Search tips
Search criteria 


Logo of ijayurvedaHomeCurrent issueInstructionsSubmit article
Int J Ayurveda Res. 2011 Jan-Mar; 2(1): 8–13.
PMCID: PMC3157120

The efficacy of Ayurvedic treatment for rheumatoid arthritis: Cross-sectional experiential profile of a longitudinal study



Allopathic and Ayurvedic physicians collaborated on a study of traditional medicine, which was sponsored by the World Health Organization.


The aim of the study was to test the efficacy and safety of Ayurvedic treatment for rheumatoid arthritis (RA).

Settings and Design:

This study was conducted at the Ayurvedic Trust, Coimbatore, India.

Materials and Methods:

In this unique study of classical Ayurvedic treatment for RA, allopathic physicians enrolled a total of 290 patients with a confirmed diagnosis of RA over a 7-year period, and once every 6 weeks evaluated Ayurvedic treatment outcomes on the basis of American Rheumatism Association criteria: grip strength, walking time, number of swollen and painful joints, joint count, functional class, erythrocyte sedimentation rate, and rheumatoid factor. Ayurvedic physicians administered individualized treatment, closely adhering to principles set forth in classical Ayurvedic texts. The duration of treatment varied from 1 to 6 months.

Statistical Analysis Used:

Due to limitations in computer technology in the 1970s, the data were not computerized. Therefore, data for 12 months at a time were analyzed, using repeated measures t-test. Measures of central tendency (means) and probability values were reported. Results from the patients enrolled and discharged at the end of the first year of the study (N = 33) are presented in this paper.


There was statistically significant improvement in all parameters from admission to discharge.


The results indicated that classical Ayurvedic treatment was effective in this first cohort of patients who completed treatment. Even patients with severe functional limitations showed significant improvement. Although there was no control group, the results are positive enough to warrant further study of classical Ayurvedic treatment for RA in controlled trials.

Keywords: Ayurveda, longitudinal study, rheumatoid arthritis


Ayurveda has been recognized by the World Health Organization (WHO) as a complete system of natural medicine, but it is not widely known that the first-ever study of a traditional medical system sponsored by WHO was of complete, classical Ayurvedic treatment for rheumatoid arthritis (RA), conducted in collaboration with the Indian Council for Medical Research (ICMR) and the Ayurvedic Trust, Coimbatore, Tamil Nadu, India, from 1977 to 1984.[1] This unblinded, longitudinal study entitled, “The WHO/ICMR Study of the Efficacy of Ayurvedic Treatment for Rheumatoid Arthritis,” enrolled 290 patients with RA over a 7-year period, and was conducted at the Ayurvedic Trust in Coimbatore, India.

This study is unique in at least two ways: (1) in an unprecedented move, the Ayurvedic Trust opened up their treatment methods to the scrutiny of allopathic physicians and allowed them to formally evaluate Ayurveda according to rigorous allopathic criteria; and (2) It was the first time that both Ayurvedic and allopathic physicians collaborated on a study of Ayurvedic medicine, in which the Ayurvedic physicians exclusively provided the treatment. Statistical analysis was performed by an expert medical statistician. Each patient was evaluated at several different time points using the American Rheumatism Association (ARA) criteria,[2] thus yielding baseline controlled longitudinal repeated measures data. Unpublished reports of the allopathic experts’ assessments and statistical analysis together indicate that, in this study, Ayurvedic treatment for RA was safe and effective, and provided symptomatic relief without harmful side effects.

There are no published studies that have tested Ayurveda as a whole system or its multiple modalities for the same disease at the same time.[3] Although there have been several studies of Ayurvedic treatment for various conditions, including RA,[46] they do not constitute true tests of classical Ayurveda because they did not allow for individualization of therapy, nor did they follow classical texts of Ayurveda for the treatment of RA.[7] Although three decades have passed since its inception, this study of classical Ayurvedic treatment for RA still remains the only one of its kind.

