6: No. 2, April 2009
Use of Complementary and Alternative Medicine Among Patients With Arthritis
Leigh F. Callahan, PhD, Elizabeth K. Wiley-Exley, MPH, Thelma J. Mielenz, PhD, Teresa J. Brady, PhD, Changfu Xiao, MS, Shannon S. Currey, PhD, Betsy L. Sleath, PhD, Philip D. Sloane, MD, MPH, Robert F. DeVellis, PhD, Joseph Sniezek, MD, MPH
Suggested citation for this article: Callahan LF, Wiley-Exley EK, Mielenz TJ, Brady TJ, Xiao C, Currey SS,
et al. Use of complementary and alternative medicine among patients with arthritis. Prev Chronic Dis 2009;6(2):A44.
Previous studies suggest that people with arthritis have high rates of using complementary and alternative medicine (CAM) approaches for managing their arthritis,
in addition to conventional treatments such as prescription medications. However, little is known about the use of CAM by diagnosis, or which forms of CAM are most frequently used by people with arthritis. This study was designed to provide detailed information about use of
CAM for symptoms
associated with arthritis in patients followed in primary care and specialty clinics in North Carolina.
Using a cross-sectional design, we drew our sample from primary care (n = 1,077)
and specialist (n = 1,063) physician offices. Summary statistics were used to
calculate differences within and between diagnostic groups, practice settings,
and other characteristics. Logistic regression models clustered at the site
level were used to determine the effect of patient characteristics on ever and
current use of 9 CAM categories and an overall category of “any
Most of the participants followed by specialists (90.5%) and a slightly smaller percentage of those in the primary care sample (82.8%) had tried at least 1 complementary therapy for arthritis symptoms. Participants with fibromyalgia used complementary therapies more often than those with rheumatoid arthritis, osteoarthritis, or chronic joint symptoms. More than 50% of patients in both samples used over-the-counter topical pain relievers, more than 25% used meditation or drew on
religious or spiritual beliefs, and more than 19% used a chiropractor. Women and participants with higher levels of education were more likely to report current use of alternative therapies.
Most arthritis patients in both primary care and specialty settings have used
CAM for their arthritis symptoms. Health care providers (especially
musculoskeletal specialists) should discuss these therapies with all arthritis patients.
Back to top
More than 1 in 5 US adults (46.4 million people) had doctor-diagnosed arthritis in 2003, and that number will grow to an estimated 67 million by 2030 (1). Arthritis is a common cause of disability in the United States, and the costs are substantial, estimated to be approximately 1.2% of the US gross domestic product (2).
Proper management of arthritis can reduce pain, functional limitations, and related problems (3). Treatment and management of arthritis can include medication, physical
or occupational therapy, patient education, weight loss, and surgery. Increasingly, complementary and alternative medicine (CAM) therapies are also being used. These therapies are a group of practices or products that are not currently used in the practice of conventional medicine. Estimates of CAM ever use among adults with
arthritis range from 59% (4) to 90% (5,6).
Many studies have looked at CAM use for arthritis (5-14), but we were particularly interested in the frequencies of use for multiple types of CAM by diagnostic category (especially within larger categories) and other characteristics, for which few data are available. Most studies provide only the averages for each category, which masks these differentiations. Herman
et al (5) found that 23.7% of people with arthritis in a sample from New Mexico used glucosamine, but only
1.2% used gamma linolenic acid. Katz and Lee (6) found that, although 42.4% of people with arthritis across the United States used mind-body interventions (such as prayer, spiritual healing, and biofeedback), only 3.7% used some form of relaxation therapy, guided imagery, or positive imagery. More information from populations from different parts of the country would provide an adjunct to these findings. We also explored the use of CAM by people with arthritis seeing
different types of health care providers.
Our main objective was to provide detailed information about ever and current use of methods of CAM for symptoms associated with osteoarthritis (OA), rheumatoid arthritis (RA), fibromyalgia (FM), and chronic joint symptoms (CJS) by demographic and disease status characteristics in a sample of 2,140 people in North Carolina. Our secondary objective was to separate and describe these findings by type of practice setting, primary care or specialty.
