Background/Purpose:  Studies have identified widespread use of complementary and alternative medicine (CAM), particularly in patients with rheumatic disease.  Most have reported results over short time frames, i.e. ”used in last year” or “ever use”; few have evaluated long-term (LT) use of CAM.  In addition, studies have often looked at all patients in a rheumatology practice rather than those with rheumatoid arthritis (RA) only.  The purpose of this study was to quantify the frequency of ever and LT use of CAM in patients with RA and to identify characteristics associated with ever and LT CAM use.

Methods:  Data from RA patients participating in the Consortium of Rheumatology Researchers of North America (CORRONA) registry, an independent prospective observational cohort with >30,000 RA patients enrolled from over 100 academic and private practices across the US, were examined.  Patients provided data regarding the use of glucosamine/chondroitin, fish oil, borage seed oil, evening primrose oil, and flax seed oil at clinic visits.  Other CAM use such as acupuncture, massage, or yoga was not collected.   Patients with at least two years of follow-up and at least three visits were included in the analysis.  Primary outcome was any CAM use within 2 years of first visit. LT use was defined as use of the same CAM at 3 consecutive visits or all visits in one year.  Logistic regression was used to calculate odds ratios (ORs) of CAM use (ever or LT) by patient demographics, disease characteristics, and medication history.

Results: 11,970 patients were included in the analysis; 35.2% reported any CAM use, but only 10.8% reported LT use (p<0.0001).  Fish oil was the most common CAM reported (27.3%).  Without adjusting for other factors, patient demographics, medication history, and lower disease activity [Disease Activity Score (DAS28), tender/swollen joint count, modified Health Assessment Questionnaire (mHAQ), Clinical Disease Activity Index (CDAI), physician/patient assessments] were associated with ever and LT CAM use (Table 1). Separate multivariate models for ever and LT CAM use had the listed predictors in common (Table 1).

 Table 1. Patient characteristics at first visit associated CAM use

 

	OR No Adjustment		OR (95% CI) With Adjustment1
Demographics	Ever CAM	LT CAM	Ever CAM	LT CAM
Older age	1.005#	1.006#	N	N
White	1.160#	1.152*	1.25 (1.09 – 1.45)#	N
Current smoker	0.611*	0.556*	0.70 (0.62 – 0.79)*	0.71 (0.57 – 0.87)#
Education: high school or less vs any college	0.641*	0.523*	0.69 (0.63 – 0.75)*	0.58 (0.51 – 0.67)*
Married – vs windowed                  – vs single	0.732* 0.815*	0.651* 0.718*	0.72 (0.62 – 0.83)* 0.86 (0.74 – 1.00)*	0.66 (0.52 – 0.84)# 0.68 (0.53 – 0.88)#
Region – Western vs Midwest              Northeast vs Midwest                      South vs Midwest	1.961* 0.817* 1.122*	2.269* 0.847* 1.141*	1.49 (1.29 – 1.72)* 0.83 (0.74-0.93)* 0.88(0.79 – 0.997)*	1.67 (1.36 – 2.03)* 0.85 (0.71-1.02)* 0.80 (0.66-0.97)*
Part-time work vs full-time	1.159#	1.244#	1.17 (1.02 – 1.35)#	1.33 (1.08 – 1.64)#
More frequent visits (8+ vs 3 visits)	1.542*	2.024*	1.83 (1.53 – 2.19)*	2.63 (1.98 – 3.50)*
First visit in 2006 or later	2.014*	2.450*	1.83 (1.67 – 2.00)*	2.21 (1.90 – 2.56)*
Disease characteristics
Duration of RA (years)	0.995#	0.994#	N	N
Older at RA onset  (years)	1.007*	1.008*	1.01 (1.01 – 1.02)*	1.01 (1.01 – 1.02)*
Deformities present	0.900#	1.003	N	1.22 (1.06 – 1.41)#
Disease activity
DAS 28	0.932#	0.873*	N	N
CDAI	0.994#	0.990*	N	N
Number of swollen joints	0.990#	0.992	N	N
Number of tender joints	0.995	0.981#	N	N
mHAQ	0.898#	0.784#	N	N
Physician assessed disease activity (scale 1-100)	0.997#	0.994*	N	N
Patient assessed disease activity (scale 0-100)	0.997#	0.994*	N	0.996 (0.993 – 0.999)#
Medical History
NSAID use	1.274*	1.418*	1.31 (1.20 – 1.43)*	1.49 (1.29 – 1.72)*
Exposure to 3+ DMARDs vs 0	1.293#	1.131	1.30 (1.13 – 1.49)#	N
Patient report anxiety/depression	1.188*	1.011	1.19 (1.09 – 1.31)#	N

¹Multivariate models built separately for ever and LT CAM using stepwise logistic regression. 

* p<0.0001, #p<0.05

N = not included in model

Conclusion: The results show significant differences in patient characteristics between CAM users and non-users, but clinical characteristics are similar suggesting that patient characteristics rather than disease severity are the driving force behind CAM use. CAM use with first visits after 2006 suggests that CAM use is increasing.  Although “ever use” of CAM was 35.2%, the LT use was significantly lower, 10.8%, implying that patients may try complementary therapy, but few continue.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/use-and-long-term-use-of-complementary-and-alternative-medicine-in-rheumatoid-arthritis-patients/