Complementary Therapies in Medicine (2008) 16, 87—91
available at www.sciencedirect.com
journal homepage: www.elsevierhealth.com/journals/ctim
Using acupuncture to treat depression:
A feasibility study
M. Whitinga, G. Leaveyb,c,∗, A. Scammella,S.Au
d, M. Kingc
aWandsworth Primary Care Research Centre, Wandsworth PCT, London, UK
bR&D Department, Barnet, Enﬁeld and Haringey Mental Health Trust, London, UK
cDepartment of Mental Health Sciences, University College London, London, UK
dTCM Healthcare, Middlesex, UK
Available online 29 September 2007
Objectives: To establish the feasibility of conducting a randomised controlled trial to evaluate
the efﬁcacy of acupuncture in the treatment of mild-to-moderate depression. Control group
intervention, blinding of treatment, outcome measures and the acceptability of such a trial
amongst participants were key factors to be addressed. The ﬁndings from this study will be
used to determine the design of a phase III randomised controlled trial.
Design: Nineteen participants were recruited through general practices in London, UK. Partici-
pants were randomly assigned on a 2:1 basis to either 12 sessions of verum acupuncture or 12
sessions of sham acupuncture (control). Sham acupuncture involved actual needling but at sites
considered to be unrelated to depression.
Main outcome measures: The Beck’s Depression Inventory (BDI) and the RAND 36 Item Health
Survey 1.0 (RAND) were completed at baseline and at the end of treatment or at treatment
dropout. All participants also attended a brief qualitative interview at the end of the study.
Results: Treatment dropout was low and there were high levels of patient enthusiasm
for a study of acupuncture. Referrals from General Practitioners (GPs) were lower than
expected. The sham control method successfully maintained participant blinding to treat-
ment and enabled the speciﬁc (or active) component of an acupuncture intervention to be
isolated and its efﬁcacy assessed. The outcome measures were sensitive enough to record
changes in depressive symptoms and quality of life and are appropriate for use in a larger
Conclusions: This feasibility study has provided important information that can be used to guide
the design and methodology of a full-randomised controlled trial.
© 2007 Elsevier Ltd. All rights reserved.
∗Corresponding author. Tel.: +44 208 442 6972.
E-mail address: Gerard.Leavey@beh-mht.nhs.uk (G. Leavey).
Depressive disorders are expected to become the second
highest cause of disease burden worldwide by 2020.1Around
70% of all primary care consultations are reported to involve
individuals with depressive disorder.2
0965-2299/$ — see front matter © 2007 Elsevier Ltd. All rights reserved.
88 M. Whiting et al.
Acupuncture may be a promising treatment in primary
care for individuals with depressive disorders3—5; however,
the evidence for use of this technique in such cases is incon-
sistent or inconclusive, and further trials are required.5—7
The objective of this study was to establish the feasibil-
ity of conducting a randomised controlled trial to evaluate
the efﬁcacy of acupuncture in the treatment of mild-
to-moderate depression. Previous report has highlighted
the need for greater methodological rigour in acupuncture
trials6; therefore, this feasibility study was designed to focus
on patient recruitment, to identify a suitable control group
intervention, to explore treatment blinding, and to ascer-
tain suitable outcome measures.
Materials and methods
Accrual of participants
Participants were recruited in seven general practices in
North London, UK. General Practitioners (GPs) provided
information to eligible participants during routine consul-
tations and posters and leaﬂets describing the study (and
inviting contact) were displayed in practice waiting rooms.
Participants were screened using a self-completion Beck’s
Depression Inventory (BDI) and those who scored at least 14
Following referral or expression of interest, the partici-
pants were screened for eligibility.
(1) Aged at least 18 years; (2) ‘‘mixed anxiety and depres-
sion’’ or ‘‘mild-to-moderate depressive episode with or
without somatic syndrome’’8as diagnosed using the Clinical
(1) Pregnancy; (2) history of substance abuse or brain dam-
age; (3) other psychiatric disorder preceding the onset of
depression; (4) in receipt of concurrent alternative treat-
ments or a talking therapy for repression; (5) in receipt of
pharmacological treatments for more than 3 months in the
Randomisation and treatment blinding
Participants with conﬁrmed eligibility attended a diagnostic
session with the acupuncturist and were then randomised.
