THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 11, Number 4, 2005, pp. 663–671
© Mary Ann Liebert, Inc.
Treatment of Fibromyalgia with Formula Acupuncture:
Investigation of Needle Placement, Needle Stimulation,
and Treatment Frequency
RICHARD E. HARRIS, Ph.D.,1XIAOMING TIAN, M.D., L.Ac.,2DAVID A. WILLIAMS, Ph.D.,1
THOMAS X. TIAN, C.M.D., L.Ac.,2THOMAS R. CUPPS, M.D.,3FRANK PETZKE, M.D.,4
KIMBERLY H. GRONER, M.S.N.,1PINAKI BISWAS, M.Stat.5RICHARD H. GRACELY, Ph.D.,1
and DANIEL J. CLAUW, M.D.1
Objectives: The objective of this study was to investigate whether typical acupuncture methods such as nee-
dle placement, needle stimulation, and treatment frequency were important factors in fibromyalgia symptom
Design/settings/subjects: A single-site, single-blind, randomized trial of 114 participants diagnosed with fi-
bromyalgia for at least 1 year was performed.
Intervention: Participants were randomized to one of four treatment groups: (1) T/S needles placed in tra-
ditional sites with manual needle stimulation (n ? 29): (2) T/0 traditional needle location without stimulation
(n ? 30); (3) N/S needles inserted in nontraditional locations that were not thought to be acupuncture sites,
with stimulation (n ? 28); and (4) N/0 nontraditional needle location without stimulation (n ? 2 7). All groups
received treatment once weekly, followed by twice weekly, and finally three times weekly, for a total of 18
treatments. Each increase in frequency was separated by a 2-week washout period.
Outcome measures: Pain was assessed by a numerical rating scale, fatigue by the Multi-dimensional Fa-
tigue Inventory, and physical function by the Short Form–36.
Results: Overall pain improvement was noted with 25%–35% of subjects having a clinically significant de-
crease in pain; however this was not dependent upon “correct” needle stimulation (t ? 1.03; p ? 0.307) or lo-
cation (t ? 0.76; p ? 0.450). An overall dose effect of treatment was observed, with three sessions weekly pro-
viding more analgesia than sessions once weekly (t ? 2.10; p ? 0.039). Among treatment responders,
improvements in pain, fatigue, and physical function were highly codependent (all p ? 0.002).
Conclusions: Although needle insertion led to analgesia and improvement in other somatic symptoms, cor-
rect needle location and stimulation were not crucial.
cupuncture has been used as a therapeutic intervention
for more than 2500 years in China, and remains an im-
portant facet of many modern Asian medical systems.1The
use of acupuncture as a complementary and alternative form
of medicine in Western countries has increased rapidly over
the last three decades.2,3The U.S. Food and Drug Admin-
1Department of Internal Medicine, Division of Rheumatology, University of Michigan, Ann Arbor, MI.
2Academy of Acupuncture and Chinese Medicine, Bethesda, MD.
3Department of Rheumatology, Georgetown University, Washington, DC.
4Department of Anesthesiology, University of Cologne, Cologne, Germany.
5Department of Biostatistics, University of Michigan, Ann Arbor, MI.
istration (FDA) estimates that Americans spend half a bil-
lion dollars per year on acupuncture treatments, most com-
monly for conditions involving pain,4despite the lack of
agreement within the scientific community on its efficacy
and mechanism of action. Of note, the most appropriate con-
trol intervention to use is undecided5and it is not clear
whether acupuncture efficacy is synergistically dependent
on where the needles are placed and whether they are stim-
ulated.6As a control some investigators use nontraditional
sites,7–9whereas others insert needles in traditional acupunc-
ture points but refrain from stimulating them.10,11
Fibromyalgia (FM), a condition characterized by chronic
diffuse pain, is the second most common rheumatologic dis-
order, affecting 2%–4% of the populations of industrialized
countries.12,13To meet American College of Rheumatology
criteria for this diagnosis, individuals must have a history
of chronic pain in all four quadrants of the body plus the
axial skeleton, and have 11 or more (of a possible 18) ten-
der points.14Although FM is partly defined by tenderness
at these specific anatomic sites, in fact this condition is char-
acterized by generalized tenderness, or hyperalgesia.15,16As
no universally effective, well-tolerated treatment exists, pa-
tients with FM often seek acupuncture.17Yet, most studies
on the efficacy of acupuncture in FM have been uncontrolled
case series,18,19and only a single high-quality controlled
trial of 70 patients randomized to electro-acupuncture or
sham electrical stimulation has provided evidence support-
ing acupuncture for this condition.11,20Although seven of
eight outcomes improved in the active treatment group com-
pared to none in the sham group, the extent to which the
sham group was blinded was not assessed.