While allopathic treatment of RA is improving, remission remains rare and treatment remains unsatisfactory.[8] Therefore, a search for effective alternative and additional therapies for this disease continues. There is a growing interest in this study[9] and importantly, it was the inspiration for a recently completed double-blind, placebo-controlled, randomized pilot clinical trial, sponsored by the National Institutes of Health, USA, of classical Ayurvedic outpatient treatment for RA.[10]

The data archive housed at the Ayurvedic Trust contains descriptive analyses for every 12 months of the study for preparing the annual reports. In this paper, we present the outcomes for the first cohort of patients who completed their treatment during the first year of the study (1977–1978).


The ICMR team was responsible for the study design, confirmation of RA diagnosis, and enrollment of patients, and for the evaluation of the efficacy of Ayurvedic treatment, based on ARA criteria.[2] The Ayurvedic physicians administered the treatment, closely adhering to the principles set forth in classical Ayurvedic texts.[11] There was no control group.

Potential participants for the study were drawn from the outpatient department of the Ayurvedic Trust hospital and selected on the basis of Ayurvedic criteria for vatarakta by the Ayurvedic panel consisting of three Ayurvedic physicians. Participants who had diseases and conditions which would likely delay or alter the Ayurvedic treatment plan for vatarakta were excluded. The exclusion criteria jointly decided by the Ayurvedic and allopathic investigators were: Jaundice, diabetes, dysentery, pleurisy, tuberculosis, cardiovascular disease, leprosy, syphilis, ulcerative colitis, severe arthritis, iritis, asthma, epilepsy, mental illness, Parkinson's disease, herpes, neuropathy, and chronic urinary infection.

The selected patients who were all confirmed as having vatarakta were then sent to the allopathic panel consisting of four physicians, who evaluated them for RA on the basis of histological, serological, biochemical, and radiological tests. Only those with a confirmed diagnosis of RA, as determined by the allopathic panel, were admitted to the study as inpatients to undergo Ayurvedic treatment and periodic evaluations by both allopathic and Ayurvedic physicians. Of the patients referred by the Ayurvedic physician, the allopathic panel confirmed approximately 66% as having RA.

Since this was a study of classical Ayurvedic treatment, the Ayurvedic physicians were free to prescribe any combination of medicines and therapies based on their clinical judgment. Thus, the patients received individualized therapy as per principles specified in the Ayurvedic classical texts. Treatment included several pharmacological dosage forms of internal herbal medicines (kashayams, arishtams, gulikas, lehyam, and choornam), specialized oil therapies (sneha-sweda), purificatory therapies (panchakarma chikitsa) including medicated enema (vasti, or basti, as it is commonly known) and therapeutic purgation (virechana), external application of analgesic herbal pastes (lepa), medicated oils (oushadha siddha taila), and dietary and lifestyle modification. Patients were in treatment until they were considered well enough for discharge, which varied between 1 and 6 months after admission.

The evaluation of the efficacy of Ayurvedic treatment was done once every 6 weeks exclusively by the allopathic panel, on the basis of criteria established by the ARA once every 6 weeks.

Outcome measures (ARA Criteria)

The following were assessed at admission, once every 6 weeks during treatment, and at discharge: grip strength (mm of mercury), number of painful joints, number of swollen joints, walking time (25 and 50 ft, s), joints with active RA, joint count, functional status, erythrocyte sedimentation rate (ESR) at 0.5 and 1 h, and rheumatoid factor (RF).

Safety assessments

Also once every 6 weeks, the following tests were done: liver function tests such as SGOT, SGPT, serum protein, serum bilirubin, serum alkaline phosphatase, and prothrombin index; renal function tests including blood urea and serum creatinine; and complete blood count measures including white blood count, lymphocytes, PCV%, and hemoglobin.


Data were recorded by both the allopathic and Ayurvedic physicians in separate records for each patient. Outcome and efficacy endpoints were extracted for analysis by the medical statistician.


Results are presented for the first cohort of patients who were discharged at the end of the first year of the study. The Ayurvedic physicians screened and diagnosed 100 patients as having vatarakta. These prospective patients underwent preliminary screening by an allopathic physician and 80 out of the 100 were provisionally diagnosed as having RA. Nine out of 80 failed to report to the allopathic hospital for the confirmation of diagnosis. The allopathic panel confirmed an RA diagnosis in 66 of these remaining 71 patients. Two of these were not willing to undergo inpatient treatment, leaving 64 patients to be admitted to the study. Since patients were recruited on an ongoing basis, not all of them completed the study at the same time. At the end of the first year, results for 33 patients were available for statistical analysis.

