Article Text

other Versions

Pilot study of the impact of massage therapy on sources and levels of distress in brain tumour patients
  1. Stephen Thomas Keir1 and
  2. Julia R Saling2
  1. 1Surgery Department, Duke University, The Preston Robert Tisch Brain Tumor Center, Durham, North Carolina, USA
  2. 2West Liberty University, Tridelphia, West Virginia, USA
  1. Correspondence to Professor Stephen Thomas Keir, Surgery Department, Duke University, The Preston Robert Tisch Brain Tumor Center, Box 3624 DUMC, Durham, NC 27710, USA; Keir0001{at}


Background Patients with brain tumours report elevated levels of distress across the disease course. Massage therapy is a commonly used complementary therapy and is employed in cancer care to reduce psychological stress and to improve quality of life (QoL). The purpose of this pilot study was to obtain a preliminary assessment of the effect of massage therapy on patient-reported psychological outcomes and QoL.

Materials and methods This study was a prospective, single-arm intervention. Participants were newly diagnosed primary brain tumour patients who reported experiencing distress and who received a total of eight massages over a period of 4 weeks. Participants completed the National Comprehensive Cancer Network's Distress Thermometer (DT) six times over a 5-week period.

Results As a group, levels of distress dropped significantly between baseline and week 3 (mean 4.19, SD 1.481, p≤0.025), with a further significant reduction in distress between week 3 and week 4 (p≤0.001). At the end of week 4, the DT scores of all participants were below the threshold for being considered distressed. By the end of the intervention, participants reported significant improvements in one test domain focused on emotional well-being.

Conclusions This study further documents that brain tumour patients report high levels of distress across the disease course. However, participants in this study reported improvements in distress level and total number of sources of distress while receiving massage therapy.

Statistics from


The clinical course of brain tumours leads not only to progressive functional deterioration but also to changes in the psychological impact on the patient.1 Those diagnosed with brain tumours frequently experience distress due to severe functional, neurocognitive and neuropsychological sequelae as a result of the progressive nature of the disease and subsequent treatment. Previous research indicates that such pathological conditions can impair a person's overall quality of life (QoL).2–5 Many of these symptoms are also found in other types of advanced cancer and may have a synergistic effect on patient morbidity.6 As a result, cancer patients have nearly twice the risk of developing psychiatric distress as compared to the general population.7 Furthermore, brain tumour patients and their spouses report having a high prevalence of comorbid mental disorders and psychosocial distress.8 In addition, this elevated distress level is apparent throughout the course of the disease.9 Despite these findings, very few studies have sought to assess QoL in terms of distress in brain tumour patients.1 One assessment tool that quickly and accurately determines an individual's distress level is the Distress Thermometer (DT), a single-item measure that asks respondents to rate their distress on an 11-point scale ranging from feeling no distress to feeling extreme distress.10 It is also possible that patients with advanced illness or significant symptoms of distress often associated with brain tumours are systematically undersampled for QoL assessment.11

Conventional treatment (surgery, radiation, chemotherapy) of adult brain tumours does not necessarily relieve symptoms of distress, and may even contribute to the distress due to its side effects.12 This, coupled with the many adverse effects for both patients and family caregivers throughout the course of the disease, highlights the need for appropriate interventions.13 Research suggests that specific interventions for symptoms together with standard of care may improve symptom impact on QoL.14 One such intervention is the use of massage, which can help relieve symptoms and improve physical and mental well-being as a part of overall QoL.15 When used in combination with standard of care, complementary therapies such as massage, can significantly improve QoL in patients with late-stage disease.16 Nevertheless, some risks are associated with massage which, although rare, are increased in those with cancer.17 However, research indicates that massage can be safely integrated into standard of care practices, especially when there is effective collaboration between the patient, a licensed massage therapist and the healthcare provider.17 ,18 Good communication is likely to maximise the benefits of massage therapy,17 which include symptomatic relief of pain, fatigue, stress/anxiety, nausea and depression.19 Massage therapy has also been linked to increased growth and enhanced immune function, as well as improved medical and psychiatric conditions.20–22 Tactile/kinesthetic stimulation is also considered a very cost-effective therapy which provides a broad range of benefits, including improved behavioural organisation.21

It is well known that brain tumour patients experience distress throughout the course of their disease, and that massage therapy can be safely administered to other cancer populations with advanced disease.9 ,16 Indeed, previous research indicates that massage therapy can relieve distress and its symptoms in other populations14–16; however, brain tumour patients are a unique subset of the cancer population due to the physical and neurocognitive issues associated with the disease and its treatment.3–5 ,13 ,15 As previous research indicates that brain tumour patients are interested in participating in interventions that use massage therapy to address their unique needs,23 and as there is a lack of research on effects and potential risk, assessment of the impact of massage therapy on QoL in brain tumour patients is required.1 ,16