To explore methodological acupuncture issues, a 2 ? 2
factorial was used design to address the following questions:
(1) Is acupuncture beneficial for FM? (2) Is the placement
of needles in traditional locations, and stimulation of the
needles, required for symptom improvement? and (3) Is
there a dependence on treatment frequency? It was hypoth-
esized that if needle location and/or stimulation were es-
sential, then an additive or synergistic effect should be de-
tected with traditional or correct needle placement and
stimulation. In addition, if acupuncture produces analgesia
via a physiologic mechanism, a frequency-dependent im-
provement in symptoms might be observed.
Participants were recruited from the Washington, DC,
metropolitan area from August 2000 through January 2002
via announcements in local newspapers and periodicals and
were screened on the telephone for study eligibility. Inclu-
sion criteria were: (1) having met ACR criteria for the di-
agnosis of FM14for at least 1 year; (2) reported widespread
pain on more than 50% of days; (3) willingness to limit the
introduction of new medications or treatment modalities for
FM symptoms. Exclusion criteria were: (1) sufficient knowl-
edge of acupuncture techniques to prevent blinding (in-
cluding previous acupuncture); (2) known bleeding diathe-
sis; (3) autoimmune or inflammatory disease; (4) regular use
of narcotic analgesics daily or a history of substance abuse;
(5) contraindication to the use of acetaminophen or
ibuprophen (rescue analgesics) ; (6) participation in other
therapeutic trials; (7) pregnancy or lactation; or (8) receiv-
ing disability payments or were involved in litigation related
to FM. Subjects were allowed to continue their normal treat-
ment regime(s), including use of antidepressants); however
they were asked not to make any additional changes and not
to seek acupuncture outside of the trial.
All participants were randomized and allocated to one of
four treatment arms using computer-generated random num-
bers in a four-block design. Treatment allocation was con-
cealed in an opaque envelope and then presented to the
acupuncturist 1 day before treatment. All procedures were
approved by the local institutional review board; participants
received full descriptions of the protocol and gave written
This study used a 2 ? 2 factorial design to examine the
individual and synergistic effects of both needle location and
needle stimulation on the efficacy of acupuncture analgesia.
The four intervention arms were: (1) T/S (traditional site
with stimulation): subjects received acupuncture, at points
lying on Traditional Chinese Medicine acupuncture merid-
ians, combined with manual stimulation of the needles; (2)
T/0 (traditional site without stimulation): participants re-
ceived the needles placed at the traditional locations but
without any needle manipulation, (3) N/S (nontraditional site
with stimulation): participants received the same number of
needles placed at the same depths and with the same degree
of manual stimulation as the T/S group, but these needles
were placed in sites that were not believed to be effective
in Traditional Chinese Medicine–based acupuncture; and (4)
N/0 (nontraditional site with no stimulation): subjects re-
ceived needles in nontraditional sites without stimulation.
Subjects were told that they would be randomly assigned to
one of these treatment groups and would receive either tra-
ditional acupuncture therapy or control acupuncture tech-
niques not known to benefit FM. The T/S arm represented
traditional Chinese acupuncture whereas the other three
arms could be considered as different control groups.
The active point formula (Du 20, LI 11, LI 4, GB 34, bi-
lateral St 36, Sp 6, Liv 3, and Ear-Shenmen)1was chosen
based on the points’ ability to relieve FM symptoms in Tra-
ditional Chinese Medicine. Needles placed in the nontradi-
tional groups were not on acupuncture meridians or points
(Fig. 1A). The choice of point locations was determined by
HARRIS ET AL.
licensed acupuncturists (X.T. and T.X.T., who each have 12
years of experience in treating FM and 17 years of experi-
ence with this form of acupuncture).
Treatments followed a forced-titration paradigm: once
weekly for 3 weeks, then twice weekly for 3 weeks, and fi-
nally 3 times weekly for 3 weeks for a total of 18 sessions.