Back to top
Samples were drawn from 2 populations based on a study protocol approved by the University of North Carolina institutional review board: a family medicine research network and a musculoskeletal database.
Family Medicine Research Network
Data from the primary care setting were gathered via the North Carolina Family Medicine Research Network (NC-FM-RN), described in detail by Sloane
et al (15). During 2001, research assistants approached all adult patients in a representative sample of 16 family practice sites during a 4-week period. Each consenting adult patient was administered a 4-page self-report survey with questions on demographics, self-reported chronic conditions,
health habits (eg, smoking and physical activity), and self-rated health.
The racial/ethnic composition of the 5,575 patients who agreed to participate reflected that of the state’s adult population in terms of African Americans, Hispanics, and adults aged 65
or older. Patients who self-reported RA, OA, FM, or CJS were asked to complete the survey (n = 2,026).
The musculoskeletal database was established in the mid-1990s as part of an ongoing, longitudinal project measuring arthritis outcomes. During an outpatient visit, patients seen in the rheumatology or orthopedic clinics at the University of North Carolina Hospitals or 13 selected private rheumatology practices in North Carolina were asked to participate. Patients who agreed to participate completed a consent form and baseline self-report
questionnaire on demographic and health-related characteristics; diagnosis and date of disease onset were provided by the patient’s physician. Patients with RA, OA, or FM who completed this process and agreed to further contact were mailed the survey (n = 2,075).
Two survey booklets were mailed to 4,101 people. The first asked about health, health beliefs, and use of health care. The second asked about use of CAM. After 3 weeks, nonrespondents were sent a second set of survey booklets, and then
were contacted by telephone if neither mail survey elicited responses. A total of 2,140 patients responded to the survey (52.2%); 1,077 were from the NC-FM-RN, and 1,063 were from the musculoskeletal database.
Demographic characteristics included age, sex, race/ethnicity, education level, location of practice (urban or rural), and marital status. Because of the small number of responses in the categories other than African American or white, responses were categorized into white, African American, or other. Education was based on self-reported number of years, and marital status was dichotomized into currently married or not.
Disease information included self-reported RA, OA, FM, or CJS for the NC-FM-RN sample. The category of CJS was used if
patients reported having had symptoms of pain, aching, or stiffness in or around joints
during the last 30 days and did not self-report having RA, OA, or FM. For the specialist sample, the primary diagnosis (RA, OA, or FM) was provided by the specialist. Each participant in the 2 samples was then assigned a primary diagnosis of RA, OA, or FM (with CJS
also assigned in the NC-FM-RN dataset only). Consistent with previous research (5),
we classified participants who had more than 1 type of arthritis in the following order of priority: 1) RA, 2) FM, and 3) OA. In the NC-FM-RN dataset, 192 patients (18%) were classified with RA, 400 (37%) with OA, 81 (8%) with FM, and 404 (38%) with CJS; in the musculoskeletal database, 489 patients were classified with RA (46%), 300 (28%) with OA, and 274 (26%) with FM.
Health Assessment Questionnaire (HAQ)
The HAQ disability scale (16,17) is a reliable and valid instrument that rates difficulty with 20 activities of daily living ranging from 0 (without any difficulty) to 3 (unable to do). We calculated an unweighted mean of these scores.
Four questions focused on sleep (“Do you have trouble falling asleep?,” “Do you wake up several times per night?,” “Do you have trouble staying asleep?,” and “Do you wake up after your usual amount of sleep feeling tired and worn out?”) (18). The scores could range from 0 (no problems) to 5 (the most problems). We calculated an unweighted mean of these scores.
Pain and fatigue
Visual analog scales (VASs) were used to measure pain and fatigue (19). For example, the amount of pain experienced
during the past week was assessed by using a 100 mm VAS anchored with “no pain” (0
mm) and “pain as bad as it could be” (100 mm).
Rheumatology Attitudes Index (RAI)
The 5-item helplessness subscale of the RAI (20) was used to measure perceived helplessness (ie, the degree to which one believes the condition of interest is controlling one’s life). Five questions
were scored on a scale from 1 to 5, with 5 being the most helpless, and an unweighted mean of these scores was calculated.
