doi: 10.1111/j.1742-1241.2007.01553.

SYSTEMATIC REVIEW

Massage therapy for the treatment of depression: a systematic review

H. F. Coelho, K. Boddy, E. Ernst

SUMMARY Background: People with depressive disorders or subsyndromal symptoms of depression (SSD) often use complementary and alternative therapies, including massage therapy (MT). This systematic review evaluates the evidence, from randomised clinical trials (RCTs), for the effectiveness of multiple sessions of classical European (Swedish) MT for the treatment of depression. Methods: Eligible RCTs were identied via eight electronic databases and manual searches of references. Two reviewers independently selected trials, assessed trial quality and extracted data. Results: Four RCTs met our inclusion criteria. Three of these RCTs compared MT with relaxation therapies, but provided insufcient data and analyses to contribute meaningfully to the evaluation of MT for depression. The fourth included RCT used MT as a control condition to evaluate a depression-specic acupuncture treatment. This trial provided limited evidence that, in the early stages of treatment, MT is less effective than acupuncture for treating depression, a treatment which itself is not accepted for this condition. Conclusions: Despite previous research suggesting that MT may be an effective treatment for depression, there is currently a lack of evidence to support this assertion from RCTs that have selected participants for depression or SSD.

Review Criteria
The study is a systematic review conducted according to a predened protocol. RCTs were identied using prespecied search terms in eight electronic databases. Reference lists of retrieved articles were manually searched to identify further trials.

Complementary Medicine, Peninsula Medical School, Universities of Exeter and Plymouth, Exeter, UK Correspondence to: Helen F. Coelho, PhD, Research Fellow, Complementary Medicine, Universities of Exeter and Plymouth, 25, Victoria Park Road, Exeter EX2 4NT, UK Tel.: + 44 1392 424942 Fax: + 44 1392 424989 Email: helen.coelho@pms.ac.uk Disclosures There are no conicts of interest to declare.

Message for the Clinic

Patients who are using (or considering using) MT for treating depressive symptoms should be provided with the current effectiveness evidence. Whilst a lack of evidence does not mean that MT is ineffective, patients should be informed that the only systematic and critical appraisal of RCTs demonstrates that there is currently no robust evidence to recommend a course of MT to alleviate symptoms of depression.

Introduction
A large number of people with clinically signicant depression do not receive formal treatment. In Europe, only 36.5% (95% CI: 32.540.5) of people with a 12-month mood disorder contact formal healthcare services, and 15.1% (95% CI: 10.619.7) of these do not receive any treatment (1). Subsyndromal depressive symptoms (SSD) are also common, are associated with reduced functioning and quality of life, and can occur as prodromal and residual (interepisodic) features of major depression (2,3). Many people use a complementary or alternative therapy (CAT) to treat their symptoms of depression. Surveys indicate that 53.6% of those with severe depression (4) and 22.4% of those with major depression (5) have used CAT to treat their symptoms in the preceding 12 months. Massage therapy (MT) is a popular CAT used by those with depression. Amongst those with severe depression, 2.1% use MT to treat their symptoms (4). The use of MT declined in the early 20th century (at the time of the pharmaceutical revolution), but its popularity has been regained, with

consumers spending increasing amounts on this treatment (6). The bio-physiological rationale for using MT for depressed mood is twofold. First, electroencephalogram activation changes to a symmetrical or left frontal pattern with MT. Such a pattern is associated with happy affect. Second, MT stimulates facial expressions and increases vagal activity, subsequently reducing depressed affect (7). There are two systematic reviews of all complementary and alternative medicine (CAM) for depression which have included MT. However, one of these reviews did not critically evaluate the included studies (8) and the other only identied one trial of MT that had selected participants for depressive symptoms (9). A meta-analysis evaluating the effectiveness of MT for all indications reported largest effects for the reduction of depression and anxiety. However, data from general medical populations who were not specically selected for having depression were included in the meta-analysis (10). Because of the limitations of these previous reviews (810) it is still not clear whether MT is effective for treating those with depressive disorders or SSD. Thus, there is a

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

325

326

MT for the treatment of depression

need for a systematic and critical review of the evidence for MT as a treatment for depression that focuses on randomised clinical trials (RCTs) using depressed populations. The objective of the current study was to address this need. Our review concentrates on multiple sessions of classical European (Swedish) massage, as a sole or adjunctive treatment, for depressive disorders or SSD.

