One of the most disabling chronic conditions for Americans is osteoarthritis. This slowly progressive degenerative disease is said to cause more physical limitation in adults than the diabetes and heart and lung diseases we see topping the health news these days (from Centers for Disease Control and Prevention).¹ Knee and hip joint replacements are now a common event on operating tables, while a plethora of ads tout the benefits of both conventional drugs and the less conventional glucosamine/chondroitin combo to stave off this end result.

The statistics on the incidence of osteoarthritis are staggering, with as many as twenty-one million Americans affected. It is esti-mated this number will climb to more than fifty million by the year 2020.² In a 1999 survey, the American Academy of Orthopedic Surgeons (AAOS) reported ten million American adults diagnosed with osteoarthritis of the knee. In this same report, it was noted that one-quarter of adults with osteoarthritis of the knee underwent some type of surgery as compared to less than one-tenth of adults without the condition. The study also revealed that women predominate in these statistics, accounting for two-thirds to three-fourths of those afflicted. And the incidence rises with age, showing up in 12 percent of American adults aged sixty-five years and older.³

Taking these statistics to heart, a team of researchers in New Jersey initiated the first study of its kind to evaluate the use of massage therapy to decrease symptoms and improve physical function in patients diagnosed with osteoarthritis of the knee. Lead researcher Adam I. Perlman, MD, of the Institute for Complementary and Alternative Medicine, University of Medicine and Den-tistry of New Jersey, collaborated with a team from Yale University’s medical school, most of who are also physicians. The study was backed by funding from the Centers for Disease Control and Prevention (CDC) and Yale’s Prevention Research Center grant.

At the outset, the team based their expectation of improvement on previous research of massage for other painful muscu-loskeletal conditions, including low-back pain and rheumatic diseases. Citing positive results from a study on spinal cord injury (Nayak, 2001), the team report states, “Massage therapy may diminish symptoms and improve the course of osteoarthritis by in-creasing local circulation to the affected joint, improving the tone of supportive musculature, enhancing joint flexibility, and relieving pain.”⁴

For this randomized, wait-list controlled trial, researchers recruited participants from area ambulatory care centers, private medical practices, and senior living facilities. Eligibility criteria included radiographical substantiation and physician diagnosis of osteoarthri-tis of the knee, as well as specific score ranges on pain measurements and an osteoarthritis questionnaire. Patients with a history of recent steroid treatments or knee surgery or diagnosed with other serious medical conditions were excluded. While the trial was open to both men and women, the majority in the final sample group of sixty-eight subjects were elderly females,⁵ reflecting the statistics from AAOS as noted above. Participants were randomly assigned to the two groups in equal number and demographics turned out to be similar between groups. Average age for the intervention group (twenty-seven females, seven males) was about seventy, and for the control group (twenty-six females, eight males) about sixty-six. In both groups, 85 percent of the subjects were Caucasian.⁶

Treatment for All

Using the wait-list control design, researchers were afforded the opportunity to include data from the control group’s subse-quent treatment as a second intervention group in addition to their initial role as a no-treatment comparison.⁷ Fortunately for the control subjects, this also allowed them to reap the benefits of massage for their painful condition. The massage intervention was administered over a period of eight weeks, with subjects receiving one-hour long sessions of full-body Swedish massage. The protocol included petrissage, effleurage, and tapotement techniques, applied as the therapist deemed appropriate. While the specific strokes were standardized to minimize practitioner variability, the sequence of strokes was left to the therapists’ discretion. The two participating therapists were licensed and certified by the National Certification Board for Therapeutic Mas-sage and Bodywork.⁸

During the initial four weeks of intervention, treatments were scheduled twice weekly to enhance a loading-dose effect. For the remainder of the trial period, subjects received one session weekly. While on the wait list, the control group continued their usual care for eight weeks, then crossed over to an intervention period as described for the original intervention group. Con-versely, the first intervention group had another eight-week period of usual care prior to the end-of-study evaluation. To mini-mize the attrition common to research studies, team members prompted participants to keep appointments on a regular schedule. In spite of this effort, a substantial percentage of subjects were lost to follow-up at the sixteen-week point.⁹