Participants were asked about 9 categories of CAM use: alternative providers, special diets, vitamins and minerals, supplements, ointments or topical rubs, body treatments (eg, copper bracelets and magnets), movement (eg, yoga), spiritual (eg, prayer), and mind-body therapies (eg, visualization). In the regression models and when totals are reported for the category of vitamins and minerals, the following were excluded because they are often prescribed or strongly
suggested by physicians for people with musculoskeletal disorders: multivitamins, calcium, folic acid, and vitamin D. The specific percentage for each of these categories, however, is provided. A final (10th) category of “any use” was computed, which was coded as yes if the participants were using any of the 9 categories of CAM. Participants were asked whether they 1) have “ever used [therapy] for your arthritis or joint symptoms,” 2) “currently use [therapy] for your arthritis or
joint symptoms,” and 3) “plan to continue to use [therapy] for your arthritis or joint symptoms.”
Summary statistics were calculated; proportions are given for categorical
variables, and means with the standard deviation are given for continuous
variables. We used χ2 and linear regression with dummy variables to determine significant differences within and between diagnostic groups, practice settings, and other demographic characteristics. Logistic regression models clustered at the site level were used to determine the effect of patient characteristics on
current use of the 9 CAM categories and “any use” by using Stata software version 9.0 (StataCorp LP, College Station, Texas). Models were adjusted for age,
sex, race, education, marital status, HAQ score, RAI score, pain VAS, fatigue VAS, and location of practice.
Back to top
Higher proportions of participants were women and were white in both samples
(Table 1). Approximately half of the participants had more than a high school education. Almost half of patients in the primary care sample received care from rural practices, and all patients in the specialist sample received care from urban practices. The mean age in the specialist sample was slightly higher (59.8 years vs 54.0 years).
Types of CAM used
More than 80% of both samples had used some form of CAM for arthritis symptoms during the course of their disease (data not shown).
Ointments or topical rubs were the most commonly used CAM
(Table 2). More than 60% of both groups had ever used rubs. Spiritual methods were the second most commonly used CAM category;
approximately 40% to 49% of participants had ever used them. Alternative providers, vitamins
and minerals, other supplements, movement, and mind-body therapies were ever used by 22% to 40% of the groups. Special diets, on the
other hand, were the least commonly used (7% to 16% of both groups ever used
Although rubs were the most common ever-used CAM category, the rates of current use were much lower (approximately half). The same
was true for alternative providers and body treatments (eg, magnets). However, rates of ever and current use were similar for special diets, spiritual methods, and mind-body therapies.
Of the most commonly used specific types of CAM (Table
3), more than 50% of both samples used Bengay, Icy Hot, or similar ointments
or rubs; more than 25% used meditation or drew upon religious or spiritual beliefs; and more than
20% had seen a chiropractor or used calcium supplementation.
In the musculoskeletal database, 90.5% had used at least 1 CAM therapy for
their arthritis symptoms during their disease course, and 75.9%
still used at least 1 CAM therapy at the time of the survey (data not shown).
For the NC-FM-RN sample, a smaller percentage (82.8%) had ever tried at least 1 CAM therapy, and 70.2% were still using at least 1 CAM therapy at the time of the interview (data not shown). Methods used by 20% of patients in both settings included
chiropractors; calcium; Bengay, Icy Hot, and similar ointments or rubs; spiritual beliefs; and meditation.
Participants with FM used CAM therapies more often than did those with RA, OA, or CJS (Table 2). Of the specific categories of CAM use (Table 3)
that showed significant differences (P < .05) in use by disease category, patients with FM used most CAM therapies significantly more often than
those with other types of arthritis.
both sets of participants with OA, meditation was also commonly used (35.8% for
primary, 34.7% for specialty), as were drawing on spiritual beliefs and meditation for
participants with FM in the NC-FM-RN setting (55.6% for both CAM therapies).
Characteristics of current CAM users
In logistic regression models adjusted for age, sex, race, education, marital status, disability, pain, fatigue, and practice location, only sex was significantly associated with current use of any CAM in all 9 categories (data not shown). Most CAM therapy categories were significantly associated with at least 2 patient characteristics; for example, sex, race, and education were associated with the current use of supplements. However, sex was the only characteristic
significantly associated with current use of special diets.