Materials and methods


Subjects were a sample of brain tumour patients being followed at an academic medical centre between November 2008 and September 2009. The inclusion criteria for study participants were as follows: (a) ≥18 years of age, (b) a diagnosis of a histologically confirmed, newly diagnosed, post-surgery, WHO grade III/IV malignant glioma (ie, glioblastoma, anaplastic astrocytoma), (c) a Karnofsky performance score of ≥70 at study onset, (d) an estimated life expectancy of ≥3 months, (e) approval from the attending oncologist, (f) the ability to speak, read and write English, (g) living within a 60-mile radius of the academic medical centre where they were receiving care, (h) having been identified as ‘distressed’ by the National Comprehensive Cancer Network's (NCCN's) DT, and (i) signed, institutional review board approved informed consent prior to initiation of study-related procedures.


The massage intervention took place at The Center for Living at the academic medical centre. The Center for Living is a local and national resource providing both medical and wellness information through a variety of programmes that improve client lifestyle. The centre offers a wide range of therapeutic massage services in a quiet and private setting. The massage therapists at the centre work closely with physicians to design appropriate interventions for patients who are receiving treatment at the local academic medical centre, including cancer patients.

Design and procedure

In this prospective, single-arm intervention study, potential participants were identified and screened for eligibility via medical record review of patients scheduled for a consultation for post-surgical primary adjuvant treatment. Immediately following their consultation and approval of the primary attending oncologist, potential participants were asked if they were interested in participating in a study involving massage therapy. Interested patients who gave their consent were screened to determine their level of distress using the NCCN's DT. Those patients determined to be ‘distressed’ were eligible for participation and received a study outline, and their study responsibilities were explained to them by the study coordinator. As a part of the study, participants were to receive two massage therapy sessions each week for 4 weeks (for a total of eight sessions). Study participants were required to complete study questionnaires at baseline, after weeks 1, 2, 3 and 4, and 1 week after the last massage session was completed (week 5).

Massage intervention

Massage therapy sessions made use of ‘classic’ Swedish massage techniques that included long strokes, kneading, friction, tapping, percussion, vibration, effleurage and shaking motions. Each massage therapy session lasted 45 min. All massage therapy sessions were conducted by two highly trained massage therapists, each having completed more than 600 h of training and holding a state license to practice massage.

Distress assessment

The NCCN's DT is a 37-item instrument designed to rapidly assess the level and sources of emotional distress. The DT has two sections, a visual analogue scale and a 36-item checklist. The DT visual analogue scale allows patients to rate their level of distress from 0 (none) to 10 (extreme). The midpoint of 5 indicates moderate distress. The 36-item DT checklist allows patients to identify the sources/causes of their distress and is divided into six subcategories of problems: physical, family, emotional, spiritual/religious, informational and practical. Distress in a cross-sectional population of brain tumour patients using the DT has been previously recorded by this group as having a mean score of 4.3

Sample size

As this study was an exploratory pilot study, it did not require formal power calculation. We recruited 25 patients based on criteria provided by massage therapists and subject to funding constraints. In addition, we worked closely with the Duke Cancer Center statistical team to determine which commonly accepted scales could be adequately powered on these numbers and time period.

Statistical analysis

A general linear model with repeated-measures analysis of variance was used to test for statistical significance. All data were initially entered in an Excel spreadsheet and then analysed using SPSS V.14.0. In addition, raw scores from the DT were converted into a standard format using z scores based on the population mean and SD. An α type 1 error rate of 0.05 was prespecified for all tests of statistical significance and 0.080 for statistical trend. χ2 Analysis was conducted to examine differences in sources of distress on a weekly basis. Additional regression analysis was performed to determine significant indicators of sources of distress for this patient population.


A total of 28 patients were approached to participate in this study. One patient declined participation due to problems involving transportation. Of the 27 remaining patients, 26 met the eligibility criteria for being ‘distressed’ according to the DT. One patient dropped out of the study because of an unrelated illness (herpes zoster (shingles)). The remaining study participants (25 in total) were able to adhere to the twice-weekly massage therapy schedule for 4 consecutive weeks. Their responses to the DT are presented below.


The majority of participants were Caucasian (N=24, 96%) and male (N=16, 64%). The median age of the sample was 49, ranging from 24 to 70. All participants had a diagnosis of a histologically confirmed, newly diagnosed, post-surgery, WHO grade III/IV malignant glioma (ie, glioblastoma, anaplastic astrocytoma) at enrolment in the study. Scores from the DT were not significantly associated with age but did differ significantly by gender such that females reported significantly higher distress than males (t=2.70, p≤0.010).


