Massage Therapy: An Old Intervention Comes of Age


Individual sessions of massage consistently bring immediate reductions in self-reported anxiety, physiologic arousal, and some somatic symptoms in cancer inpatients. A preliminary survey of 104 oncology inpatients suggested that repeated sessions of massage decreased muscle tension, negative mood, physical symptoms such as pain and nausea, and perceived isolation. The cumulative effects of systematic massage therapy, however, have not been rigorously studied in oncology inpatients. The authors have designed a study to investigate empirically the effectiveness of massage in alleviating symptom distress in patients undergoing autologous bone marrow transplantation, and to illuminate the need for systematic assessment of complementary, non-pharmacologic techniques aimed at enhancing patient care.

Denise M. Tope, Ph.D., Danette M. Hann, PHD, Briane Pinkson, LPN
Denise M. Tope is Assistant Professor of Psychiatry, Danette M. Hahn is Postdoctoral Fellow in Psycho-Oncology, Center for Psycho-Oncology Research, Department of Psychiatry, Dartmouth Medical School, and Briane Pinkson is Healing Arts Practitioner, Medical Hematology/Oncology Unit, Dartmouth-Hitchcock Medical Center, Lebanon, NH.

Reprinted with permission of the publisher
Copyright 1994, Cerenex(tm) Pharmaceuticals

Why Massage Therapy for Cancer Patients?

Hands-on massage has been receiving increasing attention as a potentially cost- effective nursing intervention amid the high technology of cancer nursing. Emphasis on the human touch is the basis of the new trend toward a "high-tech/high-touch" approach to nursing care.

Renewed recognition of the importance of direct human contact is especially relevant for oncology nursing. Cancer patients, particularly inpatients and those undergoing bone marrow transplantation, experience a significant amount of anxiety and emotional distress due to the severe side effects of chemotherapy agents and immunosuppression, which often are not adequately controlled by pharmacologic means. Efforts to maintain or improve patients' quality of life, especially during rigorous treatment, might include direct-touch interventions such as massage.

Massage as a Cancer Nursing Intervention

Massage is a systematic form of touch using specialized manipulations of the soft tissues of the body to promote comfort and healing. It is among the oldest of medical interventions, documented in writings for over 2000 years, [1] yet it had become used less often in conventional medicine as biotechnology and pharmacology improvements in patient care proliferated. In the last two decades, however, massage as a complementary medical practice has reappeared.[2,3]

Massage offers an adjunctive treatment to help control symptoms faced by many cancer patients. A primary reason why massage may be an appropriate intervention is that it directly affects symptoms that patients commonly report. Massage has been shown to affect such variables as muscle tension, autonomic arousal, pain, and sense of well-being. [4,5] Such psychological and somatic symptoms are often reported by cancer patients and are frequently not fully controlled by medications. Thus, symptoms may occur in the absence of pharmacologic treatment or remain at residual levels alter pharmacotherapy, and can be problematic for patients. They exacerbate the unpleasantness of an inherently stressful situation and significantly affect patients' quality of life.[6], Moreover, the psychological sequelae of rigorous cancer treatment can interfere with patients' performance of self-care behaviors such as mild exercise and appropriate food intake, which are assumed to assist in recovery from stressful cancer treatment.[7]

Second, medication to control mood and certain somatic symptoms is not always a feasible option for patients. For instance, the patient may be too ill to tolerate anti-emetics or anxiolytics, or such medications may be contraindicated for other reasons. If so, nonpharmacologic interventions such as massage, relaxation, and imagery may be helpful and could provide symptom relief without negative side effects.

A third reason why massage may be an appropriate intervention in the oncology setting is that the psychological distress reported by cancer patients is generally not sufficiently severe to require intensive psychological intervention. To expound on this point, we have chosen to tar- get cancer patients undergoing autologous bone marrow transplantation (ABMT). We will focus on ABMT because these patients' treatment-related symptoms are among the most severe and life-threatening of all cancer treatments, and, as we will discuss in the following paragraphs, these patients may be especially suitable for a palliative, non pharmacological intervention such as massage.