Materials and methods

Search
The rst author searched PUBMED, EMBASE, SCOPUS, ISI Web of Science, CENTRAL, PsychInfo, CINAHL and AMED from inception until June 2006, using a predened search strategy (Table 1). No language restrictions were imposed. The reference lists of all articles retrieved in full were also searched.

Study selection
Abstracts of all articles identied from electronic searches were read. Articles clearly not meeting inclusion criteria were ltered out. All other articles were obtained in full (Figure 1). HFC and KB independently considered articles written in English. EE and an independent reviewer considered articles written in other languages. Inclusion of a trial was subject to both reviewers reaching agreement that the inclusion criteria were met. Inclusion criteria were: (i) the study was a clinical trial, described as randomised; (ii) participants were selected for having a depressive disorder or SSD. Trials were excluded if participants with bipolar disorder or cyclothymic disorder were included; (iii) the intervention resembled classical European (Swedish)

MT, delivered as a series of sessions, and therefore excluding single-session trials or analyses. Trials investigating self-massage techniques and MT involving the use of mechanical devices were excluded. The primary reason for excluding mechanical devices was the fact that skin-to-skin contact, which plays a major role in classical MT, would be either reduced or absent. RCTs of MT administered as part of a multi-component therapy (including trials using aromatherapy oils) were also excluded, unless the trial design enabled the evaluation of the MT component as an adjunctive treatment [see (iv) below]. Trials providing background music, which is often utilised in MT and considered to be part of the therapeutic session, were included; (iv) the study design enabled the evaluation of the sole or additive benet of MT by using at least one appropriate control condition. Acceptable controls for sole MT included comparison treatments not involving massage, attention controls, and no treatment or waiting list groups. When MT was part of a multi-component therapy, control conditions must have been identical therapy minus the MT component; (v) there was at least one measure of the depressive disorder or SSD, or a clinical aspect or construct of such a disorder, and the measurement was made before treatment and at least once after treatment. Physiological outcomes were excluded.

Data extraction
Two reviewers (HFC and KB) independently extracted data from all included studies using data extraction forms specically designed for the study. The data extracted corresponds with those given in Tables 2 and 3. Disagreements between reviewers were resolved through discussion.

Quality assessment
Table 1 Electronic database search strategy

Steps for searching electronic databases 1. Massage OR massotherapy OR masso-therapy OR self-massage OR soft-tissue mobilization OR soft-tissue mobilisation OR soft tissue mobilization OR soft tissue mobilisation OR soft-tissue manipulation OR soft tissue manipulation 2. Depression OR depressive OR MDD OR PND OR dysthymic OR dysthymia OR affective OR mood 3. #1 AND #2 4. #3 AND (randomised OR randomized OR trial) For databases enabling limits 5. #3 AND limit RCT/clinical trial 6. #4 OR #5 When searching CENTRAL only steps 13 of the search strategy will be performed (because CENTRAL is a register of controlled trials)

The Jadad et al. scale (11), modied to allocate one point for single blinding of the outcome assessor rather then for double blinding (maximum score 4), was used to evaluate the methodological quality of each included RCT. To supplement the Jadad scale, Table 2 provides additional methodological details for each trial, such as those described in Verhagen et al. (12).

