Factors associated with the use of complementary and alternative medicine in juvenile idiopathic arthritis




To describe the frequency of complementary and alternative medicine (CAM) use in patients with juvenile idiopathic arthritis (JIA) and to explore whether CAM was associated with patient-specific characteristics, parent-specific characteristics, and medical management factors.


Parents of children with JIA completed questionnaires that addressed the use of CAM, adherence and perceived problems, parental distress, and coping with childhood illness. Clinical variables were abstracted from the patients' medical files.


One hundred-eighteen children with JIA, mean ± SD age of 10.4 ± 3.9 years and mean ± SD disease duration of 4.5 ± 3.5 years, participated in the study. Ever use of CAM was 33.9% and was higher in patients whose parents used CAM (odds ratio [OR] 5.1, 95% confidence interval [95% CI] 1.4–19.5) and among those who considered themselves as “Canadian” as opposed to belonging to a specific ethnic group (OR 10. 7, 95% CI 1.2–99.8). Adherence to conventional treatment was high for both users and nonusers of CAM.


Use of CAM is common among patients with JIA. CAM use is not related to any decrease in adherence to conventional medical treatment.


Juvenile idiopathic arthritis (JIA) is a relatively common chronic medical condition of childhood (1). Its management requires a multidisciplinary treatment approach often involving a number of different therapies, including several types of antirheumatic drugs, physical therapy, and orthoses (2). Despite this, some parents of ill children believe that the side effects of medications outweigh their potential benefits and are reluctant to use them (3). Some may seek complementary and alternative medicine (CAM) because of their reluctance to administer what they perceive as “unnatural” medications to their children, fear of side effects with conventional medications, and a perception that the child's condition is not improving with such therapy (3, 4). Disease severity and progression may also influence whether parents explore other types of therapy (3, 4). Poor adherence to treatment regimens may compromise benefits of conventional treatments (5–7). A concern with CAM use is that it might be associated with decreased use of conventional therapies, either because of the parental belief system or because of drug interactions between conventional medications and CAM.

CAM therapies include many treatments or products that are currently not considered part of conventional medicine and are classified in 5 domains: alternative medical systems, mind–body interventions, biologically based therapies, manipulative and body-based methods, and energy therapies (8). CAM may include the following: acupuncture, homeopathy, herbal therapies, chiropractic, massage, high-dose vitamins, spiritual healing, lifestyle diet, relaxation, imagery, energy healing, folk remedies, and hypnosis (9).

Only 2 studies have reviewed the use of CAM in children with JIA, and neither attempted to measure its effect on adherence to conventional treatment (10, 11). Southwood et al (10) surveyed 53 patients with JIA attending arthritis youth camps and reported that 70% used unconventional remedies. Of these, the most common therapies cited were copper bracelets, dietary alterations, and patent medicines (e.g., seaweed extracts, cod liver oil, megavitamins, and herbal remedies). Hagen and associates (11) reported that 64% of their cohort of 141 patients with various pediatric rheumatic diseases used CAM, most commonly vitamin and mineral supplements.

The objectives of this study were to describe the frequency of CAM use in a cohort of patients with JIA, and to explore whether the use of CAM was associated with any patient-specific characteristics (age, sex, disease severity, disease duration), parent-specific characteristics (revenue, maternal education, ethnicity, parental use of CAM, psychological distress, coping with childhood illness), or medical management issues (adherence to conventional prescribed treatment, perceptions of problems with medical management). Based on the literature (3, 11–13), we hypothesized that CAM use would be associated with increased disease duration, higher parental income, parental use of CAM, and perceived problems with medical management.



Parents of patients with JIA who attended the JIA clinic at 2 pediatric hospitals (Montreal Children's Hospital–McGill University Health Centre and British Columbia's Children's Hospital) were eligible for the study. Parents were approached to participate if they had a sufficient comprehension of either English or French and if their children were currently undergoing treatment.

Data collection.

Parents of children with JIA completed several questionnaires: 1) a questionnaire package that included the Parent Adherence Report Questionnaire (PARQ) (14), which addresses adherence to treatment, a demographics questionnaire, and a questionnaire on the use of CAM and problems encountered with the health care system; 2) the Symptom Checklist-90-Revised (SCL-90-R) (15), which evaluates a caregiver's psychological status; and 3) the Coping Health Inventory for Parents (CHIP) (16, 17), which documents caregiver's coping with their child's illness. Clinical information, obtained from the patient's chart, included type of JIA, active joint count, date of diagnosis, and prescribed treatments.