ABMT patients generally experience side effects from toxicity and immunosuppression associated with the procedure, such as high fevers, nausea and vomiting, painful skin rashes, and debilitating fatigue, along with other idiosyncratic symptoms. In addition, they remain in the hospital for an extended period of time (at least 3 weeks), and for a portion of that hospitalization are unable to leave the protected environment. The combination of unpleasant symptoms associated with high-dose chemotherapy and ABMT, restrictions associated with prolonged hospitalization, and the ever-present potential of mortality can readily elevate levels of anxiety and depression in these patients.

Three general findings have emerged in research addressing psychological distress in cancer inpatients, including those undergoing ABMT. First, it has been clearly documented that psychological distress parallels physical symptoms in these patients. [8-10] That is, psychological distress increases as the number and severity of physical symptoms increase. Second, the level of distress experienced by patients is moderate, on average, not severe, although the amount of distress varies greatly from patient to patient. Larson et al, [11] for example, found that psychological distress assessed at pre-hospitalization using a standard mood questionnaire varied greatly among their sample of 30 patients. This variability indicated that some patients were in a great deal of distress, whereas others exhibited only minimal negative psychological response in anticipation of the procedure. Third, the distress present at the start of hospitalization remains throughout the procedure. [11,12] In fact, Tope et al [12] assessed psychological distress in ABMT patients via standard depression and anxiety questionnaires done on hospital admission, at the mid-treatment point, and immediately prior to discharge. Results indicated that psychological distress at the mid-treatment point had increased from pre -treatment levels, but tended to remain in the moderate range.

< In sum, these findings support the conclusion by Ahles and Shedd [13] that in general, patients undergoing ABMT experience a moderate and perhaps "normal" level of distress relating to the procedure, but whereas most patients cope adequately with ABMT, a subset of patients may experience significant distress. That is, most cancer patients do not meet criteria for any psychiatric disorders. For them, a less intensive intervention, such as massage, may be sufficient and even more effective than formal psychological intervention.

Preliminary Survey of Massage Therapy in Cancer Inpatients

The first step in systematically addressing the question of effectiveness of massage therapy in cancer patients in general and ABMT patients in particular was to survey patients who were receiving massage by a nurse who specialized in its delivery. The survey had two purposes: 1) to document that these very ill patients could tolerate repeated sessions of massage, and 2) to explore the range of variables or outcomes in a cancer population that might be affected by massage.

The self-report survey was conducted over a 4-year period on a convenience sample of 104 oncology inpatients (31% male, 69% female) who had received massage therapy. Average age of the sample was 43.2 years, ranging from 18 to 74. Patients with prolonged hospitalizations constituted much of the sample: Average number of days in the hospital was 27.9, with a median of 23 days and a range of 3 to 90 days. Approximately 60% of the sample were ABMT patients. All patients had received two or more sessions of massage during their hospitalization.
Table 1.--Components of a Massage Therapy Session
  1. Patient lies supine in bed
  2. Massage therapist (MT) sits on a stool behind the patient's head
  3. MT applies almond oil to [her] hands for lubrication
  4. MT places [her] hands on patient's shoulders to begin massage
  5. Massage consists of combination of effleurage, petrissage, and acupressure
    • Effleurage: Smooth, long, rhythmic strokes up either side of the spine and out across the shoulders and/or up the neck and off the head
    • Petrissage: Gentle kneading of the shoulders, trapezius, neck, and scalp .
    • Acupressure: Manual pressure applied to the upper back, shoulders, neck, and face.
  6. MT closes the session by placing one hand under the patient's neck, and the other hand lightly on the chest