A 2-week wash-out period was inserted between each treat-
ment dose (Fig. 1B). Sterile, 2.54-cm, single-use, dispos-
able, 38-gauge stainless steel needles were used (HBW Sup-
ply Inc., Hemet, CA). Stimulation was performed manually
by lifting and thrusting while evenly rotating the needle
(?12 times at 180 degrees clockwise and counterclockwise).
Characteristic de qi needling sensations1resulting from nee-
dle manipulation were elicited in the T/S group but not in
the N/S group. There was no intention from the acupunc-
turist to elicit needle sensations in the N/S group. De qi was
assessed by the acupuncturist’s sensation of needle grasp-
ing and by the subject’s report to a research assistant. Nee-
dles were retained for 20 minutes. More than 95% of the
acupuncture sessions were performed by X.T., and the re-
maining sessions were performed by licensed acupunctur-
ists trained for this study.
To ensure and to verify adequate blinding throughout the
study multiple measures were taken. Participants from dif-
ferent treatment arms were never present in the acupuncture
clinic at the same time. Subjects were asked to refrain from
seeking additional information on acupuncture and from dis-
cussing acupuncture with other participants, during the time
in the study. All subjects were blindfolded during treatment
and no verbal communication was allowed between the
acupuncturist and the subjects. A nonblinded research as-
sistant was present during all sessions to monitor and en-
sure treatment integrity. This study was single-blinded. The
acupuncturist knew the treatment groups and study hypoth-
esis; however all evaluators of study outcomes were blinded
to treatment allocation. Blinding was assessed by question-
naire following the third acupuncture session. This ques-
tionnaire read: “What type of procedure do you think you
are receiving, acupuncture, placebo (sham), or can’t tell.”
A priori outcomes
Pain (primary outcome). Subjects completed a 101-point
numeric rating scale (NRS) which ranged from 0 to 100
points in 5-point increments. The scale was anchored at 0
by “no pain” and 100 by “worst pain imaginable.” Clini-
cally meaningful treatment responders were determined us-
ing the criteria suggested by Farrar et al.21: either a 20-point
reduction or a 30% improvement in pain from baseline. On
average 85.5% of responders met both criteria, while 9.8%
met the 30% improvement criterion only and 4.6% met the
20-point reduction criterion only.
Fatigue. Participants completed the Multi-Dimensional
Fatigue Inventory questionnaire22and clinically meaningful
changes in fatigue were calculated using the Reliability of
Change Index (RC).23A change of 3.67 in the General Fa-
tigue score from baseline was used to classify fatigue re-
sponders. The General Fatigue scores range from 4 to 20
with larger scores indicating more fatigue.
Function. Physical function was assessed with the phys-
ical components score calculated from the Short Form-36
questionnaire.24This score ranges from 0 to 100 with higher
scores indicating better function. Previous work has sug-
gested that a change of seven points in the physical com-
ponents score is clinically meaningful and therefore we used
this value as a cutoff for function responders.
A priori it was estimated that a sample size of 30 patients
per group was needed to detect 30% differences between
ACUPUNCTURE FOR FIBROMYALGIA
comes. A. Body maps of approximate needle locations. #, non-
traditional points; *, inserted bilaterally. B. Representation of
study outcomes and acupuncture sessions by week. NRS, nu-
meric rating scale; MFI, Multi-dimensional Fatigue Inventory;
SF-36, Short Form–36.
Needle locations and temporal outline of study out-
groups with a power of 0.82 and a significance level of 0.05.
Longitudinal data were analyzed according to intention to
treat. Incomplete data were handled via a last observation
carried forward in time. Data were either hand entered or
scanned via Teleform (Cardiff Software Inc., Bozeman, MT)
and double-checked. Analysis was carried out using SAS
(Cary, NC) and SPSS (Chicago, IL) software.
Analyses of variance were performed to determine dif-
ferences in age, years of disease diagnosis, number of treat-
ments attended, and baseline pain; ?2tables were con-
structed for responder status, male/female ratio, and
dropouts. Paired t-tests were performed to determine sig-
nificant changes in NRS pain scores between weeks 3 and
4, 8 and 9, and 13 and 15.