The study was approved by the Institutional Review Boards of the Montreal Children's Hospital and British Columbia's Children's Hospital.


CAM use specifically for the child's arthritis was evaluated in terms of “ever use” and use in the past 3 months.

Psychological distress was measured with the SCL-90-R, a widely used and validated 90-item self-report measure of symptoms (9 symptom clusters) experienced during the past week (15). The clusters are summed to create a global severity index. Clinically important distress corresponds to T scores ≥63 (15). Test-retest reliability intraclass correlation (ICC) scores range from 0.47 to 0.69 and internal consistency Cronbach's alpha range from 0.72 to 0.91 (18).

Parental coping with child illness was measured with the CHIP (16, 17), a self-report measure of parents' perception of their response to management of family life when a child member is seriously or chronically ill. Coping behaviors are grouped into 3 subscales: 1) maintaining family integration, cooperation, and an optimistic definition of the situation; 2) maintaining social support, self esteem, and psychological stability; and 3) understanding the medical situation through communication with other parents and consultation with the medical staff. Higher scores are indicative of higher utilization of specific coping behaviors. The CHIP has good internal consistency with alphas >0.70 for the 3 subscales and fair concurrent validity as evidenced by significant correlations with independent measures of family environment and changes in child's health. Test-retest reliability is 0.75, 0.63, and 0.71, respectively, indicating sufficient sensitivity for changes in the situation (19).

Adherence to treatment was evaluated as caregiver's response on the PARQ to a 100-mm visual analog scale to the question of how often they followed treatment recommendations as prescribed by the health care provider. This was asked for each type of treatment (medications, exercises, and wearing of splints). Caregivers also rated difficulty in following treatments and perception of helpfulness for each type of treatment. The PARQ has been shown to have satisfactory construct validity and test-retest reliability (ICCs ranged between 0.60 for medications and 0.88 for exercises) (14). Finally, caregivers were asked about any problems that they encountered with respect to the medical management of their child. Problems were defined as organizational (e.g., long waits, difficulty in getting appointments, seeing different doctors each time, etc.), communication in nature (e.g., lack of understanding of clinician's explanations and instructions, etc.), and treatment-related (e.g., sideeffects, difficulty with the treatment, etc.).


Descriptive statistics were calculated to characterize use of CAM in general and within subgroups (patient-specific characteristics, parent-specific characteristics, and medical management characteristics). Logistic regression was employed to determine whether any of these characteristics were associated with the use of CAM. All analyses were performed using SPSS software (Chicago, IL).


There were 158 patients who were approached to participate (100 from Montreal and 58 from Vancouver). A total of 118 children was enrolled, 76 from Montreal and 42 from Vancouver. There were no differences between nonparticipants and participants with respect to age at disease onset or age at recruitment; however, there were slightly more girls who withdrew from the study than boys, and mean disease duration was slightly higher in nonparticipants (5.0 versus 4.6 years). Mean ± SD age of the children participating in the study was 10.4 ± 3.9 years (range 1.8–18.0 years); 66.7% were female. The percentage of children within each JIA classification type was 36.1% oligoarthritis, 23.1% polyarthritis, 5.6% systemic arthritis, 14.8% enthesitis-related arthritis, 15.7% psoriatic arthritis, and 4.6% other. Mean ± SD disease duration was 4.6 ± 3.5 years (range 0.2–15.6 years). The cohort is described in Table 1.

Table 1. Description of the cohort of JIA patients*
  • *

    JIA = juvenile idiopathic arthritis; GSI = global severity index.

Active joint count  
Family income  
Maternal education  
 High school or less3136.9
 Some college3035.7
 University degree2327.4
 French Canadian2731.8
 English Canadian3844.7
Parental psychological distress  
 GSI < 639688.9
 GSI ≥ 63 (high distress)1211.1

Parent-reported adherence was relatively high for medications (84.9 ± 27.8) and considerably lower for exercises (45.9 ± 32.0) and splints (42.0 ± 39.4). The mean ± SD psychological distress score among caregivers was 47.9 ± 11.6; well below the clinically significant level of 63. Average coping scores were 2.2 ± 0.5 on the family integration subscale, 1.7 ± 0.6 on the social support subscale, and 1.8 ± 0.7 on the health care subscale, indicating high use of coping strategies to maintain family integration, cooperation, and an optimistic definition of the situation. Of the 73 problems cited by 52 families who encountered problems, 34 were organizational, 9 were communication, and 30 were treatment-related.