Massage therapy was delivered by a licensed practical nurse who had received extensive training in massage therapy. She performed massage as part of her nursing duties. One type of massage adopted by the nurse for working with cancer patients is outlined in Table 1. Massage therapy, for the purpose of this survey, was tailored to meet the needs of the individual patient. Sessions of massage were usually limited to 30 minutes, because of time constraints in the nurse's schedule and other demands on patients' time. Common sites of massage included the back, shoulders and neck, and feet. Whereas patients often associate the term "massage" with "back rub," back massage was often precluded because of patients' intravenous apparatus or symptoms such as nausea. The self-report questionnaire was administered prior to discharge at the termination of their' course of massage therapy. It assessed the degree to which patients' massage therapy program had affected their anxiety, tension, overall comfort level, and sense of isolation. The brief questionnaire provided an opportunity for patients to respond to open-ended questions on "the most beneficial aspects of massage therapy" and "negative aspects of massage therapy."
Table 2.--Effects of Massage Therapy in Cancer Patients: Excerpts From a Preliminary Survey

"I have used relaxation tapes, drugs, and hypnosis to help me during these intense times. However, I truly appreciate the long-lasting help (and no bad side effects)that massage affords."

"The massages helped me to relax and released tension in my body. For a bone marrow transplant patient, massage can be extra important because of the prolonged hospitalization and the problems that accompany the procedure."

"It makes a long hospital stay very much better. Allows better sleeping and relaxes you when you may be nervous about your disease and recovery."

"She was able to relax a couple of pressure points that medication was unable to relieve."

Symptom Management
"Unlike other hospital treatments of pain, this is very comforting. Pills are often hard to swallow and have side effects. Also good for general attitude."

"It relaxed me and alleviated pain--no drug was able to do that. It was the most positive aspect of my health care here."

Improved Mood
"A great source of physical and emotional comfort in relieving pain in body and mind."

"The most beneficial aspect of massage therapy was that l could close my eyes and totally relax, and for a short period of time, forget all the pain and discomfort of this treatment. Massage therapy should be part of a patient's routine and holistic health care."

"The massage enabled me to totally relax and feel much less anxiety, which l was unable to do before the massage."

Decreased Isolation
"Feeling of warmth, connectedness to another person. Awareness of my body. I am much more than a cancer patient."

"This was a wonderful service. It is important for cancer patients to physically connect."

"It was nice to feel someone touching you when you're isolated. It's a very untouchable time."

In their responses, 99% of the patients spontaneously mentioned "relaxation" or "release of muscle tension" as one of the most beneficial effects of massage therapy. In addition, 35% commented on improved mood or sense of well-being, 22% mentioned assistance in symptom management (control of pain, inflammation, nausea), and 15% made reference to a decreased sense of isolation as one of the most beneficial aspects of massage therapy. None of the 104 patients who responded to the survey acknowledged any negative effects of massage. In fact, several spontaneously pointed out that massage had no negative side effects. Selected excerpts from survey questionnaires are presented in Table 2.

In sum, these findings indicate that oncology inpatients will participate in repeated massage therapy. They also suggest, for the purpose of research, that patients would be willing to complete serf-report questionnaires about that experience. Perhaps most importantly, the results of the survey indicate that muscle tension, anxiety, and general mood are the variables most noticeably affected by massage therapy. The Findings of our survey were then compared with the existing empirical literature about the effects of massage in cancer patients, described briefly below.