Primary analyses contained two strategies: (1) assessment
of changes in mean scores and (2) assessment of changes in
binary responder classification. These two approaches were
taken to examine the robustness of findings. To examine the
effects of location, stimulation, and treatment frequency (dose)
on pain, a repeated-measures model was fit to the data by cal-
culating the change in response from baseline to weeks 3, 8,
and 13, including fixed effects for a combination of dose, lo-
cation, and stimulus. Model 1 appeared as follows:
yijkl? xl? ? ? ?ijk? ?ijkl
where ?ijk, i ? 1, 2, 3; j ? 1, 2 and k ? 1, 2 represents the
effect for the ithdose level, jthstimulus, and kthlocation; xl
is the baseline response for the lthpatient. A general corre-
lation structure was assumed across the dose levels to ac-
count for within-patient variability across time, as time and
dose were confounded in the experiment.
To analyze data from the binary pain outcome (respon-
ders versus nonresponders), a similar repeated-measures
model was used (model 2). This model used a binomial as-
sumption for the distribution and a logit link leading to a
generalized linear model (GLM), as follows:
logit pijk? ? ? ?ijk
where pijkrepresents the probability of responding. The other
terms have the same interpretation as model 1. Here, a re-
sponse value of 1 indicates a 20-point or 30% change on the
NRS pain scale of the current pain rating compared to base-
Baseline demographics and dropouts
Subject demographics are displayed in Table 1. There
were no significant group differences in duration of fi-
bromyalgia (F ? 0.070; p ? 0.976) or age (F ? 2.186; p ?
0.094). Not unexpectedly, the majority of subjects were fe-
male (106/114); however the ratio of female to male sub-
jects was not significantly different across groups (?2?
5.019, p ? 0.136). No differences were detected in pain
scores between groups (F ? 0.215; p ? 0.886).
Figure 2 depicts the flow of participants through the
study. Thirty-three percent (33%; 38/114) of subjects
dropped out of the study before completion. The majority
of dropouts occurred in the first 4 weeks (21/38; 55%); the
primary reason for dropping out was the time commitment
(11/38). There were no significant group differences in
dropout rates (?2? 3.039, p ? 0.386), and the total number
of treatment sessions attended did not vary by group (F ?
0.122; p ? 0.947; Table 1).
To assess whether subjects could determine to which
treatment arm they had been randomized, they were asked
in week 4, if they thought they were receiving either (a)
acupuncture or (b) placebo or (c) could not tell. A ?2analy-
sis revealed no significant differences (Fisher’s exact ?2?
7.531, p ? 0.259), indicating that participants remained
blinded to treatment assignment at week 4.
HARRIS ET AL.
TABLE 1. SOCIODEMOGRAPHIC CHARACTERISTICS OF FIBROMYALGIA (FM)
STUDY PARTICIPANTS AND AVERAGE NUMBER OF TREATMENTS
ParameterT/ST/0 N/S N/0
Years of FM diagnosis*
T/S, traditional needle locations with manual stimulation; T/0, traditional needle location without manual stimulation; N/S, nontra-
ditional needle location with manual stimulation; N/0, nontraditional needle location without manual stimulation.
*Values are mean (standard deviation) by treatment group.
Needle location and stimulation
Mean pain, fatigue, and function scores are shown in
Table 2 and responders are displayed in Table 3. Based on
analysis using model 1, no significant effects were obtained
for either needle stimulation (weeks 3, 8, 13: t ? 1.03; p ?
0.307) or location (weeks 3, 8, 13: t ? 0.76; p ? 0.450).
Similarly no significant effects were obtained from model
2 for the binary response pain variable for either needle stim-
ulation (weeks 3, 8, 13: ?2? 3.60; p ? 0.058) or location
(weeks 3, 8, 13: ?2? 0.20; p ? 0.657).
Figure 3A shows a graphic representation of the per-
centage of pain responders by time. No group consistently
had more improvement than another. Taken together these
results suggested that there was no additional benefit de-
rived from placing needles in the correct location, or stim-
ulating the needles, at these time points.
Treatment frequency (cumulative dose)
As exploratory analyses we tested for effects of treat-
ment frequency in data from all four groups combined. A
fit of model 1 to the continuous NRS pain scores yielded
a significant overall effect for treatment frequency between
weeks 4 and 15 (t ? 4.81; p ? 0.045) and between weeks
3 and 13 (t ? 4.92; p ? 0.039). No significant effects were
ACUPUNCTURE FOR FIBROMYALGIA
needle location; T/S, traditional needle location; T/0, traditional
needle location without stimulation; N/0, nontraditional needle
without manual stimulation.