Female sex was positively associated with most categories of CAM use, while higher levels of education were positively associated with 5 categories of CAM use and negatively associated with current use of
ointments or topical rubs. Of the other characteristics included in the adjusted analyses, the categories of African Americans, whites, and other race were positively associated with 3 categories of current CAM use: supplements,
ointments and topical rubs, and spiritual. Rural location of the practice was negatively associated with
current use of 2 categories: CAM providers and body treatments (eg, magnets). Disability, measured by the HAQ, was positively associated with spiritual and mind-body
therapy categories. Helplessness, measured by the RAI, was positively associated with body treatments.
Back to top
In this survey of 2,140 people with arthritis in North Carolina, most had used some form of CAM for their arthritis symptoms. This finding is close to other estimates (5,6) of 90.2% and 80% of ever use or use within the past month, although it is much higher than findings
of 34% to 68% from many earlier studies (7,8,12,14,21).
Some of the differences between our study and earlier studies that reported much lower levels of ever use of CAM may be attributable to our inclusion of prayer. In our study, 13.7% of the family practice group and 17.4% of the
specialty group prayed about their arthritis. Almost half (40.6%) of the sample with OA of the knee from Katz and Lee (6) used prayer. The numbers reported by Cronan
et al (22) also included prayer as a form of CAM, and their findings of ever use were
However, this inclusion does not seem to explain all of the difference, because Herman
et al (5) did not include prayer but still had similar findings. They attribute their higher percentage of use to differing definitions of CAM, noting that they surveyed for a broader array of mind-body therapies, energy therapies, and CAM movement therapies than most other studies. They also suggested that the differences between their study and earlier studies
were attributable to geographic location,
noting that CAM use is often higher in the Western regions of the United States, where their study took place.
A larger proportion of participants from the specialty setting had used CAM than had participants from the family practice setting. This
finding is not surprising because patients seeing specialists have more severe disease (23) and are probably in need of greater pain relief. Our findings corroborate a study by Breuer
et al (11) that noted significantly more CAM use by patients with FM and a study by Herman
et al (5) that reported a higher number of CAM therapies used by patients
with FM and RA than those with OA. The higher use of CAM therapies by participants with FM compared with participants who have other forms of arthritis is also not surprising. Few good pharmacologic treatments are available for FM, and people with FM are often encouraged to participate in exercise regimens and meditation, which could account for some of the higher levels of use (24-26). In addition, people with FM experience a wide variety of symptoms, such as nonrestorative sleep, mood
disturbance, irritable bowel syndrome, headache, and paraesthesias (25,27). These symptoms may catalyze the use of a broader range of therapies.
Participants in our survey tried a variety of therapies, and although many tried rubs,
alternative providers, and body treatments, they often were not currently using those methods. Ever and current use of special diets, spiritual methods, and mind-body therapies, on the other hand, were similar. This could suggest that people with arthritis are more satisfied with dietary, spiritual, and mind-body methods. More research in this area might explore what
it is about these methods that promotes continued use.
Several limitations should be noted when interpreting these results. Most prominently, the CAM questions in our survey asked whether respondents used CAM for arthritis or joint symptoms. Participants conceivably could have misread the question as asking whether they had ever used CAM for any reason. This issue has arisen in previous research (5), and validation of this aspect of the questionnaire is needed. Similarly, the self-reported nature of the diagnoses for
participants in the
family practice group is potentially problematic. Self-reported data for arthritis reportedly have moderate sensitivity (71%) and specificity (70%), but few studies address the issue (28).
This study also is limited in its ability to determine the use of CAM among races/ethnicities other than African American and white. Other studies have looked more closely at this issue (5,6). Although our study’s ethnic composition at enrollment paralleled that of the state’s adult population, oversampling of some races/ethnicities, such as Asians and Hispanics, would have enabled us to say more about these populations. In addition, these findings are based on a cross-sectional
survey. The findings from previous research show that people frequently change their patterns of CAM use (7). For this and other reasons, we have focused on both ever and current use in this
Because almost every participant in our study used CAM at some point for his
or her arthritis symptoms, it may be useful for practitioners to invite discussion of what therapies patients might be using for their symptoms and to assist them in evaluating risks.
