Learning From the Largest Study on Cancer and Massage

By Tracy Walton, LMT, MS
May 29, 2009

Learning From the Largest Study on Cancer and Massage

By Tracy Walton, LMT, MS
May 29, 2009

The body of research on cancer and massage is growing. One study often cited to support massage therapy programs for cancer patients was performed by the Memorial Sloan-Kettering Cancer Center (MSKCC) in New York City. Authored by Barrie Cassileth and Andrew Vickers, it's titled "Massage Therapy for Symptom Control: Outcome Study at a Major Cancer Center," and is the largest published study on cancer and massage to date. MSKCC is not new to the massage arena. Therapists have provided Swedish massage, light-touch massage and foot massage since 1999, and both inpatients and outpatients receive the work.

The "Big Five" Cancer Symptoms

Health care for cancer patients focuses on what some people call "The Big Five" symptoms patients face: pain, nausea, fatigue, anxiety and depression. Medications can help somewhat, but these five symptoms still can cause much suffering along the cancer journey. Massage therapists have offered anecdotal reports of symptom relief in their clients. If their experiences turn out to be true for significant numbers of people, this indeed will be news.

So far, only small studies have suggested a link between massage and symptom relief, and it's too early to claim "proof." Cassileth and Vickers strengthen the suggested link with this observational study of their clinical offerings, documenting their patients' responses to massage in a systematic way.

In this study, symptom cards were distributed to patients. These cards asked them to rate their symptoms on a 0-10 scale at baseline (pre-massage) and post-massage, five to 15 minutes afterward. Three years' worth of patients led to a large sample size.

Cards were returned for several thousand massage sessions, and the study staff pared them down to only the initial sessions for 1,290 different patients. Because of when the cards were completed, they supplied data only on immediate effects on symptoms, if any. To see about sustained effects on symptom relief, investigators followed up with approximately one-quarter of the patients by phone, 24 to 48 hours after their massage session. A large amount of data was collected.

Control Group or No Control Group?

It's important to note the absence of a control group in this study. This was not a "randomized, controlled clinical trial (RCT)." In an RCT, patients in the study are randomized to either an intervention (massage) group or a non-intervention (control) group, the intervention is applied (or not, in the case of the control), and the same measurements are taken from both groups for comparison. A control group is a key feature of a study because, if treatment X appears to be effective for symptom Y, it's extremely important to know whether symptom Y would have improved without treatment X. Symptoms tend to come and go, and symptoms improve for all sorts of reasons. Thus, a control/comparison group is vital if you want to isolate any effects that are specific to massage.

In class, I often am asked, "Why did this group carry out such a large study without bothering to include a control group? Isn't it a lot of wasted work?" This is an important question. For the goals of the study, a control group wasn't necessary. One goal was to see whether existing clinical services seemed to be helping people. Another was to check feasibility: whether massage therapy could indeed be delivered at high volume in a major cancer center. Even though the massage program had been up and running and was theoretically feasible, because it already was happening, numbers like this make feasibility real. This observational self-study was the perfect design for these particular goals.

A controlled clinical trial of this size would be very costly. However, such an observational study lays a foundation for one, paving the way for funding. The authors mentioned their plans for an RCT in the paper, and a look at the MSKCC Web site shows that one currently is underway on massage at the end of life. Moreover, the data from this observational study support not only the researchers themselves, but also the rest of us in seeking funding and support for RCTs on cancer and massage. So, their efforts were in no way wasted.

What Did They Find?

The researchers found what you might expect − immediate, dramatic reductions in all five symptoms. Notably, in patients who initially scored a given symptom at 4 or more, the average improvements in that symptom ranged from 42.9 percent in fatigue to 59.9 percent in anxiety. Patients who had Swedish and light-touch massage had stronger responses than those who received foot massage, but there was little difference in the outcomes between Swedish and light-touch massage.

Those were the immediate, post-massage effects. Follow-up scores looking for sustained effects were obtained from inpatients two to five hours after treatment and from outpatients 24 to 48 hours later. Improvement in outpatients' symptoms persisted over that time period. In contrast, inpatient scores, which initially had improved, started to worsen in just a few hours after massage treatment. This is an interesting difference!

Although it's tempting to focus only on massage benefits, other data about the massage protocols and other factors also were interesting. For example, investigators found that Swedish massage and foot massage were more commonly administered than light-touch massage, and that foot massage was used more often for inpatients than outpatients. The latter may reflect practical issues in massage with inpatients − being able to easily reach the feet of a patient surrounded by equipment, no need for repositioning, and so on. Swedish massage and light-touch massage were balanced between in- and outpatients. Moreover, the average length of the massage session for an inpatient was just 20 minutes, while the average session for outpatients was 60 minutes in length. This is a wide range in dose, an important clinical factor. In my experience, massage therapists are good for some lively conversation about the needed, tolerated and best massage dose for any given symptoms!

Take-Home Messages

These data provide rich opportunity for speculation. Why did the outpatients seem to do better than the inpatients? Is it a function of the difference in massage dose? Is it a function of the type of massage protocols or how ill the patients were in the first place? Is it harder to sustain the benefits of massage in a hospitalized patient in an acute health crisis than in an outpatient? These questions call for further study.

The investigators themselves stated, "Major, clinically relevant, immediate improvements in symptom scores were reported following massage therapy. Given the observational nature of this study, we cannot make conclusions about the cause of this effect." Their caution is well-advised. If you cite this study in support of massage therapy for this population, always mention it was an observational study, rather than a controlled trial that would establish clearer cause and effect. Use the word "suggest" rather than "prove." However, also note that this study offered clinical outcomes similar to smaller controlled trials in this population. See my summary of two such massage trials in the May 2006 and November 2006 issues of Massage Today.

Even without a control group, this study offers therapists, hospital administrators and health care providers a stronger foundation for massage. If you are building a case for a massage therapy program in your facility, note that MSKCC found it feasible for inpatients and outpatients at high volume. If your prospective client is nervous about receiving massage during cancer treatment or isn't sure it would help, a study like this suggests other people found it safe and helpful. This study gathers together 1,290 valuable, individual stories of massage into one place and offers them to us to scrutinize, learn from and appreciate. Studies such as this move the work forward. They inspire us by their example, move us to ask further questions and help us to envision a future when massage therapy is part of regular cancer care.


Author's Note: The article is indexed at www.pubmed.gov. Search the author to yield the abstract and ordering information, or request a reprint from the author in writing at MSKCC. Cassileth BA, Vickers, AJ. Massage therapy for symptom control: outcome study at a major cancer center. Journal of Pain and Symptom Management 2004;28(3):244-9. Memorial Sloan-Kettering Cancer Center, Integrative Medicine. "Our Research." Available at www.mskcc.org/mskcc/html/1990.cfm. Accessed 12-06.