Flow of participants through study. N/S, nontraditional
TABLE 2. SCORES FOR PAIN, FATIGUE, AND FUNCTION BY ACUPUNCTURE GROUP AND WEEK
T/ST/0 N/S N/0Total
Mean (standard deviation) scores for pain, fatigue and function outcomes by treatment group and week are shown. No treatment
group showed significant improvement over the others.
T/S, traditional needle locations with manual stimulation; T/0, traditional needle location without manual stimulation; N/S, non-
traditional needle location with manual stimulation; N/0, nontraditional needle location without manual stimulation.
aMean (SD) in change score (pain: week 0–13; fatigue and function: week 0–15).
bModel 1, p value for comparison of groups: T/S ? T/0 vs. N/S ? N/0 (weeks 3, 8, and 13).
cModel 1, p value for comparison of groups: T/S ? N/S vs. T/0 ? N/0 (weeks 3, 8, and 13).
observed from either once to twice per week or from twice
to three times per week (all p ? 0.05). No dose effect was
observed within any of the treatment groups when ana-
lyzed separately (all p ? 0.05) possibly because of the
smaller sample size per group compared to the entire
Figure 3B depicts the overall frequency dependence for
NRS pain responders from all four groups combined. A
trend of an increasing number of responders was observed
and a noticeable decrease in analgesia was detected dur-
ing the washout periods. This decrease in analgesia was
significant from week 3 to 4 (t ? ?2.787; p ? 0.007) and
from weeks 13 and 15 (t ? ?2.396; p ? 0.019). Together
these data suggest that increasing treatment frequency de-
creased pain and that the cessation of treatment increased
Codependence of outcome domains
To determine whether responder status was independent
within the pain, fatigue, and function domains, multiple 2 ?
2 ?2tests of treatment responders were performed. Signifi-
cant codependence in (1) pain–fatigue (X2 ? 15.352; p ?
0.001), (2) fatigue–function (X2 ? 10.469; p ? 0.001);
and (3) pain–function (X2 ? 10.045; p ? 0.002) was ob-
served. These data suggest that responder status in all three
outcome domains was highly interrelated.
Both randomized controlled trials, and longitudinal ob-
servation of individuals treated at tertiary care centers sug-
gest that fibromyalgia is difficult to treat.25,26As a result,
no single intervention is completely accepted by the med-
ical community, and no one intervention has significantly
outperformed others.27,28The present data provide evidence
that a subset of individuals obtained improvements in clin-
ical pain, fatigue, and function from acupuncture, although
efficacy was not related to needle placement or stimulation.
These results differed from Deluze et al.,11which may in
part be caused by their use of electro-acupuncture, which is
inherently difficult to blind and is a different treatment
modality from the manual stimulation used here.
Needle location and stimulation
When performing sham acupuncture, investigators some-
times place needles on nontraditional acupuncture points us-
ing a technique termed “minimal acupuncture.”6However
HARRIS ET AL.
TABLE 3. PERCENTAGE OF RESPONDERS BY ACUPUNCTURE GROUP AND WEEK
T/S T/0 N/SN/0 Total
Number (percentage) of responders for each outcome by treatment group and week are shown. Clinically significant improvements
in pain were observed in 25%–35% of subjects.
T/S, traditional needle location with manual stimulation; T/0, traditional needle location without manual stimulation; N/S, non-
traditional needle location with manual stimulation; N/0, nontraditional needle location without manual stimulation.
aPercentage of responders changed between first and last assessment week (week 4 to week 15).
bModel 2, p value for comparison of groups: T/S ? T/0 vs. N/S ? N/0 (weeks 3, 8, and 13).
cModel 2, p value for comparison of groups: T/S ? N/S vs. T/0 ? N/0 (weeks 3, 8, and 13).
needle insertion, irrespective of location, may elicit a re-
sponse that is functionally identical to that in traditional
points.29–33Indeed, sham needle sites have been shown to be
more effective than inert controls,34which implies that min-
imal acupuncture may not be inert. This may have been the
case for the nontraditional groups, and perhaps the use of a
more inert control intervention35,36would clarify this issue.
Acupuncture advocates may suggest that FM is an inap-
propriate condition for the study of the specifics of needle
placement. In Traditional Chinese Medicine, Ashi points are
commonly used for the treatment of pain originating away
from the major meridians, and needles are placed exactly
where the pain is located, regardless of whether the pain lies
on a meridian or does not correspond to a particular acupunc-
ture point.1Because FM pain is diffuse and located through-
out the body, our nontraditional locations may have actually
functioned as Ashi points.