As a group, participant-reported levels of distress had dropped by the end of the massage intervention at week 4. Levels of distress dropped significantly from baseline to the end of week 3 (mean difference 2.62, p≤0.025), week 4 (mean difference 4.23, p≤0.001) and 1 week after completion of the intervention (week 5) (mean difference 3.40, p≤0.001). In addition, participants reported a significant reduction in distress between weeks 3 and 4 (mean difference 1.61, p≤0.001). At the end of week 4, DT scores for all participants were below the threshold for experiencing distress. Distress scores did not significantly increase after completion of the massage intervention between weeks 4 and 5 (mean difference 0.88, p≤0.153). Mean scores and SDs of DT by week are presented in table 1.

Table 1

Mean Distress Thermometer scores and scores for items of concern

Items of concern

As a group, participant-reported items of concern dropped significantly in number between baseline and the end of week 3 (p≤0.001, mean 9.30, SD 4.177), week 4 (p≤0.001, mean 7.70, SD 3.846) and week 5 (p≤0.001, mean 8.60, SD 5.232). The total number of reported items of concern did not significantly increase at week 5 after completion of the massage intervention (p≤0.252). The total number of concerns reported by participants was directly correlated with the patient's level of distress. Regression analysis was used to determine significant predictors of distress, which included worry (β=0.337, p<0.05), sadness (β=0.317, p<0.05), nervousness (β=0.307, p<0.05) and pain (β=0.298, p<0.05). At the baseline assessment, 75% of participants reported the following items of concern: sadness, worry, fatigue, nervousness, pain, sleep and getting around, while 50% reported concerns with insurance, fears, depression, dry/itchy skin, work, transportation, eating, constipation, tingling in hands and feet, and nausea. At the end of week 3, 75% of participants reported only one item of concern (pain), and 50% of participants reported only two items of concern (fatigue and worry). At completion of the intervention at week 4, only 50% of participants reported concerns with fatigue. At week 5 after completion of the massage intervention, 50% of participants reported an increase in the following concerns: worry, sadness, nervousness, pain, insurance and fears. Mean scores, SD and total DT items of concern are presented in tables 1 and 2.

Table 2

Distress Thermometer items of concern by percent*


This study demonstrates a high level of distress in brain tumour patients, as evidenced by all but one participant meeting the criteria for being deemed ‘distressed’ on the DT. Distress scores decreased significantly from baseline throughout the study, suggesting an additive or cumulative effect for massage therapy on distress level reduction. This is highly supported by the fact that, at week 4, no patients were considered distressed and that at week 5 distress levels were still lower than at baseline. Although there was no control group for comparison, previous studies by Keir indicate that distress continues to negatively affect brain tumour patients throughout the course of their disease.3 ,9 However, this study suggests that the massage intervention reduced distress overall for study participants.

The number of distress-related concerns reported by individuals directly correlated with their level of distress. The number of concerns and number of participants affected decreased throughout treatment from baseline to week 4. At baseline, 75% of respondents listed seven items of concern, while 50% listed an additional 10 items. However, at week 4 half the participants listed only one concern (fatigue). At week 5 when massage therapy had ended, the majority of concerns reported at baseline returned but to a lesser extent. This result emphasises the importance of massage therapy for reducing distress levels in cancer patients.

The effect of massage on general well-being has been well documented in previous research. This study validates those findings in a brain tumour population, as participants in this study reported a reduction in factors contributing to distress, leading to improved emotional, social and physical well-being. However, when massage was discontinued, the overall reduction in factors contributing to distress became insignificant. Nevertheless, concerns were still below baseline 1 week after massage had ended.

The study has several limitations, including the small sample size and lack of controls for comparison. This is due in part to funding restrictions and because participant recruitment was restricted to a 60-mile radius around the treatment facility, which may have led to selection bias and restricts generalisation of results to other brain tumour patients. Therefore, the reduction in distress and improvements in QoL reported in this study may be unique to this set of study participants.

Recommendations for future studies include using a crossover methodology or a control group administered a different complementary intervention which may also lower distress levels and is safe for brain tumour patients. As this study was restricted to measured outcomes over 5 weeks, future research should examine therapy protocols over a longer period and continue to track outcomes post-intervention. A greater number of participants should be recruited to offset the possibility of sampling error. In addition, objective assessment using physiological and biological markers should be incorporated to improve study design.


This pilot study supports the fact that receiving a brain tumour diagnosis and then living with the disease causes distress. Participants in this study reported improvements in their overall QoL in terms of distress and sources/items of concern while receiving massage therapy as a part of a twice weekly intervention programme. This pilot study demonstrates that massage therapy improves QoL in this unique subpopulation of cancer patients.


We thank the Massage Therapy Foundation for support.

  • Received 22 February 2012.
  • Accepted 22 August 2012.
View Abstract


  • Contributors STK is responsible for the overall content of this manuscript. STK contributed to the planning, conduct and reporting of the work described in this article, and secured funding for the project. JRS primarily contributed to the reporting and writing of the manuscript.

  • Funding The Massage Therapy Foundation financially supported this study.

  • Competing interests None.

  • Ethics approval The Duke Cancer Center Institutional Review Board approved this study.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.