Data presentation and interpretation

Table 3 provides general information and the main ndings of all included trials. Only results of between-group comparisons, using appropriate statistical techniques to evaluate the effectiveness of multiple-session MT, are included. If a study did not provide between-group comparisons, but presented sufcient data, we analysed the data using

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

MT for the treatment of depression

327

Unique articles identified from literature searches: n = 595 Articles excluded from abstracts: n = 562 Full text articles obtained for detailed evaluation: n = 33 Studies excluded (n = 30) Reasons for exclusion: (1) Not selected for depression: n = 20 (2) Review /commentary on excluded trial: n = 2 (3) Ongoing RCT: n = 1 (4) Participants with cylothymia/bipolar disorder included: n = 1 (5) Single session of MT and physiological outcome only: n = 1 (6) (1) and not randomised: n = 1 (7) (1) and outcome not depression relevant: n=1 (8) (1) and not MT: n = 1 (9) (1) and excluded type of MT: n = 1 (10) (6) and excluded type of MT: n = 1 Additional RCTs identified for inclusion from reference lists: n = 1

RCTs identified for inclusion: n = 3

All included RCTs: n = 4

Figure 1 Selection of articles for inclusion

software from the Cochrane Collaboration (RevMan 4.2).

sessions ranging from 20 to 30 min. None of the included trials reported using background music as part of the MT.

Results
Thirty-three articles were retrieved in full for a detailed evaluation (Figure 1). Of these, 30 were excluded. Reasons for exclusion were: the study was a trial but was not randomised (13,14), participants were not all selected for depressive symptoms (1337), participants with bipolar disorder or cyclothymia were included within the depressed sample (38), the intervention was not MT (36), the intervention was an excluded type of MT (14,28), the intervention was a single session of MT (39), there were no non-physiological outcomes (39), the outcome was not depression or a symptom of depression (37), the study was a review of, commentary on, or record of, another trial which had already been excluded (40,41), the study was an ongoing RCT (42). The remaining three RCTs were included in this review. One further RCT was identied from manual searches of reference lists. All four of the included RCTs were reported in English. Jadad scores ranged from 1 to 3 (Table 2). The number of MT sessions ranged from 5 to 32 sessions, with the length of the

MT vs. relaxation therapy

Three of the included trials compared MT with relaxation therapy (4345). Participants were depressed adolescents (43), postnatally depressed women (44) and women with depression during pregnancy (45). The latter trial included a second control group who received usual prenatal care (Table 3). The three studies (4345) used multiple outcome measures at various time points, resulting in numerous analyses: for each outcome, both within- and between-group analyses were conducted to assess change over the rst and the last session (single-session effects), and to assess change over the series of sessions (multiple-session effects). However, for the outcomes relevant to this review, all three studies omitted to report the between-group, pair-wise analyses that would have enabled evaluation of multiplesession MT. Also, all three of the studies comparing MT with relaxation therapies did not provide standard deviations alongside the mean values given in tables (4345), precluding reanalysis of the data.

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

328

MT for the treatment of depression

Jadad score (modied, maximum score 4)

MT vs. acupuncture
Highlighted columns combine to determine the modied Jadad score. *Blinding of outcome assessors where possible (not for observations during sessions). Blinding of outcome assessors was only mentioned in relation to excluded physiological outcomes, unclear if blinding occurred for remaining outcomes. Described as modied intent-to-treat, did not include all participants initially randomised but all participants attending at least one assessment.

Table 2 Methodological quality of trials

Was the trial randomised?