Nonetheless no difference in clinical pain was detected
regardless of whether the needles were stimulated. This in-
dicates that the needle manipulation used in this study is not
an essential component to acupuncture therapy in FM, and
it follows that needle sensations elicited by manual manip-
ulation are not critical in this setting. However it should be
noted that some forms of acupuncture use no manual stim-
ulation. The form of acupuncture administered here was
highly simplified for experimental purposes; in a real clin-
ical setting, elements such as needle stimulation might en-
hance the effectiveness of this therapy.
A hallmark of effective pharmacologic interventions is
the dose–response effect. In theory, interventions that effect
changes in physiological substrates should result in dose de-
pendency, in which a greater effect is elicited with an in-
crease in the magnitude of the therapeutic intervention. In
exploratory analyses, data were reported supporting this hy-
pothesis for an overall effect of treatment frequency in FM.
This effect was modest, as it was only observed when all
four treatment groups were combined. With the absence of
a no-treatment arm, one cannot exclude regression to the
mean, natural history, or other time effects as possible mech-
anisms for the frequency dependence because time and dose
were confounded in this study. However this seems unlikely
for two reasons: first, an overall decay in analgesia was ob-
served during the wash-out periods (Fig. 3B), suggesting
that treatment was related to pain relief; and second, the nat-
ural trend of pain fluctuation in fibromyalgia does not match
the magnitudes observed here.25
Reduction of pain in fibromyalgia by a “placebo
effect” of acupuncture
In light of the fact that no differences were observed for
either stimulation or location of the needles within this study,
one may be tempted to conclude that the effects of acupunc-
ture occur largely because of a “placebo effect.” Indeed the
criteria used to designate responders from nonresponders
lies well within the range of the placebo effect.37The
placebo effect may have diluted any specific effects of
acupuncture observed in this trial; however further studies
will be required to confirm or refute these findings.
In this controlled trial, a subset of FM subjects received
symptom improvement with acupuncture needle insertion,
but no specific effects of needle placement or stimulation
were noted. Alternative designs containing more “inert”
comparison groups or trials with fewer dropout rates may
be more helpful in assessing which specific factors are nec-
essary to engender these effects. More specialized treatment
regimens tailored to individual FM participants may also
yield more positive findings.
ACUPUNCTURE FOR FIBROMYALGIA
quency, and baseline depressive symptoms on pain responders.
(A) Plot of percentage of NRS pain responders by group for in-
creasing treatment frequency (1? per week, 2? per week, and
3? per week). No one group consistently displayed more re-
sponders than another. (B) Plot of percentage NRS pain respon-
ders by increasing treatment frequency for all four groups com-
bined. An increasing trend for responders was found. Solid
symbols equal final week for each acupuncture dose (weeks 3, 8,
and 13) while open symbols are assessment weeks (weeks 4, 9,
and 15). Seventy two percent (72%) of responders after 3? per
week were also responders after 2? per week. NRS, numeric rat-
ing scale. acu., acupuncture.
Effects of needle location, stimulation, treatment fre-
The authors thank Jack Kalbfleisch, John Pezzullo, and
Cora MacPherson for their statistical advice and Dedra
Buchwald for careful reading of the manuscript. This pro-
ject was supported by the National Institutes of Health (NC-
CAM) grant R01 AT00004, the Department of Defense
DAMD-17002-0018, and Georgetown University GCRC
grant 98-186. R.E.H. was supported by National Institutes
of Health (NCCAM) K01 AT01111-01.
1. Xinnong C. Chinese Acupuncture and Moxibustion. Bejing:
Foreign Languages Press, 1999.
2. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alter-
native medicine use in the United States, 1990–1997: Re-
sults of a follow-up national survey. JAMA 1998;280:1569–
3. Emslie MJ, Campbell MK, Walker KA. Changes in public
awareness of, attitudes to, and use of complementary therapy
in North East Scotland: Surveys in 1993 and 1999. Comple-
ment Ther Med 2002;10:148–153.
4. NIH Consensus Conference. Acupuncture. JAMA 1998;280:
5. Streitberger K, Vickers A. Placebo in acupuncture trials. Pain
6. Stux G, Hammerschlag R. Clinical Acupuncture: Scientific
Basis. New York: Springer, 2001.