“Ever use” of CAM for JIA was 46 of 118 (33.9%). Of the 14 (11.9%) of 118 who had used CAM in the previous 3 months, the types of CAM used were naturopathy (4), acupuncture (3), chiropractic (2), diet therapy (2), homeopathy (2), massage (1), spiritual healing (1), and other (3). Of those parents whose children used CAM, 92% sought pain relief for their child, and 93% indicated that the children continued with standard therapy even while using CAM. Almost half (43.7%) of parents reported also having used CAM for themselves. Among those who expressed having specific problems related to management of their child's arthritis, 52.0% used CAM; 55.6% among those who expressed communication problems, 36.4% among those with organizational problems, and 31% among those who cited treatment-related problems. Use of CAM was similar for both sites (32.9% in Montreal, 35.7% in Vancouver; P = 0.76). Factors associated with the use of CAM in univariate analyses (Table 2 and Figure 1) were parental use of CAM (P < 0.001) and ethnicity (P = 0.006), with the use of CAM being highest among families who identified themselves as being French Canadian. Those families that used CAM tended to have more difficulty following the medical treatment, although this was not statistically significant (P = 0.09). There was also a nonstatistically significant trend to higher CAM use in families with higher family income. Logistic regression analysis (Table 3) indicated that use of CAM was strongly associated with both parental use of CAM and being Canadian (as opposed to identifying with an ethnic group). There was a slight tendency for longer disease duration to be associated with CAM use (odds ratio 1.13, 95% confidence interval 1.00–1.43).

Table 2. Characteristics of users and nonusers of CAM*
 Used CAMDid not use CAMP
  • *

    Data reported as mean (standard error). CAM = complementary and alternative medicine; VAS = visual analog scale.

Active joint count1.81.20.33
Disease duration, years4.7 (0.5)4.4 (0.4)0.63
Parental psychological distress, global severity index47.6 (2.1)48.1 (1.4)0.83
Parental coping   
 Family integration subscale2.2 (0.1)2.2 (0.1)0.93
 Social support subscale1.6 (0.1)1.8 (0.1)0.35
 Health care subscale1.7 (0.1)1.8 (0.1)0.57
Adherence to treatment, 100-mm VAS   
 Medications86.0 (4.6)84.1 (3.7)0.74
 Exercises45.1 (6.0)45.5 (4.4)0.96
 Splints42.0 (18.6)42.0 (11.0)1.00
Difficulty following treatment, 100-mm VAS   
 Medications23.7 (5.1)14.4 (3.0)0.09
 Exercises32.3 (5.5)37.7 (4.4)0.46
 Splints28.6 (14.9)39.2 (10.8)0.58
Perceived helpfulness of treatments, 100-mm VAS   
 Medications80.1 (4.7)83.2 (3.4)0.60
 Exercises67.3 (6.2)64.6 (4.1)0.71
 Splints58.7 (18.9)67.1 (10.1)0.67
Figure 1.

The percentage of those with the described characteristic in each subgroup: users of complementary and alternative medicine (CAM) versus nonusers of CAM.

Table 3. Factors associated with use of CAM in JIA (logistic regression)*
 Adjusted odds ratio95% confidence interval
  • *

    CAM = complementary and alternative medicine; JIA = juvenile idiopathic arthritis.

Parental use of CAM5.131.35–19.46
Disease duration, months1.011.00–1.03
Income >$45,0002.410.35–16.56
Difficulty following treatment (medications)1.520.21–11.01
Disease severity (active joint count)1.150.89–1.48


In this study, 33.9% of patients with JIA used some form of CAM. The use of CAM was related to parental use of CAM and French Canadian ethnicity. Forty-four percent of families encountered problems with the management of their child's JIA. Difficulties with medications may be predictive of CAM use among JIA patients; however, adherence to conventional treatment remained high.