Yes Yes Yes Yes

In the fourth RCT, MT was used as a control condition when investigating the effectiveness of a depression-specic acupuncture treatment (SPEC) for depressed pregnant women (46). A non-specic acupuncture control condition (NSPEC) was also used. Analyses conducted for an 8-week acute treatment phase found that SPEC resulted in a greater proportion of participants responding to treatment than MT (see Table 3 for criteria used to dene treatment response; Fisher exact p 0.03, number to treat (NTT)2 effect size 2.7, 95% CI: 1.516.0). Because of the fact that MT and NSPEC were intended as control treatments, similar pair-wise analysis were not conducted between these two groups. Using the data provided in the trial report (46), our analysis demonstrated no signicant difference between MT and NSPEC in proportion of participants responding to treatment following the acute treatment phase (OR 0.51, 95% CI: 0.141.92). The authors of the trial (46) analysed Beck Depression Inventory (BDI) data, obtained weekly during the acute treatment phase, by tting individual regression lines to each participants data (to represent individual change in BDI scores), and then comparing rates of change between groups using t-tests. These t-tests demonstrated a greater rate of improvement with SPEC than MT during the rst month of acute treatment (t 1.72, p 0.047 and d 0.70), but not during the second month (p 0.083, d 0.54). Similar t-tests were not conducted between MT and NSPEC, but insufcient data were available to enable reanalyses. Following the acute treatment phase, Manber et al. (46) offered continuation treatment, to treatment responders, until 8 weeks after labour. Participants who did not receive continuation treatment also participated in the continuation phase assessments and were included in subsequent analyses. Using baseline and postcontinuation-phase data, repeated measures ANOVA demonstrated no signicant group by time interaction effects on the BDI or the Hamilton Rating Scale for Depression. Likewise, pair-wise analyses conducted for this review revealed no signicant difference between MT and SPEC, or between MT and NSPEC, in the odds of achieving full remission (OR 0.33, 95% CI: 0.052.10; OR 2.00, 95% CI: 0.498.24 respectively; see Table 3 for criteria used to dene full remission).

Was a power calculation described? Was an analysis conducted on the intent-to-treat sample? In addition to stating the number of withdrawals/dropouts, were reasons given for each group? Was the number of withdrawals/ dropouts in each group mentioned? Were blind outcome assessments conducted? Were groups similar at baseline on prognostic indicators? Was the treatment allocation concealed? Was the randomisation procedure described, and was it appropriate?

Not Not Not Not

described described described described

Not Not Not Not

clear clear clear clear

Yes Yes Not clear Yes

Yes* Not clear Not clear Yes

No No No Yes

No No No Yes

Not clear Not clear Not clear No

No No No No

2 1 1 3

Field et al. (43) Field et al. (44) Field et al. (45) Manber et al. (46)

Discussion
This systematic review cannot generate conclusive evidence for the effectiveness of MT as a treatment for depression. We identied only four relevant

References

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

Table 3 Randomised clinical trials of massage therapy for the treatment of depression

References

Depression type Participants

MT details

Comparison details

Timeframe

Relevant measures (and assessment times)

Results

Field et al. (43) Depression or dysthymia

5 days; 5 sessions (B) Relaxing videotape watched with psychology (1 30 min session per day) student; n 10 (B) Relaxation therapy (yoga + PMR); n 16 5 weeks; 10 sessions (2 30 min sessions per week)

Field et al. (44) Postnatal depression

Pre- and postsession at rst Analyses were unclear. and last session: Insufcient data provided (1) Behaviour Observation Scales in tables to enable reanalysis (2) POMS Analyses were unclear. Pre- and postsession at rst Insufcient data provided and last session: (1) Behaviour Observation Scales in tables to enable reanalysis (2) POMS Signicant group time (over multiple sessions) MANOVA [F (3,66) 24.32, p < 0.001], reported as favouring massage, but appropriate pair-wise analyses not provided. Insufcient data provided in tables to enable reanalysis

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

Field et al. (45) Depression in pregnancy

Hospitalised adolescents (A) Clothed back MT; meeting DSM-III-R administered by criteria for depression psychology students; or dysthymia; n 36 n 26 Adolescent mothers (A) MT to head, neck, meeting DIS criteria arms, hands, torso, for dysthymia; BDI legs, feet; administered score > 16; n 32 by trained therapist; n 16 Pregnant women. (A) MT to head, neck, arms, Criteria for establishing hands, back, legs, feet; depression not administered by signicant provided; n 84 others (taught by trained therapist); n 28 (B) PMR; n 28 (C) Usual prenatal care; n 28