7. Ballegaard S, Pedersen F, Pietersen A, et al. Effects of
acupuncture in moderate, stable angina pectoris: A controlled
study. J Intern Med 1990;227:25–30.
8. Tavola T, Gala C, Conte G, Invernizzi G. Traditional Chinese
acupuncture in tension-type headache: a controlled study. Pain
9. Takeda W, Wessel J. Acupuncture for the treatment of pain of
osteoarthritic knees. Arthritis Care Res 1994;7:118–122.
10. Thomas M, Eriksson SV, Lundeberg T. A comparative study
of diazepam and acupuncture in patients with osteoarthritis
pain: A placebo controlled study. Am J Chin Med 1991;
11. Deluze C, Bosia L, Zirbs A, et al. Electroacupuncture in fi-
bromyalgia: results of a controlled trial. BMJ 1992;305:
12. Wolfe F, Ross K, Anderson J, et al. The prevalence and char-
acteristics of fibromyalgia in the general population. Arthritis
13. Jacobsen S, Bredkjaer SR. The prevalence of fibromyalgia and
widespread chronic musculoskeletal pain in the general pop-
ulation. Scand J Rheumatol 1992;21:261–263.
14. Wolfe F, Smythe HA, Yunus MB, et al. The American Col-
lege of Rheumatology 1990 Criteria for the Classification of
Fibromyalgia. Report of the Multicenter Criteria Committee.
Arthritis Rheum 1990;33:160–172.
15. Granges G, Littlejohn G. Pressure pain threshold in pain-free
subjects, in patients with chronic regional pain syndromes, and
in patients with fibromyalgia syndrome. Arthritis Rheum
16. Yunus MB. Towards a model of pathophysiology of fi-
bromyalgia: Aberrant central pain mechanisms with peripheral
modulation. J Rheumatol 1992;19:846–850.
17. Berman BM, Swyers JP. Complementary medicine treatments
for fibromyalgia syndrome. Baillieres Best Pract Res Clin
18. Waylonis GW. Long-term follow-up on patients with fibro-
sitis treated with acupuncture. Ohio State Med J 1977;
19. Lautenschlager J, Schnorrenberger CC, Muller W. Dtsch Zschr
20. Berman BM, Ezzo J, Hadhazy V, Swyers JP. Is acupuncture
effective in the treatment of fibromyalgia? J Fam Pract
21. Farrar JT, Young JP Jr, LaMoreaux L, et al. Clinical impor-
tance of changes in chronic pain intensity measured on an 11-
point numerical pain rating scale. Pain 2001;94:149–158.
22. Smets EM, Garssen B, Bonke B, De Haes JC. The Multidi-
mensional Fatigue Inventory (MFI) psychometric qualities
of an instrument to assess fatigue. J Psychosom Res 1995;
23. Jacobson NS, Truax P. Clinical significance: A statistical ap-
proach to defining meaningful change in psychotherapy re-
search. J Consult Clin Psychol 1991;59:12–19.
24. Ware JEJ, Snow K, Kosinski M, Gandek B. SF-36 Health Sur-
vey Manual and Interprestation Guide. Lincoln, RI: Quality-
Metric, Inc., 2000.
25. Wolfe F, Anderson J, Harkness D, et al. Health status and dis-
ease severity in fibromyalgia: Results of a six- center longitudi-
nal study [see comments]. Arthritis Rheum 1997;40:1571–1579.
26. Fitzcharles MA, Costa DD, Poyhia R. A study of standard care
in fibromyalgia syndrome: A favorable outcome. J Rheuma-
27. Crofford LJ, Appleton BE. The treatment of fibromyalgia:
A review of clinical trials. Curr Rheumatol Rep 2000;2:
28. Rossy LA, Buckelew SP, Dorr N, et al. A meta-analysis of
fibromyalgia treatment interventions. Ann Behav Med 1999;
29. Leibing E, Leonhardt U, Koster G, et al. Acupuncture treat-
ment of chronic low-back pain—a randomized, blinded,
placebo-controlled trial with 9-month follow-up. Pain 2002;
30. Goddard G, Karibe H, McNeill C, Villafuerte E. Acupuncture
and sham acupuncture reduce muscle pain in myofascial pain
patients. J Orofac Pain 2002;16:71–76.