The acupuncturist elicited the underlying imbalances of the
participants without reference to ‘depression’. A member
of the research team (S.A.) was independently responsible
for randomisation off-site, and used a block method on a
2:1 ratio with two-thirds of participants receiving verum
The member of the research team responsible for ran-
domisation informed the acupuncturist by telephone of
each patient’s randomisation after they had completed
their diagnostic session. All other researchers, including
assessors, and the participants were blinded to treatment.
The acupuncturist received training from a member of
the research team who has extensive clinical trial expe-
rience. The training included advice on how to maintain
professionalism and conﬁdentiality about assignment of
Participants were randomised to receive either 12 sessions
of manual verum acupuncture or 12 sessions of manual
sham acupuncture (control). Each session lasted 40 min
and was provided during usual clinic hours by an accred-
ited TCM Healthcare practitioner with 10 years’ clinical
Verum acupuncture—–rationale and method
A formula combined with two additional discretionary points
was used in the acupuncture treatment. Box 1 (pub-
lished in online version at http://sciencedirect.com/ as
supplementary data) shows the points used in the verum
group, the range of needling depth, and the dimensions of
the needles used. Choice of points and needling depth was
individualised according to TCM Healthcare principles10 and
body mass of the participant.
Sham acupuncture—–rationale and method
Sham acupuncture involves actual needling but at sites con-
sidered to be unrelated to depression. The same sham points
were used on all participants in the sham group (Box 1,
published in online version at http://sciencedirect.com/ as
supplementary data). In addition, only shallow needling was
undertaken, there was no needle stimulation after insertion
and de-qi was avoided.
Needle retention time was 20 min for both verum and
sham control treatments.
The primary outcome measure was the BDI.11 The secondary
outcome measure was the RAND 36 Item Health Survey
1.0 (RAND).12 Both the BDI and the RAND were completed
at baseline and at the end of treatment or at treatment
dropout. All participants also attended a brief qualita-
tive interview at the end of the study. The interview was
designed to explore the participant’s experiences of the
study. The participants were also asked to indicate to which
study group they believed they had been assigned. All base-
line outcome measures were assessed face-to-face at the
participant’s general practice. Where possible, end point
assessments were also conducted face-to-face. In cases
where this was not possible, end point assessments were
conducted by telephone.
Brief descriptive analysis of the BDI and the RAND was used
to examine changes in mean scores. Qualitative data were
coded into the categories ‘acupuncture treatment had a
positive impact’, ‘acupuncture treatment had a negative
impact’ and ‘acupuncture treatment was ineffective’.
Using acupuncture to treat depression: A feasibility study 89
Ethical approval was obtained from Barnet, Enﬁeld and
Haringey Local Research EthicsCommittee. All participants
provided signed informed consent.
Twenty patients were recruited into the study over a period
of 15 months. One patient withdrew from the study before
randomisation. Thirteen participants were randomised into
the verum group and six into the sham-control group. End-
point data were available for all 13 participants in the verum
group and for four of the six participants in the sham-
control group. The ﬂow of participants through each stage
of the study is shown in Fig. 1. Most participants were self-
The mean age of participants was 39.9 years (S.D. = 12.1) in
the verum group (nine female, four male) and 48.5 years
(S.D. = 10.2) in the sham-control group (ﬁve female, one
male). Participants in the verum group had slightly higher
depression scores on the BDI and slightly lower wellbeing
scores on the RAND than participants in the sham-control
group (Table 1). The TCM Healthcare diagnosis of each
participant’s condition at enrolment is given in Box 2 (pub-
lished in online version at http://sciencedirect.com/ as
Adherence to treatment
Treatment adherence was good: 11 of 19 participants (58%)
completed all 13 study sessions and 16 of 19 participants
(84%) completed at least 6 study sessions (Table S1, pub-
lished in online version at http://sciencedirect.com/ as
supplementary data). There was no correlation between
early treatment dropout of participants and the treatment
they were receiving.
There was a decrease in mean scores on the BDI in both study
groups, which indicates a reduction in depressive symptoms.