16 weeks; 32 sessions Pre- and postsession at rst (2 20 min sessions and last session: (1) POMS per week) At rst and last session: (2) CES-D

MT for the treatment of depression 329

330

Table 3 (continued )

References Depression type Participants

MT details

Comparison details

Timeframe

Relevant measures (and assessment times) Results

MT for the treatment of depression

Manber et al. (46)

Depression in pregnancy

Pre-, mid- and postacute (A) MT to back, Pregnant women (B) Depression-specic 8-week acute treatment phase: face, head, meeting DSM-IV acupuncture; n 20 phase; 12 sessions (1) Response to treatment neck, shoulders, criteria for a (2530 min sessions); (C) Non-specic feet; unclear who acupuncture; n 21 continuation phase: 817 dened by: current, Failure to meet DSM-IV MDD administered non-psychotic, additional sessions for criteria treatment*; major depression acute phase responders 50% reduction in HRSD episode; HRSD n 20 score score 14; n 61 HRSD score 14 Weekly during acute phase: (2) BDI Postcontinuation phase: (3) HRSD (4) BDI (5) Full remission dened by HRSD score 8

Acute phase results: (1) B favoured over A (Fisher exact p 0.03, NTT2 effect size 2.7, 95% CI: 1.516 Pair-wise contrast between A and C not provided. Analysis conducted for this review found no signicant difference between A and C (OR 0.51, 95% CI: 0.141.92) (2) Rate of change greater in B than A in 1st month of treatment (t 1.72, p 0.047, d 0.70), but NSD between A and B in 2nd month (p 0.083, d 0.54) Pair-wise contrasts between A and C not provided. Insufcient data provided to enable analysis Continuation phase results: (3) Repeated measures ANOVA demonstrated no signicant treatment by time interaction effect [F(2,41) 2.2, p 0.12] (4) Repeated measures ANOVA demonstrated no signicant treatment by time interaction effect [F(2,41) 1.6, p 0.21] (5) Pair-wise contrasts between A and B, and between A and C, not provided. Analyses conducted for this review found no signicant difference between A and B (OR 0.33, 95% CI: 0.052.10), or between A and C (OR 2.00, 95% CI: 0.498.24)

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

*MT details were not included in the trial report but were provided by R. Manber (personal communication). POMS, Prole of Mood States; CES-D, Centre for Epidemiologic Studies Depression Scale; BDI, Beck Depression Inventory; HRSD, Hamilton Rating Scale for Depression.

MT for the treatment of depression

331

trials, all of which, through methodological, reporting or design limitations, could not be used to evaluate the effectiveness of MT as a treatment for depression. Three of the studies (4345) either did not conduct, or did not clearly describe, appropriate between-group analyses to enable evaluation of the multiple-session effects of MT. The fourth trial evaluated acupuncture (MT was intended as a control treatment) (46). The authors state that their study is not a realistic assessment of MT, because the MT sessions were designed to be comparable in length to the SPEC and NSPEC sessions, and were therefore shorter than is typical in clinical practice (46). However, the MT sessions resembled classical European massage (R. Manber, personal communication) and were of a similar length to those used in the other trials (4345). There is also little information on optimal dosing for MT for the treatment of depression, and the data from all trials should be considered irrespective of the length of treatment sessions. Nevertheless, the results (46) have little bearing on the conclusions that can be drawn about the effectiveness of MT for the treatment of depression because there is currently insufcient evidence to suggest that acupuncture is effective for this condition (47). Comparison between acupuncture and MT only serves as a comparison between two treatments of unknown efcacy. Our nding that there is currently no good evidence to suggest that MT is an effective treatment for depression differs from the conclusions of two previous reviews, one of which focused on all CAM for depression (8), and the other of which focused on MT for all indications (10). The latter review concluded that reductions in trait anxiety and depression following a course of treatment were among MTs largest effects (10), but included trials where depression was measured as an outcome even if participants were not selected for having depressive symptoms, and did not include any of the four trials identied here. The other review (8) summarised two of the trials included here (43,44) and concluded that MT appears to have short-term benets, and that its longer-term effects have not been evaluated. It is apparent that the reviewers include both singlesession and multiple-session effects under short-term benets, reserving the phrase longer-term effects for the long-term follow-up of participants. The most likely reason for the different conclusions drawn in this previous review (8) and by ourselves is that by focusing on a specic therapy we could devote our review to the critical evaluation of the methodology and analyses of the included trials. In providing a general overview of CAM for depression,

