TO THE EDITOR: This article repeated a common misconception about the Glucosamine/Chondroitin Arthritis Intervention Trial (GAIT), interpreting it as showing that the combination of glucosamine and chondroitin is effective for treating moderate to severe osteoarthritis.1 The study found that glucosamine and chondroitin, separately or in combination, were not more effective than placebo. Only one of 10 subgroups showed statistically significant results, and these were for moderate to severe arthritis. With 10 subgroups, it is likely that chance alone would produce false-positive results in one of these groups. The authors of the study warned that it was not powered to differentiate among subgroups, and that no clinical recommendations should be made based on that finding.1
The AFP article recommends a combination of glucosamine and chondroitin as the third step in a stepped-care approach for the treatment of osteoarthritis. This is not justified by the evidence. In addition, the rationale is suspect because glucosamine and chondroitin are produced by the body. The amount added by taking supplements is only a minuscule fraction of what is already present. These are not essential nutrients like vitamins, of which taking a small amount is likely to make a large difference.
IN REPLY: I appreciate Dr. Hall's thoughtful comments. She is correct that the GAIT trial did not demonstrate a beneficial effect of glucosamine combined with chondroitin in most patients with osteoarthritis of the knee. However, there was a statistically significant benefit in the subgroup of patients with moderate to severe osteoarthritis. Of the 1,583 patients in the trial, 354 (22 percent) fell into this subgroup.1
I find that forest plots are useful in helping me visualize the effectiveness of various treatments.2 There is a set of forest plots in Figure 2 of the GAIT trial article.1 The figure shows that the combination of glucosamine and chondroitin was more effective than placebo for moderate to severe osteoarthritis based on two scoring systems: the Western Ontario and McMaster Universities Osteoarthritis Index, and the Outcome Measures in Rheumatology Clinical Trials and Osteoarthritis Research Society International. These are well-accepted clinical measures, and the differences reached statistical significance in each case.1
Dr. Hall states that “with 10 subgroups, it is likely that chance alone would produce false-positive results.” However, if statistical significance is defined as a P value of .05 or less, that corresponds to a rate of one in 20 being false-positive by chance alone, not one in 10.
Additional studies are needed to confirm the effectiveness of glucosamine and chondroitin for the treatment of osteoarthritis. For now, I stand by the recommendation for a brief trial of combined glucosamine and chondroitin in patients who have progressed to moderate or severe osteoarthritis, although the stepped-care approach in my article should state that this is only for osteoarthritis of the knee, per the evidence in the GAIT trial.