Characteristics of patients at admission

There were more females (61%) than males. About 75% of the patients were in the age range of 15–44 years, and approximately 60% were classified as being in functional class III (limited only to little or none of the duties of usual occupation or self-care) and functional class IV (incapacitated, largely or wholly; bedridden or confined to a wheelchair; little or no self-care). Approximately, 25% of patients had had RA for more than 5 years, 50% for 1–4 years, and the rest for less than 1 year [Table 1]. The length of treatment varied from 1 to 6 months.

Table 1
Demographic and background characteristics at admission (N = 33)

Efficacy outcomes

The outcome measures for assessing the change between admission and discharge included grip strength, walking time (50 feet and 25 ft), number of swollen and painful joints, joint count, ESR, and RF. Table 2 shows the change in parameters from admission to discharge as well as the scoring system used to calculate the joint count.

Table 2
Mean values of the change from admission to discharge (N = 33)

Grip strength

Grip strength was measured by a sphygmodynamometer inflated to 30 mmHg. A mean of six readings, three for each hand measured alternatively was taken. The mean grip strength of both hands was less than 150 mmHg in 91% of patients at the time of admission. At discharge, this proportion decreased to 73%. The mean grip strength for all 33 patients was 82 mmHg at admission and 111 mmHg at discharge, a statistically significant increase (P < 0.001).

Walking time

Walking time is measured as the time in minutes and seconds taken to walk a distance of 25 ft and 50 ft. At admission, 12% of the patients took 15 s or more to walk a distance of 25 ft and 12% could not walk at all. At the time of discharge, however, all the patients could walk, and 94% were able to walk a distance of 25 ft in less than 10 s. The mean walking time for 25 ft was 7.4 s at admission and 4.8 s at discharge, a statistically significant decrease (P < 0.001).

Swollen joints

The proportion of patients with at least 10 swollen joints was 27% on admission, but was only 6% at discharge. The mean number of swollen joints was 6.6 initially and 4.3 at discharge, a significant reduction (P < 0.001).

Painful joints

Whereas 33% of the patients had at least 10 painful joints initially, the figure was only 6% at discharge. In terms of mean values, the number of painful joints decreased from a mean of 7.8 at admission to 3.6 at discharge (P < 0.001).

Joint count

A joint count of 75 or more was observed in 52% of the patients initially, compared with only 18% at discharge. Conversely, no patient had a joint count of 0 to begin with, but a third of the patients did so at discharge. The mean joint count decreased from 74 to 30, which is statistically significant (P < 0.001).

Erythrocyte sedimentation rate

Considering the findings at 1 h, 48% of the patients had an ESR of 50 mm or more initially, compared with 27% at the time of discharge; the means were 59 and 37, respectively (P < 0.01). Findings at 0.5 h showed the same pattern, with a statistically significant reduction in the mean from 30 mm to 17 mm (P < 0.01).

Functional class

Whereas 19 (58%) of the 33 patients were classified as belonging to functional class III or IV at the time of admission, only 4 were classified as such at the time of discharge [Table 3]. In all, there were 16 patients (48%) for whom no change in the functional class occurred, including 7 classified as functional class I. An improvement by one class was observed in 11 patients (33%), by two classes in 2 (12%), and by three classes in 2 (6%). The overall improvement in the group was highly significant (P < 0.001).

Table 3
Functional class at admission and discharge (N = 33)

The influence of various background and pre-treatment factors (such as sex, age, disease duration, and functional class) was examined. In general, the improvement in males was better than in females [Table 4]. In the case of grip strength, walking time, and joint count, the contrasts were appreciable and statistically significant. Age did not have an impact on progress (results not shown). There was some evidence that patients who were in functional classes III and IV improved more than those in functional classes I and II, especially with respect to grip strength, walking time, and joint count [Table 5]. Although none of the differences was statistically significant, they were all in the direction of improvement.

Table 4
Influence of gender on progress from admission to discharge (N = 33)
Table 5
Influence of functional class on progress (N = 33)

There was clear evidence that patients with a short history of RA (less than 1 year) improved more than those with longer histories [Table 6]. The differences were highly significant in the case of grip strength, walking time, and joint count.