the authors (8) succinctly summarised, but did not critically appraise, the available data for each therapy. Our ndings do support, and indeed update, the conclusions of another previous review (9), which identied one of the RCTs included here (43), and concluded that the data are insufcient to judge the value of massage for depression. Thus, despite the popularity of MT, and the identication of three more trials, including two that have been conducted since the publication of this previous review (9), the conclusions still stand. It is not entirely unexpected that a critical evaluation of MT trials would identify methodological problems and lead to difculties in drawing conclusions regarding the effectiveness of MT. This is because it is difcult to provide an adequate control condition for this intervention. However, direct comparisons can be made between MT and standard evidence-based treatments, such as cognitive-behavioural treatment (CBT). However, it is still impossible to ensure that patients and therapists are blind to treatment allocation in such trials. This is undoubtedly an obstacle in MT research, and may account for the general paucity of trials in this area. Some limitations of our review require discussion. We did not conduct hand searches of the literature, nor did we contact relevant experts or organisations to identify further trials. However, grey literature does appear in some of the databases searched. Although we did not exclude non-English language articles, we did not search non-English language databases so we may have unknowingly excluded trials. Finally, our standardised quality assessments reect the information provided in the trial reports, and may not fairly represent the trials themselves (11). We did supplement the Jadad score with other methodological information, but this too was based upon the reports rather than the trials.

Conclusions
Although a recent review (8) and meta-analysis (10) have indicated that MT may be an effective treatment for depression, this critical evaluation demonstrates a lack of evidence, from RCTs which have selected participants with depressive symptoms, for multiple-session Swedish MT (as either a sole or additive treatment). Of course, lack of evidence does not equate to lack of effectiveness, and Swedish MT delivered by a trained therapist is generally safe (48). Potential indirect risks of MT should also be considered (e.g. hindering access to mainstream services), particularly when patients use MT on its own. Any future trials aiming to estab-

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

332

MT for the treatment of depression

lish whether MT is benecial for the treatment of depression should draw upon, and seek to improve, the methodological and reporting limitations of the previous research.

Acknowledgements
HFC has a research fellowship from the Pilkington Charitable Trusts, who had no role in this research other than in nancing the fellowship. We would like to thank Barbara Wider who, in addition to EE, independently considered non-English language trial reports for inclusion in this review.