31. Ernst E. Acupuncture as a symptomatic treatment of os-
teoarthritis. A systematic review. Scand J Rheumatol 1997;26:
32. Shlay JC, Chaloner K, Max MB, et al. Acupuncture and
amitriptyline for pain due to HIV-related peripheral neuropa-
thy: A randomized controlled trial. Terry Beirn Community
Programs for Clinical Research on AIDS. JAMA 1998;280:
33. Margolin A, Kleber HD, Avants SK, et al. Acupuncture for
the treatment of cocaine addiction: A randomized controlled
trial. JAMA 2002;287:55–63.
HARRIS ET AL.
34. Fargas-Babjak A. Acupuncture, transcutaneous electrical
nerve stimulation, and laser therapy in chronic pain. Clin J
35. Streitberger K, Kleinhenz J. Introducing a placebo needle into
acupuncture research. Lancet 1998;352:364–365.
36. Sherman KJ, Hogeboom CJ, Cherkin DC, Deyo RA. De-
scription and validation of a noninvasive placebo acupuncture
procedure. J Altern Complement Med 2002;8:11–19.
37. Turner JA, Deyo RA, Loeser JD, et al. The importance of
placebo effects in pain treatment and research. JAMA
Address reprint requests to:
Richard E. Harris, Ph.D.
Department of Internal Medicine
Division of Rheumatology
University of Michigan
Chronic Pain and Fatigue Research Center
24 Frank Lloyd Wright Drive, P.O. Box 385
Ann Arbor, MI 48106
ACUPUNCTURE FOR FIBROMYALGIA
This article has been cited by: Download full-text
1.Howard H. Moffet . 2009. Sham Acupuncture May Be as Efficacious as True Acupuncture: A Systematic Review of Clinical
TrialsSham Acupuncture May Be as Efficacious as True Acupuncture: A Systematic Review of Clinical Trials. The Journal of
Alternative and Complementary Medicine 15:3, 213-216. [Abstract] [PDF] [PDF Plus]
2.Vitaly Napadow , Andrew Ahn , John Longhurst , Lixing Lao , Elisabet Stener-Victorin , Richard Harris , Helene M. Langevin
. 2008. The Status and Future of Acupuncture Clinical ResearchThe Status and Future of Acupuncture Clinical Research. The
Journal of Alternative and Complementary Medicine 14:7, 861-869. [Abstract] [PDF] [PDF Plus]
3.Roland Staud. 2008. Mechanisms of acupuncture analgesia: Effective therapy for musculoskeletal pain?. Current Rheumatology
Reports 9:6, 473-481. [CrossRef]
4.Michael Francis Johnston, Bo Xiao, Ka-Kit Hui. 2007. Acupuncture and fatigue: Current basis for shared communication between
breast cancer survivors and providers. Journal of Cancer Survivorship 1:4, 306-312. [CrossRef]
5.Roland Staud. 2007. Treatment of fibromyalgia and its symptoms. Expert Opinion on Pharmacotherapy 8:11, 1629-1642. [CrossRef]
6.Daniel S Rooks. 2007. Fibromyalgia treatment update. Current Opinion in Internal Medicine 6:3, 328-334. [CrossRef]
7.Hélène M. Langevin , Richard Hammerschlag , Lixing Lao , Vitaly Napadow , Rosa N. Schnyer , Karen J. Sherman
. 2006. Controversies In Acupuncture Research: Selection of Controls and Outcome Measures In Acupuncture Clinical
TrialsControversies In Acupuncture Research: Selection of Controls and Outcome Measures In Acupuncture Clinical Trials. The
Journal of Alternative and Complementary Medicine 12:10, 943-953. [Citation] [PDF] [PDF Plus]
8.Dina Dadabhoy, Daniel J Clauw. 2006. Therapy Insight: fibromyalgia—a different type of pain needing a different type of
treatment. Nature Clinical Practice Rheumatology 2:7, 364-372. [CrossRef]
9.Roland Staud, Donald D Price. 2006. Mechanisms of acupuncture analgesia for clinical and experimental pain. Expert Review of
Neurotherapeutics 6:5, 661-667. [CrossRef]
10.Daniel J Clauw. 2006. Does acupuncture help reduce pain in patients with fibromyalgia?. Nature Clinical Practice Rheumatology
1:2, 76-77. [CrossRef]
11.2005. Literature WatchLiterature Watch. Alternative and Complementary Therapies 11:5, 273-274. [Citation] [PDF] [PDF Plus]