Empirical Findings on Massage Therapy in Cancer Populations

Research findings support the use of massage to alleviate symptoms in cancer patients. Massage has been demonstrated to affect symptoms in three general areas: psychophysiologic arousal, anxiety and self-reported mood states, and certain somatic complaints. For instance, in a recent study in patients with advanced cancer, massage was effective in reducing heart rate and blood pressure, and increasing skin temperature--all consistent with a relaxation effect.[14] These findings have been paralleled in other investigations. [15] Massage intervention has also been shown to be helpful in relieving anxiety, as well as other negative mood symptoms. [15,16] Finally, it has been shown that an individual session of massage can be effective in reducing serf-reported distress, nausea, insomnia, fatigue, lack of concentration, and bowel problems, [16] as well as pain. [17] Although these latter effects are not consistently statistically significant compared with a no-massage control group, they do suggest that massage may be helpful in relieving multiple symptoms and is worthy of further study. Empirical studies of massage have produced promising results; yet the existing research has methodologic shortcomings that limit the conclusions. For instance, each study described above assessed only the immediate effects of a single session of massage; none systematically assessed any cumulative effects of multiple, consecutive massage sessions. In cancer patients enduring longer cancer treatments and prolonged hospitalization, massage is likely to be repeated. Moreover, massage delivered in the commercial or fee-for-service realm is generally administered on a regular basis. To our knowledge, the effects of a longer-term massage therapy program have never been assessed in a medical population, and certainly not in an oncology setting. Second, in most of the studies described above, massage was administered by nurses who had received only several hours of training in a particular technique. The findings of these pioneer studies might have been even stronger if the intervention had been delivered by a nurse more extensively trained and experienced in massage. Finally, of the existing studies, many did not incorporate a control condition, and most involved very small samples. Each of these methodologic shortcomings limits the conclusions that can be drawn from these studies.

Perhaps the most rigorous study to date on the efficacy of massage therapy as an adjunct to medical and pharmacologic treatment--although it did not target cancer patients or even physically ill subjects--involved the use of massage in child and adolescent psychiatric patients. [18] Adolescents in a residential treatment program were randomly assigned to either a treatment group, consisting of 30 minutes of massage therapy daily for 5 days, or a control group designed to control for therapist attention and reduced activity level, in which participants viewed relaxation tapes with a therapist. Outcome was measured through self-reported mood and behavioral observations of the participating adolescents. Overall, the findings suggested that massage therapy can reduce anxiety and depression immediately following a session of massage, even when compared with a control group. Even more interestingly, this study illustrated that massage therapy may produce a cumulative, global effect, in that reductions in anxiety and depression occurred between the first and fifth day of intervention. Moreover, the effects were not limited to variables directly affected by massage, but had generalized to other behaviors such as interactions with peers. The generalization and strength of these longer-term effects of massage were not anticipated by the authors, but they indicate the potential for improving psychosocial outcome from a comprehensive massage therapy program.

In summary, massage therapy in patients in general and cancer patients in particular has potential as a non-pharmacologic intervention for alleviating emotional distress, controlling psychophysiologic arousal, and possibly affecting physical symptoms. Moreover, our survey results suggested that patients report good response to multiple massage sessions. To date, however, no massage therapy program for physically ill individuals has been studied with acceptable rigor. Given the methodologic issues described in previous paragraphs, we developed a more systematic, controlled study of the short and long-term effects of massage in cancer patients undergoing autologous bone marrow transplantation.

A Randomized, Controlled Study of Massage Therapy in Cancer Inpatients

This study was designed to measure

  1. the impact of individual sessions of massage on self-reported anxiety, physiologic arousal, and symptom reporting;
  2. any quantitative effects of massage over a 3-week period on anxiety and physiologic arousal; and
  3. any global effects of a massage therapy program on patient mood (anxiety and depression) approximately midway through treatment as well as prior to discharge from the hospital.

To meet these objectives, 30 patients scheduled for admission for ABMT will be randomly assigned to one of two groups:

  1. massage therapy or
  2. a standard-treatment (no-massage) control.

Patients in the massage therapy group receive three 20-minute sessions per week for 3 weeks, in addition to their usual treatment. The structure of the massage has been standardized and incorporates some basic components of massage therapy. Patients in the standard-treatment control group receive treatment as usual. The immediate effects of massage will be measured via self-reported situational anxiety, ratings of physical symptoms, and indices of psychophysiologic arousal before and after individual sessions of massage. The global effects of regular massage therapy will be investigated via general anxiety, depression, and general emotional distress pre-hospitalization, mid-treatment, and pre-discharge. Data collection is in progress.