There was also a slight increase in mean scores on the RAND
in both study groups, which indicates an improvement in
quality of life; however, a reduction in mean scores for the
RAND categories of ‘physical functioning’ and ‘pain’ was
seen in the sham-control group only.
All participants welcomed the opportunity to participate
in the current study. There were no differences between
the verum and sham-control groups with regard to self-
Figure 1 Flow of participants through the trial.
90 M. Whiting et al.
Table 1 Mean scores (S.D.) in BDI and RAND 36 Item Health Survey 1.0 at baseline and end point
Verum baseline (n=13) Verum end point (n=13) Sham baseline (n= 6) Sham end point (n=4)
BDI 30.7 (5.3) 21.3 (10.5) 24.7 (8.4) 13.5 (11.6)
RAND 36 Item Health Survey
Physical functioning 63.2 (27.8) 64.6 (33.9) 73.5 (24.2) 61.1 (41.8)
Role limitations physical 26.9 (29.8) 50.7 (48.5) 54.2 (40.1) 75.0 (35.4)
Role limitations emotional 2.6 (9.6) 47.2 (43.7) 27.8 (39.0) 73.3 (43.5)
Energy/fatigue 21.0 (12.5) 35.3 (22.9) 25.3 (33.2) 62.3 (23.0)
Emotional well-being 29.8 (11.1) 50.2 (23.7) 30.7 (13.5) 55.2 (29.3)
Social functioning 25.0 (15.3) 46.9 (23.9) 31.3 (29.5) 55.0 (36.0)
Pain 51.7 (32.3) 63.3 (35.5) 63.3 (29.3) 56.0 (21.0)
General health 34.2 (16.1) 46.3 (15.8) 50.8 (28.9) 60.0 (12.4)
reported positive or negative impacts of acupuncture.
Eleven participants (eight verum, three sham) reported
that the acupuncture treatment had a positive impact on
their wellbeing, particularly on levels of stress and energy.
One participant in the verum group reported that nee-
dle insertion caused pain, and two participants in the
verum group reported that the acupuncture process had
made their depressive symptoms worse. Three other partic-
ipants (two verum, one sham) considered acupuncture to be
There was no correlation between a participant’s ability to
judge correctly their study group and the treatment they
This feasibility study has increased our understanding of the
challenges associated with conducting a full trial of the efﬁ-
cacy of acupuncture in the treatment of depression, and
has provided important guidance on how to proceed. The
study has been particularly useful in providing information
about recruitment, control group design and outcome mea-
sures. The outcome measures used in this study appear to
be suitable for use in a larger trial. In such a trial, the
BDI and RAND will be completed following an extended
follow-up period to clarify the long-term efﬁcacy of verum
acupuncture. The main purpose of this feasibility study was
to provide methodological guidance for future work. The
study comprised only a small population; therefore, the
ﬁndings of the study need to be treated with caution and
no deﬁnitive conclusions about the efﬁcacy of acupuncture
in the treatment of depression can be drawn. The number
of GP referrals into this study was lower than expected,
which is consistent with other trials,13 and may be a prob-
lem related to time pressures and workload priorities14,15
rather than a lack of commitment to the study. Most partic-
ipants were self-referred after seeing an advertisement for
the study in their practice waiting rooms. More consistent
referral channels will need to be established for recruitment
of participants into a full trial. Retrospective recruitment
methods have been used successfully in Complementary and
Alternative Medicine (CAM) trials16; however, these meth-
ods were not used in the current study because many
recorded patients may be on alternative treatments that
would exclude them from the study, and retrospective
recruitment would exclude patients who have not yet been
diagnosed with depression by the GP. Recruitment of partic-
ipants into a large phase III trial is recommended to occur
through a combination of patient screening in general prac-
tice waiting rooms and targeted advertising in primary care
The use of a sham control method is recommended
in a deﬁnitive randomised controlled trial. However, non-
acupuncture points should also be used to show that the
sham control needling sites are completely ineffective in
the treatment of depression.