Measuring Success

Researchers found significant positive changes for the massage intervention, but before examining their results let’s take a look at the measurements used. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) assesses pain, stiffness, and physical functional disability in patients with hip and knee osteoarthritis through a self-administered questionnaire. Results are evaluated in the three major domains (pain, stiffness, functionality) as well as through a mean global score for all of the twenty-four questions of the index. In addition to the WOMAC, a visual analog scale (VAS) was used for self-report of pain levels. Measures also included data for time in seconds to walk a fifty-foot line and range of motion as assessed by a trained research assistant. For both groups, measurements were collected at baseline and at the eight-week and sixteen-week completion points after interven-tion. Although participants were instructed to maintain a daily diary of medication use, the team says compliance was question-able.¹⁰

Baseline scores were comparable between the two groups with the exception of a slightly higher mean WOMAC pain score in the intervention group. WOMAC global scores improved significantly for this group after eight weeks of treatment, with the greatest improvements seen in pain. In contrast, there was no change in the WOMAC domains for control subjects during their eight weeks of usual care. The team reports similar trends in both the VAS scores and clinical assessment of range of motion, with a high cor-relation between changes in the VAS and WOMAC global score.¹¹

Following the control group’s crossover to eight-week intervention, results of both groups were pooled to determine within-group intervention effects for the full sixteen-week trial period. Again, the team found significant improvement in WOMAC global and domain scores overall along with the same positive trend in VAS scores and range of motion. Bolstering these re-sults is the finding that for the original intervention group, the positive effects demonstrated in all measures at the eight-week completion point continued to persist at sixteen-week evaluation—a full eight weeks after cessation of treatment.¹²

A Promising Alternative

Although this study is not without limitations, the promise of the results seems to far outweigh the few potential flaws. Overall, the researchers state, “This study suggests that massage therapy using the Swedish technique is safe and effective for reducing pain and improving function in patients with symptomatic osteoarthritis of the knee.”¹³ With so many sufferers of this condition already in the hands of massage therapists, it’s reassuring to be on the road to scientific proof of efficacy. In noting the study’s flaws, the writers also bring up a number of positive observations related to their work that could influence success in future research in this area.

One potential weakness is the use of wait-list controls converting to intervention for the pooled experimental data, rather than hav-ing a placebo comparison group that would be experiencing the same amount of direct personal contact throughout the intervention period. However, they say, the intervention gains were significant for both groups and generally persisted for the original experimen-tal group up to eight weeks post-treatment. With little established precedent for selection of duration, frequency, or type of massage for this study, further work in this area may more clearly define these factors and lead to refinement of efficacy.¹⁴

Homogeneity of the sample group is another area of concern in that it may limit application of benefits to the general popula-tion.¹⁵ In light of AAOS statistics, however, the sample’s gender bias of predominantly women actually leans toward the reported higher incidence of osteoarthritis in women as compared to men above age fifty. The researchers also note that inaccuracy in sub-jects’ maintenance of their medication diaries left the team without a reliable means to examine potential influence of changes in medication. But, they say, “It seems unlikely that the massage intervention would have caused participants to increase their medi-cation in such a way as to lead to significant improvement in pain and function compared to the control group.”¹⁶

For those with osteoarthritis of the knee, these results offer a seemingly more pleasant alternative to treating disease symp-tomatology and progression than the current conventional approaches. Recently, there have been reports of potentially lethal side effects from several pharmaceuticals used to treat this condition. Surgery remains the last resort, but one that some may take as an inevitable path without hope for another route. Authors of the study also cite nonconventional treatments being used, whether proven or not. Although trials on glucosamine/chondroitin have had varying results, one recent study did sug-gest the combination to be effective for moderate to severe cases. “Other nutriceutical treatments, including devil’s claw and ginger, have yet to be proven effective,” they write. Studies utilizing acupuncture have been encouraging as well (Witt, 2005; Berman, 2004), but Perlman’s team says that the effects documented in the massage trial are even greater than those re-ported by Witt et al. in their larger acupuncture trial of similar design.¹⁷

Given the limitations and potential adverse effects of pharmacological and nonpharmacological treatments for osteoarthritis, massage therapy seems to be a viable option as an adjunct to more conventional treatment modalities,” the team writes. They even suggest its potential as an alternative to the drug approach. With the framework of this study now in place, further re-search is needed to more clearly define protocol, establish absolute efficacy, and expand generalization to other populations. For those diagnosed with this condition, it’s a promise worth keeping.¹⁸