References
1 Alonso J, Angermeyer MC, Bernert S et al. Use of mental health services in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr Scand 2004; 109 (Suppl. 420): 4754. 2 Judd LL, Akiskal HS, Paulus MP. The role and clinical signicance of subsyndromal depressive symptoms (SSD) in unipolar major depressive disorder. J Affect Disord 1997; 45: 518. 3 Kennedy N, Abbott R, Paykel ES. Longitudinal syndromal and sub-syndromal symptoms after severe depression: 10-year followup study. Br J Psychiatry 2004; 184: 3306. 4 Kessler RC, Soukup J, Davis RB et al. The use of complementary and alternative therapies to treat anxiety and depression in the United States. Am J Psychiatry 2001; 158: 28994. 5 Unutzer J, Klap R, Sturm R et al. Mental disorders and the use of alternative medicine: results from a national survey. Am J Psychiatry 2000; 157: 18517. 6 Eisenberg DM, Davis RB, Ettner SL et al. Trends in alternative medicine use in the united states, 19901997. Results of a followup national survey. JAMA 1998; 280: 156975. 7 Field TM. Massage therapy effects. Am Psychol 1998; 53: 127081. 8 Jorm AF, Christensen H, Grifths KM, Rodgers B. Effectiveness of complementary and self-help treatments for depression. Med J Aust 2002; 176: S8496. 9 Ernst E, Rand JI, Stevinson C. Complementary therapies for depression. Arch Gen Psychiatry 1998; 55: 102632. 10 Moyer CA, Rounds J, Hannum JW. A meta-analysis of massage therapy research. Psychol Bull 2004; 130: 318. 11 Jadad AR, Moore RA, Carroll D et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 112. 12 Verhagen AP, de Vet HCW, de Bie RA et al. The Delphi list: a criteria list for quality assessment of randomised clinical trials for conducting systematic reviews developed by Delphi Consensus. J Clin Epidemiol 1998; 51: 123541. 13 Hernandez-Reif M, Field T, Ironson G et al. Natural killer cells and lymphocytes increase in women with breast cancer following massage therapy. Int J Neurosci 2005; 115: 495510. 14 Hirakawa Y, Masuda Y, Kimata T et al. Effects of home massage rehabilitation therapy for the bed-ridden elderly: a pilot trial with a three-month follow-up. Clin Rehabil 2005; 19: 207. 15 Ahles TA, Tope DM, Pinkson B et al. Massage therapy for patients undergoing autologous bone marrow transplantation. J Pain Symptom Manage 1999; 18: 15763. 16 Alnigenis MNY, Bradley JD, Wallick J, Emsley CL. Massage therapy in the management of bromyalgia: a pilot study. J Musculoskelet Pain 2001; 9: 5567. 17 Brattberg G. Connective tissue massage in the treatment of bromyalgia. Eur J Pain 1999; 3: 23544.