There are several concerns about the use of sham control
interventions in acupuncture studies. For example, sham
control methods may be perceived to reduce the experi-
ence of acupuncture solely to needling, rather than enabling
the holistic nature of acupuncture to be assessed.17 Also,
the possibility that ‘expectation’ of treatment has a phys-
iological effect on the brain, thus mediating a potentially
powerful non-speciﬁc clinical response to acupuncture, has
been suggested during pain trials.18
The study described here was designed to separate the
non-speciﬁc effects of the intervention from the effects of
needling at the correct depth in the correct place and with
manual stimulation, all of which are speciﬁc to acupunc-
ture treatment. Furthermore, the study showed that the
sham method was a plausible intervention for the control
group and successfully maintained participant blinding to
treatment, an issue that is central to improving the method-
ological quality of trials assessing the effects of acupuncture
in individuals with depression.6
The use of a sham-control group also enabled the speciﬁc
(or active) component of an acupuncture intervention to be
isolated and its efﬁcacy assessed.
Although practitioner engagement is an essential part
of acupuncture,19 correct needling must be considered
the most important factor. Examination of the impact of
inserting needles into speciﬁc energy points in the body
is imperative and will provide valuable evidence about the
efﬁcacy of acupuncture.
Using acupuncture to treat depression: A feasibility study 91
This feasibility study has provided important information
that can be used to guide the design and methodology of
a full-randomised controlled trial.
The authors would like to thank the acupuncturist, study
participants and GPs. The North Central London Community
Research Consortium funded this study.
Appendix A. Supplementary data
Supplementary data associated with this arti-
cle can be found, in the online version, at
1. Brown P. Effective treatments for mental illness are not being
used, WHO says. BMJ 2001;323:769.
2. Robinson WD, Geske JA, Prest LA, Barnacle R. Depression treat-
ment in primary care. J Am Board Fam Pract 2005;18:79—86.
3. Ernst E, Rand JI, Stevinson C. Complementary thera-
pies for depression: an overview. Arch Gen Psychiatry
4. Jorm AF, Christensen H, Grifﬁths KM, Rodgers B. Effectiveness
of complementary and self-help treatments for depression.
Med J Aust 2002;176(Suppl):S84—96.
5. Hagelskamp C, Scammell A, Au S, Leavey G. Acupuncture as
treatment for depression in primary care: current position and
future hopes. Primary Care Ment Health 2003;1:75—9.
6. Smith CA, Hay PPJ. Acupuncture for depression. Cochrane
Database Syst Rev 2004, 3 CD004046.pub2.
7. Mukaino Y, Park J, White A, Ernst E. The effectiveness
of acupuncture for depression—–a systematic review of ran-
domised controlled trials. Acupunct Med 2005;23:70—6.
8. World Health Organisation. International Statistical Classiﬁca-
tion of Disease and Related Health Problems, 10th Revision.
retrieved 3rd October 2006.
9. Lewis G, Peloise AJ, Araya R, Dunn G. Measuring psychiatric
disorder in the community: a standardised assessment for use
by lay interviewers. Psychol Med 1992;22:465—86.
10. DeShen W. Manual of international standardisation of acupunc-
ture. Beijing: China Academy of Traditional Chinese Medicine;
11. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An
inventory for measuring depression. Arch Gen Psychiatry
12. Hays RD, Sherbourne CD, Mazel R. The RAND 36-item health
survey 1.0. Health Econ 1993;2:217—27.
13. MacPherson H, Thorpe L, Thomas K, Geddes D. Acupuncture
for depression: ﬁrst steps toward clinical evaluation. J Altern
Complement Med 2005;10:1083—91.
14. Croughan M. Factors inﬂuencing physician participation in
practice-based research network studies—–a call for further
research. J Fam Pract 2001;50:978—9.
15. Bell-Syer SEM, Klaber Moffett JAK. Recruiting patients to ran-
domized trials in primary care: principles and case study. Fam
16. McCarney R, Fisher P, van Haselen R. Accruing large numbers
of patients in primary care trials by retrospective recruitment
methods. Complement Ther Med 2002;10:63—8.
17. Paterson C, Dieppe P. Characteristic and incidental (placebo)
effects in complex interventions such as acupuncture. BMJ
18. Pariente J, White P, Frackowiak RSJ, Lewith G. Expectancy
and belief modulate the neuronal substrates of pain treated
by acupuncture. Neuroimage 2005;25:1161—7.
19. Gould AJ, MacPherson H. Patient perspectives on outcomes fol-
lowing treatment with acupuncture. J Altern Complement Med