Conclusions and Suggestions for Future Research

Massage has been used to alleviate tension and other symptoms for thousands of years, with its popularity waxing and waning in the 20th century. Nevertheless, its benefits have essentially been presumed, with little or no scientific foundation. In fact, there has been little systematic investigation into the actual effects of massage and the mechanisms through which these effects occur. Its usefulness in enhancing patients' quality of life, alleviating symptoms, and improving mood makes intuitive sense. However, this and other non-pharmalogic interventions must be investigated empirically before formal claims can be made about their beneficial effects. As the focus of caring for patients with cancer continues to move from length of survival to the quality of survival, there is increasing need for systematic clinical investigations into methods of alleviating some of the distress that these patients encounter, and thereby improving the overall quality of patient care.

ACKNOWLEDGMENT -- Preparation of this article was supported in part by an exploratory research grant (1 R21 RR09557-01) from the Office of Alternative Medicine, National Institutes of Health, and a Special Institutional Grant (SIG 19) from the American Cancer Society.


1. Tappan F. Healing Massage Techniques: A Study of Eastern and Western Methods. Heston, Va: Reston Publishing; 1980.

2. Barnett KA. A theoretical construct of the concepts of touch as they relate to nursing. Res. 1972:21:102-109.

3. McCorkle R. Effects of touch of seriously ill patients. Nurs Res. 1974;23:125-132.

4. Longworth J. Pyschophysiological effects of slow stroke back massage in normotensive females. Adv Nurs. Sci. 1982;7:44-61.

5. Michelesen D. Giving a great backrub. Am J. Nurs 1978:78:1197-1199.

6. Aaronson NK, Meyerowitz BE, Bard M, et al. Wording group report: quality of life research in oncology: past achievements and future priorities. Cancer. 1991 ;67:839-843.

7. Massie M J, Shakin E.J.. Management of depression and anxiety in cancer patients. In: Breitbart W, Holland JC. eds. Psychiatric Aspects of Symptom Management in Cancer Patients. Washington, DC: American Psychiatric Press; 1993.

8. Andrykowski MA, Altmaier, EM, Barnett RL,, et al. Cognitive dysfunction in adult survivors of allogenic marrow transplantation: relationship to dose of total body irradiation. Bone Marrow Transplant. 1990:6:269-276.

9. Andrykowski MA, Henslee PJ, Farrall MG. Physicial and psychosocial functioning of adult survivors of allogeneic bone marrow transplanation. Bone Marrow Transplant. 1989;4:75-81.

10. Wolcott DL, Wellisch DK, Fawzy Fl et. al. Adaptation of adult bone marrow transplant recipient long-term survivors. Transplantation. 1986:41:478-484.

11. Larson PJ, Viele CS, Coleman S, et al. Comparison of perceived symptoms of patients undergoing bone marrow transplant and the nurses caring for them. Oncol Nurs Forum 1993;20:81

12. Tope DM, Ahles TA, Goulden I,C, et al. A prospective study of the neuropsychological and psychological impact of autologous bone marrow transplant in cancer patients. Poster presented at the annual meeting of the Association of the Advancement of Behavior Therapy; Atlanta, Ga: 1993.

13. Ahles TA, Shedd P. Psychosocial impact of bone marrow transplantation in adult patients: pre-hospitalization and hospitalization phases. In: Whedon M, ed. Bone Marrow Transplantation. Boston, Mass: Jones and Bartlett Publishers; 1991 ;280-292.

14. Meek SS. Effects of slow stroke back massage on relaxation in hospice clients. IMAGE: J Nurs Scholarship. 1993;25:17-21.

15. Ferrell-Torry AT, Glick OJ. The use of therapeutic massage as a nursing intervention to modify anxiety and the perception of cancer pain. Cancer Nurs. 1993;16:93-101.

16. Sims S. Slow stroke back massage for cancer patients. Nurs Times. 1986;82:47-50.

17. Weinrich SP, Weinrich MC. Effect of massage on pain in cancer patients. Appl Nurs Res. 1990;3:140-145.

18. Field T, Morrow C, Valdeon C, et al. Massage reduces anxiety in child and adolescent psychiatric patients. J Am Acad Child Adolescent Psychiatr. 1992;31:125-131.