Table 6
Influence of disease duration on progress from admission to discharge (N = 33)

Rheumatoid factor

Only 3 (9%) patients never had a positive finding and 22 patients (67%) had a titer of 1/80 or more, including 5 (15%) with a titer of 1/640 or 1/1280. The RF was determined on admission and discharge in only 15 (45%) of the 33 patients. In 12 patients, there were at least two doubling dilution steps. Two patients showed no change and one had a rise in titer from 1/20 to 1/80. The improvement in the group as whole (12 decreases compared to 1 increase) was statistically significant (P < 0.01).

Follow-up of patients was not done routinely for all patients during the first year of the study. Consequently, follow-up data were available for only 14 (42%) of patients 2-4 months after discharge (not shown). Except for grip strength, which showed a statistically significant increase (111 to 129, P < 0.01), all other parameters showed a very slight decline from discharge to follow-up, and bordered on statistical significance. Compared with mean values at admission, however, follow-up values indicated improvement.


This first-ever study of traditional medicine sponsored by the WHO and conducted almost three decades ago remains the only known study of complete classical Ayurvedic inpatient treatment for RA. This analysis of the first cohort of patients discharged from the study clearly indicates that they improved considerably as a result of Ayurvedic treatment. There was improvement (statistically significant) in the patients’ condition from admission to discharge, according to all the ARA criteria used to evaluate the effectiveness of Ayurvedic treatment.

Women constituted approximately two-thirds of the sample. They were affected more severely by the disease and improved more slowly than men. Pretreatment factors such as disease duration and functional class showed interesting patterns of improvement. While overall improvement was greatest in those who were in the early stages of RA, even those who were in the more advanced stages (functional classes III and IV) showed significant improvement. Records indicate that in general, reduction in swelling was noted within a month, and 80% reported relief from pain in the first month after starting treatment. There was no evidence of liver, renal, or other toxicity due to Ayurvedic treatment.

Given the form of the data available in the archives, we have been limited to presenting the results of only a subset of patients in this paper. Other limitations related to the overall study design must be acknowledged and considered when interpreting the results of this study. First, there was no control group, which unfortunately limits the generalizability of this study. Second, patients taking steroids were also enrolled, and instead of having the drug gradually tapered and stopped as is medically recommended, the study protocol called for stopping the drugs abruptly upon admission. The resulting severe withdrawal symptoms delayed full Ayurvedic treatment in these patients (steroid group) by a few weeks. In this first cohort, steroid and nonsteroid patients were not compared with each other, but data were analyzed separately in subsequent cohorts for steroid and nonsteroid groups. Those analyses showed that the progress of the steroid group was slower than those who were not on steroids. The average length of treatment was 3 months for the nonsteroid group compared with 6 months for steroid group. Third, the admission criteria did not exclude children who would be considered as suffering from juvenile RA, a disease entity treated differently from adult RA.

Despite the abovementioned limitations, it is a testament to the efficacy of Ayurvedic treatment for RA that the allopathic outcomes measured in this study were statistically significant and positive. It is reasonable to believe that classical Ayurveda, with its hallmark individualized, holistic treatment, was responsible for the improvement noted in this varied group of patients regardless of disease duration, functional status, and age.

An unexpected consequence of the study's methodology underscored Ayurveda's strength in diagnosing RA in the prodromal stages when symptoms are very subtle. Only two-thirds of those initially diagnosed by the Ayurvedic physicians were confirmed by the allopathic panel as having RA and included in the study. The Ayurvedic physicians independently and successfully treated the excluded individuals for vatarakta. As these patients were in the very early stages of the disease, they benefited greatly from the treatment, supporting the allopathic view that treating RA in its early stages yields better outcomes.

