-Blind Method/

10. Single-Blind Method/

11. Comparative Study/

12. Empirical Research/

13. Treatment Outcome/14. Comparative Effectiveness Research/

15. or/1-1416. exp Hypnosis/17. exp ∗Pain/

18. 15 and 16 and 1719. Remove duplicates from 1820. Limit 19 to English language21. Limit 20 to “review articles”22. 20 not 21

2. CENTRAL

1. Medical Subject Heading descriptor Pain explode all trees2. (pain∗):ti,ab,kw3. MeSH descriptor Hypnosis explode all trees4. (hypnos∗):ti,ab,kw5. (#1 OR #2)6. (#3 OR #4)7. (#5 OR #6)

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


3. PsycINFO

1. exp ∗Pain/

2. exp Hypnosis/3. 1 and 24. limit 3 to English language5. limit 4 to all journals

Eine Meta-Analyse zu Hypnose bei Chronischen Schmerzsyndromen –EinVergleich zwischen Hypnose, Standardbehandlung und anderen

psychologischen Interventionen

Tomonori Adachi, Haruo Fujino, Aya Nakae, Takashi Mashimo undJun Sasaki

Abstrakt: Hypnose ist als effektive Behandlung psychologischer und phys-iologischer Beschwerden angesehen. Dennoch wurde die Effizienz alsStrategie zur Behandlung chronischen Schmerzes bisher noch nicht durchmeta-analytische Methoden bewertet. Das Ziel der aktuellen Studie wardie Durchführung einer Metaanalyse, um die Effizienz der Hypnoseim Management chronischen Schmerzes zu erheben. Verglichen mit derStandardbehandlung zeigte die Hypnose einen leichten Behandlungsvorteil.Außerdem zeigte die Hypnose einen leichten Vorteil im Vergleich mitanderen psychologischen Interventionen bei einer Nicht-Kopfschmerz-Gruppe. Die Ergebnisse lassen daarfu schließen, daß Hypnose bei derBehandlung chronischen Schmerzes effizient ist. Es wurde eine großeHeterogenität der eingeschlossenen Studien festgestellt. Die Natur derHypnosebehandlung wird weiterhin diskutiert.

Stephanie Reigel, MD

Une méta-analyse de l’hypnose dans les cas de douleurs chroniques - unecomparaison entre l’hypnose, les soins habituels et d’autre

interventions psychologiques

Tomonori Adachi, Haruo Fujino, Aya Nakae, Takashi Mashimo etJun Sasaki

Résumé: L’hypnose est considérée comme un traitement efficace des affec-tions psychologiques et physiques. Pourtant, son efficacité comme stratégiede gestion de la douleur chronique n’a pas été évaluée par des méthodesméta-analytiques. L’objectif de cette étude consistait à mener une méta-analyse afin d’évaluer l’efficacité de l’hypnose dans la gestion de la douleurchronique. Comparativement aux soins standards, l’hypnose en tant quetraitement était modérément bénéfique. De plus, le traitement par l’hypnoses’est révélé supérieur à la moyenne, comparativement à d’autres interven-tions psychologiques auprès d’un groupe ne souffrant pas de maux de tête.Ces résultats semblent indiquer que l’hypnose est efficace dans la gestionde la douleur chronique. Étant donné la grande hétérogénéité des études

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5


comprises dans cette recherche, la nature du traitement par l’hypnose y estlonguement examinée.

Johanne ReynaultC. Tr. (STIBC)

Un meta-análisis de la hipnosis para problemas de dolor crónico: Unacomparación entre hipnosis, tratamiento estándar, y otras

intervenciones psicológicas

Tomonori Adachi, Haruo Fujino, Aya Nakae, Takashi Mashimo, yJun Sasaki

Resumen: La hipnosis es considerada como un tratamiento efectivo paraenfermedades físicas y psicológicas. Sin embargo, su eficacia como estrate-gia para el manejo de dolor crónico no ha sido evaluada a través de métodosmeta-analíticos. El objetivo del presente estudio fue la realización de unmeta-análisis para evaluar la eficacia de la hipnosis para el manejo de dolorcrónico. Al compararla con el tratamiento estándar, la hipnosis brindó un ben-eficio moderado. La hipnosis también mostró un efecto superior moderadoal compararla con otras intervenciones psicológicas para un grupo que nopadecía dolores de cabeza. Los resultados sugieren que la hipnosis es eficazpara el manejo de dolor crónico. Considerando la gran heterogeneidad de losestudios incluidos, se profundiza en la discusión sobre la naturaleza de lahipnosis como tratamiento.

Omar Sánchez-Armáss Cappello, PhDAutonomous University of San Luis Potosi,Mexico

Dow

nloa

ded

by [

Joan

nes

Mer

tens

] at

03:

33 0

7 A

ugus

t 201

5

C:iToolsWMSTandF-Journals4521288WorkingFolderNHYP A … · chest pain, orofacial pain, osteoarthritis pain, spinal cord injury, temporomandibular disorders, and other forms of chronic

Documents