Back to top
The NC-FM-RN is an organization dedicated to fostering practice-based research and is jointly sponsored by the Department of Family Medicine, the Thurston Arthritis Research Center, and the Cecil G. Sheps Center for Health Services Research at the University of North Carolina at Chapel Hill, in collaboration with the North Carolina Academy of Family Physicians. Participating family practices have included Biddle Point Health Center, Charlotte; Bladen
Medical Associates, Elizabethtown; Blair Family Medicine, Wallace; Chatham Primary Care, Siler City; Community Family Practice, Asheville; Dayspring Family Medicine, Eden; Goldsboro Family Physicians, Goldsboro; Henderson Family Health Center, Hendersonville; North Park Medical Center, Charlotte; Orange Family Medical Center, Hillsborough; Person Family Medical Center, Roxboro; Robbins Family Practice, Robbins; South Cabarrus Family Physicians, Harrisburg, Concord, Mt. Pleasant, and Kannapolis;
and Summerfield Family Practice, Summerfield.
We also thank the following physicians for encouraging their patients to participate in our musculoskeletal database and outcomes studies: H. Vann Austin, Franc Barada, Robert Berger, Mary Anne Dooley, William Gruhn, Robert Harrell, Tatiana Huguenin, Beth Jonas, Joanne Jordan, Fathima Kabir, Elliott Kopp, Andrew Laster, Kara Martin, Gwenesta Melton, Nicholas Patrone, Kate Queen, Westley Reeves, Hanno Richards, Alfredo Rivadeneira, William Rowe, Gordon Senter, Paul Sutej, Claudia Svara, Anne
Toohey, William Truslow, John Winfield, and William Yount.
This study was funded by the the Centers for Disease Control and Prevention, cooperative agreement no. U48/CCU409660.
We especially thank Jennifer Milan Polinski, MPH, and Carla J. Herman, MD, MPH, Division of Geriatrics, Department of Internal Medicine, University of New Mexico Health Sciences Center.
Back to top
Corresponding Author: Leigh F. Callahan, PhD, Thurston Arthritis Research Center, CB #7280, 3300, Thurston Building, University of North Carolina School of Medicine, Chapel Hill, NC 27599-7280. Telephone: 919-966-1939. E-mail:
Author Affiliations: Elizabeth K. Wiley-Exley, Thelma J. Mielenz, Changfu Xiao, Shannon S. Currey, Betsy L. Sleath,
Philip D. Sloane, Robert F. DeVellis, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; Teresa J. Brady, Joseph Sniezek, Centers for Disease Control and Prevention, Atlanta, Georgia.
Back to top
- Hootman J, Helmick C.
Projections of US prevalence of arthritis and associated activity limitations. Arthritis Rheum 2006;54:226-9.
- Centers for Disease Control and Prevention.
National and state medical expenditures and lost earnings attributable to arthritis and other rheumatic conditions — United States, 2003. MMWR Morb Mortal Wkly Rep 2007;56(1):4-7.
- Brady T, Kruger J, Helmick C, Callahan L, Boutaugh M.
Intervention programs for arthritis and other rheumatic diseases. Health Educ Behav 2003;30(1):44-63.
- Saydah S, Eberhardt M.
Use of complementary and alternative medicine among adults with chronic diseases: United States 2002. J Altern Complement Med 2006;12(8):805-12.
- Herman C, Allen P, Hunt W, Prasad A, Brady T. Use of complementary therapies among primary care clinic patients with arthritis. Prev Chronic Dis 2004;1(4).
- Katz P, Lee F.
differences in the use of complementary and alternative medicine in patients with arthritis. J Clin Rheumatol 2007;13(1):3-11.
- Rao J, Kroenke K, Mihaliak K, Grambow S, Weinberger M.
Rheumatology patients' use of complementary therapies: results from a one-year longitudinal study. Arthritis Rheum 2003;49(5):619-25.
- Rao J, Mihaliak K, Kroenke K, Bradley J, Tierney W, Weinberger M.