This study served to bring attention to Ayurveda, at a time when it was trying to transcend centuries of colonial neglect,[12] and even when judged by allopathic criteria, it showed that Ayurvedic treatment was successful in treating a complex, chronic disease like RA. Of equal or perhaps greater importance might be the fact that the Ayurvedic physicians documented treatment and outcomes (according to Ayurvedic criteria) meticulously and innovatively. They designed questionnaires based on Ayurvedic principles, defined variables according to doshas and stage of disease, and performed quantitative analysis of Ayurvedic outcome measures. The Ayurvedic physicians were also able to test their own theories in this way. For example, the expected degree of improvement based on Ayurveda's stage-wise classification and its prediction that patients of a certain constitution type (a vata-pitta combination) would be more prone to developing the disease were supported. The equivalence of the classical Ayurvedic diagnosis of vatarakta[11] and the allopathic diagnosis of RA as per ARA criteria[2] was also confirmed. Thus, this study provides credence to the observation that a multifaceted traditional medicine system can lend itself to quantitative and objective analysis.

Approximately 160 full sets of patient records (Ayurvedic and allopathic) are available in the study archives. There is a wealth of longitudinal data from clinical allopathic examinations, separate Ayurvedic evaluations, and self-evaluations of patients. The fact that Ayurvedic treatment was documented in such great detail makes it a valuable data source for learning more about classical Ayurvedic treatment for RA and for designing controlled studies in the future to reveal its full therapeutic potential.

A computerized database of this rich and unique data resource can help researchers generate and test hypotheses regarding the efficacy of Ayurvedic treatment in ways not envisioned by the original investigators, by taking advantage of advances in statistical software and methodology. Given the detailed documentation of Ayurvedic treatment, there is also great potential for qualitative analysis and serving as a valuable learning resource for students of Ayurveda.


Source of Support: Nil

Conflict of Interest: None declared.


1. World Health Organization/Indian Council for Medical Research Collaborative Study on the Efficacy of Ayurvedic Treatment in Rheumatoid Arthritis. Coimbatore, India: The Ayurvedic Trust; 1984. Ayurvedic Trust.
2. Ropes M, Bennett GA, Cobb S, Jacox R, Jessar RA. 1958 Revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis. 1958;9:175–6. [PubMed]
3. Hardy M, Coulter I, Venuturupalli S, et al. Ayurvedic Interventions for Diabetes Mellitus: A Systematic Review.Evidence Reports/Technology Assessments, No. 41. Rockville (MD): Agency for Healthcare Research and Quality (US); 2001. Sep,
4. Chopra A, Lavin P, Patwardhan B, Chitre D. Randomized double blind trial of an Ayurvedic plant derived formulation for treatment of rheumatoid arthritis. J Rheumatol. 2000;27:1362–5. [PubMed]
5. Kulkarni RR, Patki PS, Log VP, Gandage SG, Patwardhan B. Efficacy of an Ayurvedic formulation in rheumatoid arthritis: A double blind, placebo controlled, cross-over study. Indian J Pharmacol. 1992;24:98–101.
6. Sander O, Herborn G, Rau R. Is H15 (extract of Boswellia serrata,“incense”) an efficient supplementation to established drug therapy in RA? Results of a double blind pilot trial. Z Rheumatol. 1998;57:11–6. [PubMed]
7. Hardy ML. Research in Ayurveda: Where do we go from here? Altern Ther Health Med. 2001;7:34–5. [PubMed]
8. Firestein GS, Budd RC, Harris ED, McInnes IB, Sergent JS. Kelley's Textbook of Rheumatology. 8th ed. Philadephia PA: WB Saunders; 2008.
9. Venkatraman M, Assefi N. Ayurveda: General Principles and Treatment of Rheumatoid Arthritis. Alt Med Alert. 2002;5:144–7.
10. Furst DE, Venkatraman MM, McGann M, Manohar PR, Booth-Laforce C, Sarin R, Sekar PG, Raveendran KG, Mahapatra A, Gopinath J, Kumar PR. Double-blind, randomized, controlled, pilot study comparing classic Ayurvedic medicine, methotrexate, and their combination in rheumatoid arthritis. J Clin Rheumatol. 2011 Jun;17:185–92. [PubMed]
11. Varier KS. Ayurvedic Treatment of Rheumatoid Arthritis. Swasth Hind. 1980;5:334–6.
12. Banerji D. The place of indigenous and western systems of medicine in the health services of India. Soc Sci Med. 1981;15A:109–14. [PubMed]

Articles from International Journal of Ayurveda Research are provided here courtesy of Medknow Publications