18 Diego MA, Field T, Hernandez-Reif M et al. HIV adolescents show improved immune function following massage therapy. Int J Neurosci 2001; 106: 3545. 19 Diego MA, Field T, Hernandez-Reif M et al. Spinal cord patients benet from massage therapy. Int J Neurosci 2002; 112: 13342. 20 Field T, Diego M, Cullen C et al. Fibromyalgia pain and substance P decrease and sleep improves after massage therapy. J Clin Rheumatol 2002; 8: 726. 21 Field T, Hernandez-Reif M, Taylor S et al. Labor pain is reduced by massage therapy. J Psychosom Obstet Gynecol 1997; 18: 28691. 22 Field T, Sunshine W, Hernandez-Reif M et al. Massage therapy effects on depression and somatic symptoms in chronic fatigue syndrome. J Chronic Fatigue Syndr 1997; 3: 4351. 23 Field T, Hernandez-Reif M, Quintino O et al. Elder retired volunteers benet from giving massage therapy to infants. J Appl Gerontol 1998; 17: 22939. 24 Goodfellow LM. The effects of therapeutic back massage on psychophysiologic variables and immune function in spouses of patients with cancer. Nurs Res 2003; 52: 31828. 25 Hernandez-Reif M, Field T, Krasnegor J et al. Children with cystic brosis benet from massage therapy. J Pediatr Psychol 1999; 24: 17581. 26 Hernandez-Reif M, Field T, Krasnegor H et al. High blood pressure and associated symptoms were reduced by massage therapy. J Bodywork Mov Ther 2000; 4: 318. 27 Hernandez-Reif M, Martinez A, Field T et al. Premenstrual symptoms are relieved by massage therapy. J Psychosom Obstet Gynecol 2000; 21: 915. 28 Hernandez-Reif M, Ironson G, Field T et al. Breast cancer patients have improved immune and neuroendocrine functions following massage therapy. J Psychosom Res 2004; 57: 4552. 29 Hernandez-Reif M, Field T, Theakston H. Multiple sclerosis patients benet from massage therapy. J Bodywork Mov Ther 1998; 2: 16874. 30 Khilnani S, Field T, Hernandez-Reif M, Schanberg S. Massage therapy improves mood and behavior of students with attention-decit/hyperactivity disorder. Adolescence 2003; 38: 62338. gberg H. Tactile massage after a stroke 31 Olsson I, Rahm V-A, Ho and improved quality of life. Vard I Nord Nurs Sci Res Nord Countr 2004; 24: 216. 32 Post-White J, Kinney ME, Savik K et al. Therapeutic massage and healing touch improve symptoms in cancer. Integr Cancer Ther 2003; 2: 32344. 33 Stallibrass C, Sissons P, Chalmers C. Randomized controlled trial of the Alexander technique for idiopathic Parkinsons disease. Clin Rehabil 2002; 16: 695708. 34 Sunshine W, Field T, Quintino O et al. Fibromyalgia benets from massage therapy and transcutaneous electrical stimulation. J Clin Rheumatol 1996; 2: 1822. thlin C, Ko nig M. Efcacy of massage therapy in 35 Walach H, Gu chronic pain: a pragmatic randomized trial. J Altern Complement Med 2003; 9: 3746. 36 Walsh E, Wilson C. Complementary therapies in long-stay neurology in-patient settings. Nurs Stand 1999; 13: 325. 37 Williams A-L, Selwyn PA, Liberti L et al. A randomized controlled trial of meditation and massage effects on quality of life in people with late-stage disease: a pilot study. J Palliat Med 2005; 8: 93952. ller-Oerlinghausen B, Berg C, Scherer P et al. Effects of slow38 Mu stroke massage as complementary treatment of depressed hospitalized patients. Dtsch Med Wochenschr 2004; 129: 13638. 39 Jones NA, Field T. Massage and music therapies attenuate frontal EEG asymmetry in depressed adolescents. Adolescence 1999; 34: 52934. 40 Boylan M. Massage enhances positive mood and reduces stress for spouses of cancer patients. J Aust Tradit Med Soc 2005; 11: 812. 41 Maher J. A randomised controlled trial to evaluate the use of aromatherapy massage and relaxation therapies in improving quality of life in patients with advanced cancer. Natl Res Regist

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333

MT for the treatment of depression

333

42 43

44

45

2000. http://www.nrr.nhs.uk/ViewDocument.asp?IDN0143030200 (accessed 29 September 2006). Poland R, Daar E. Treatment of depression with massage in HIV. ClinicalTrials.gov 2004. Field T, Morrow C, Valdeon C et al. Massage reduced anxiety in child and adolescent psychiatric patients. J Am Acad Child Adolesc Psychiatry 1992; 31: 12531. Field T, Grizzle N, Scarfadi F, Schanberg S. Massage and relaxation therapies effects on depressed adolescent mothers. Adolescence 1996; 31: 90311. Field T, Diego MA, Hernandez-Reif M et al. Massage therapy effects on depressed pregnant women. J Psychosom Obstet Gynecol 2004; 25: 11522.

46 Manber R, Schnyer RN, Allen JJB et al. Acupuncture: a promising treatment for depression during pregnancy. J Affect Disord 2004; 83: 8995. 47 Smith CA, Hay PPJ. Acupuncture for depression. Cochrane Database Syst Rev 2004, Art. No. CD004046. DOI: 10.1002/ 14651858.CD004046.pub2. 48 Ernst E. The safety of massage therapy. Rheumatology 2003; 42: 11016.

Paper received June 2007, accepted August 2007

2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd Int J Clin Pract, February 2008, 62, 2, 325333