Our results were similar to those of a Danish study, where 4 (36%) of 11 children with joint diseases used CAM (20). However, this is considerably lower than the 64–70% described in 2 earlier studies of children with rheumatic diseases (10, 11). Our study included patients with JIA only, whereas the previous studies surveyed patients with a variety of pediatric rheumatology conditions. Also, the definition of CAM used may have differed between studies. Hagen et al (11) listed 20 products and services as types of CAM, with 5 of these being specific nutritional supplements and dietary changes. In our study, we listed 11 defined categories (as well as an “other, please specify” category) of which only 1 referred to dietary changes (special diets, vitamins, etc.); we therefore may not have captured all those who used specific vitamins and mineral supplements. Hagen et al (11) contended that had they excluded use of vitamins and minerals, CAM use would decrease to 51.8%. Southwood et al (10) found that copper bracelets, followed by dietary interventions, were the most frequently used types of CAM among adolescents who attended specialized arthritis youth camps in Australia and Canada. Their study included older participants with more severe disease than those in the present study, which may in part account for the higher level of CAM use. Moreover, CAM use was significantly higher in their subjects from Australia and New Zealand (86%) compared with those from Canada (39%). Considerably lower estimates of CAM use have been previously reported for children with musculoskeletal conditions (15%) in Montreal (2), “correctable pediatric conditions” (24%) in Arizona (21), pediatric oncology cases (42%) in Vancouver (22), and primary care pediatric patients (33%) in St. Louis (23).

Parents sought CAM primarily for pain relief for their children, similar to reasons reported for CAM use among adults with rheumatologic conditions (24). In our study, CAM use was not associated with any patient-specific characteristics, which concurs for the most part with Hagen et al (11). Hagen et al found that disease duration was strongly associated with CAM use. Although it might be reasonable to infer that patients with more protracted disease would be more likely to experiment with CAM use, we found only a questionable relationship between these 2 variables. It should be noted, however, that disease duration does not really address the issue of how well the disease is controlled; a patient who has carried a diagnosis of JIA for many years might have actually only had active arthritis for a short period within this timeframe. Although we were able to evaluate the number of active joints at the time of study, we are not able to determine whether there was any relationship between CAM use and duration of persistent arthritis.

CAM use was highly associated with parental use of CAM. This association has been shown in previous studies in other pediatric conditions (3, 21, 25–27). Family income may also be a factor related to use of CAM, because our study suggests that families with lower incomes used CAM less frequently than those with higher incomes. The universal health care system in Canada insures all residents for conventional treatments (including physical and occupational therapy) but does not generally cover CAM, for which parents must pay themselves. An association between higher household income and CAM use has been described in other studies (13, 28). We found no evidence to suggest that use of CAM was associated with parental distress or how the parents were coping with their child's illness. To our knowledge, this is the first study to have examined these psychosocial factors in relation to CAM use in JIA.

English Canadians tended to use CAM less than French Canadians in Montreal. It is possible that CAM use is valued more in French Canadian society. Other ethnic groups used CAM rarely, possibly because many of these are recent immigrants with fewer resources to expend on CAM. These findings associated with ethnicity concur with data on children with other physical disabilities (unpublished data by the author D. Feldman) and adults with rheumatologic problems (29).

Adherence to conventional treatment remained high in the CAM group and use of CAM was unrelated to perceived problems with medical management, although there was a nonsignificant trend toward higher CAM use in those perceiving difficulty with conventional medication use. Nevertheless, adherence to treatment in all 3 categories (medications, exercise, and splints) and perceptions of helpfulness of treatments did not differ between CAM users and nonusers. This concurs with the finding that CAM use does not seem to be associated with negative attitudes toward conventional therapy in adults (9).

This study was conducted within the context of a larger study on adherence to treatment. No causal inferences can be made due to the cross-sectional nature of the data. Our study is based on parental reports regarding CAM use for their children, which may underestimate actual use. It is possible that some of the patients who declined to participate in the study were greater users of CAM. If this was the case, and because those who declined had slightly higher duration of disease, our results may underestimate the association between disease duration and CAM use.

In conclusion, the use of CAM is common in patients with JIA and is related to parental use of such therapies, French Canadian ethnicity, possibly longer disease duration, and higher levels of family income. Although difficulties with medications may be predictive of CAM use among JIA patients, adherence to conventional treatment nevertheless remained high in this group. CAM use was not related to any decreased adherence to conventional medical treatment. Possible side effects of CAM and interactions with medical therapy make it important for rheumatologists to inquire about the use of CAM in their patients.