Use of complementary therapies for arthritis among patients of rheumatologists. Ann Intern Med 1999;131(6):409-16.
- Ahmed S, Anuntiyo J, Malemud C, Haqqi T.
Biological basis for the use of botanicals in osteoarthritis and rheumatoid arthritis: a review. Evid Based Complement Alternat Med 2005;2(3):301-8.
- Boisset M, Fitzcharles M.
Alternative medicine use by rheumatology patients in a universal health care setting. J Rheumatol 1994;21(1):148-52.
- Breuer G, Orbach H, Elkayam O, Berkun Y, Paran D, Mates M, et al.
Perceived efficacy among patients of various methods of complementary alternative medicine for rheumatologic diseases. Clin Exp Rheumatol 2005;23(5):693-6.
- Dente J, Herman C, Allen P, Hunt W. Ethnic differences in the use of complementary and alternative therapies among adults with osteoarthritis. Prev Chronic Dis 2006;3(3).
- Fautrel B, Adam V, St-Pierre Y, Joseph L, Clarke A, Penrod J.
Use of complementary and alternative therapies by patients self-reporting arthritis or rheumatism: results from a nationwide Canadian survey. J Rheumatol 2002;29(11):2435-41.
- Quandt S, Chen H, Grzywacz J, Bell R, Lang W, Arcury T.
Use of complementary and alternative medicine by persons with arthritis: results of the National Health Interview Survey. Arthritis Rheum 2005;53(5):748-55.
- Sloane P, Callahan L, Kahwati L, Mitchell C.
Development of a practice-based patient cohort for primary care research. Fam Med 2006;38(1):50-8.
- Fries J, Spitz P, Kraines R, Holman HR.
Measurement of patient outcome in arthritis. Arthritis Rheum 1980;23:137-45.
- Wolfe F.
Which HAQ is best? A comparison of the HAQ, MHAQ and RA-HAQ, a difficult 8 item HAQ (DHAQ), and a rescored 20 item HAQ (HAQ20): analyses in 2,491 rheumatoid arthritis patients following leflunomide initiation. J Rheumatol 2001;28(5):982-9.
- Jenkins C, Stanton B, Niemcryk S, Rose R.
A scale for the estimation of sleep problems in clinical research. J Clin Epidemiol 1988;41(4):313-21.
- Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Outcome measures for health education and other health care interventions. Thousand Oaks (CA): Sage Publications; 1996.
- DeVellis R, Callahan L.
A brief measure of helplessness in rheumatic disease: the helplessness subscale of the Rheumatology Attitudes Index. J Rheumatol 1993;20:866-9.
- Boisset M, Fitzcharles M.
Alternative medicine use by rheumatology patients in a universal health care setting. J Rheumatol 1994;21:148-52.
- Cronan T, Kaplan R, Kozin F.
Factors affecting unprescribed remedy use among people with self-reported arthritis. Arthritis Care Res 1993;6:149-55.
- Mazzuca S, Brandt K, Katz B, Dittus R, Freund D, Lubitz R, et al.
Comparison of general internists, family physicians, and rheumatologists managing patients with symptoms of osteoarthritis of the knee. Arthritis Care Res 1997;10(5):289-99.
- Jones KD, Adams D, Winters-Stone K, Burckhardt CS.
A comprehensive review of 46 exercise treatment studies in fibromyalgia (1988-2005). Health Qual Life Outcomes 2006;4:67.
- Mannerkorpi K, Iverson MD.
Physical exercise in fibromyalgia and related syndromes. Best Pract Res Clin Rheumatol 2003;17(4):629-47.
- Mannerkorpi K.
Exercise in fibromyalgia. Curr Opin Rheumatol 2005;17:190-4.
- Clauw D, Crofford L.
Chronic widespread pain and fibromyalgia: what we know and what we need to know. Best Pract Res Clin Rheumatol 2003;17:685-701.
- Bombard J, Powell K, Martin L, Helmick C, Wilson W.
Validity and reliability of self-reported arthritis, Georgia Senior Centers, 2000-2001. Am J Prev Med 2005;28(3):251-8.
Back to top