The major purpose of this meta-analysis was to update and critically evaluate the effects of TC training on arthritic symptoms and physical function in older patients with OA. Our meta-analysis suggests that 12-week TC significantly improves pain, stiffness, and physical function in patients with knee OA, which indicates that TC has benefits in the management of OA and should be available in rehabilitation programs as an alternative approach for patients with knee OA.
The primary goals in the management of OA are currently to alleviate arthritic symptoms, including pain and stiffness, maintain or improve joint mobility and quality of life, increase muscle strength, and minimize the disabling effects of OA 
. Rehabilitation is regarded as an effective non-pharmaceutical therapy in the management of OA 
. However, very few OA patients participate in any type of rehabilitation for fear of falling and exacerbating arthritic symptoms, which results in deconditioning and loss of physical function 
. Even those who participate in a rehabilitation program show poor adherence 
. For individuals with OA, rehabilitation intervention should be pursued cautiously because general exercise can apply either injurious or beneficial effects on the joints. Recent studies have evaluated the role of TC, which enhances balance, strength, flexibility, and self-efficacy, and decreases pain and stiffness in various patients with chronic conditions. TC is a potential option for the management of OA and is superior to other forms of rehabilitation for elders because it involves a series of gentle fluid movements reputedly good for maintaining mobility and gradually improves muscle strength and range of motion without exacerbating arthritic symptoms 
. Growing evidence suggests that TC may reduce arthritic symptoms and/or improve physical function in patients with OA 
. However, other trials failed to investigate these positive effects and were unable to draw a positive conclusion 
Our results showed that 12-week TC is effective at reducing pain and stiffness and improving physical function in patients with knee OA. Subgroup analyses suggested that 8–10 weeks of short-term TC can significantly improve pain and physical function, and 18–24 weeks of TC improves physical function. Theoretically, TC could be more effective over the long-term, but the positive effects of 12-week TC were not sustained after 6–12 weeks duration, which is consistent with previous findings 
. This change with the long-term TC exercise is interesting, but additional studies are needed to investigate the long-term effects of TC in patients with knee OA. In addition, most of the patients in the RCTs included in our study were elderly females and the OA site was primarily the knee, which is consistent with the current epidemiology of OA 
Recent efforts have suggested a minimal clinically important difference (MCID) for WOMAC scores from both pharmacological and rehabilitation trials. Changes of 20–25% in the WOMAC score are considered to be clinically relevant 
, but the most recent study suggested that a 16–18% reduction in the WOMAC score is associated with the MCID and should be appropriate for use in the interpretation of clinical studies, as well as in clinical care 
. Our results indicate that a reduction of 32.2–36.4% from baseline was greater than the MCID of 16–18% or 20–25%, which suggests that TC has beneficial effects on pain, stiffness, and physical function in patients with knee OA.
Our results are similar to the latest SR 
. In detail, this previous SR showed that TC may be effective at controlling pain and improving physical function in patients with knee OA. However, the authors did not compare their results with the MCID because the results were difficult to compare quantitatively due to the use of different assessment measures for evaluating outcomes. Therefore, we pooled the outcome measures (e.g., pain, stiffness, and function) assessed by the same WOMAC score in order to compare the results with the MCID. Our results indicate the presence of sufficient clinical evidence of reduced pain and stiffness and improved physical function.
The possible mechanisms responsible for the beneficial effects of TC that differ from other forms of exercise are still unclear. TC harmonizes yin-yang and promotes homeostasis between body and mind. TC is a lower intensity exercise of flowing circular movements, balance and weight shifting, deep breathing regulation and meditation, and visualization, and focuses on internal awareness 
. TC encourages patients to move fluidly with less strain, and improved joint stability and decreased joint pain may be beneficial for patients with knee OA. In addition, the movement characteristics of slowness, quietness, and stillness inherent to TC and its steady rhythm and slow movements aid in relaxation and offer beneficial changes in symptoms and mood, which may promote psychological well-being and positively influence chronic pain in patients with knee OA 
. Therefore, the nature of TC and the multiple potential effects on the body and mind that differ from other conventional exercise may account for these beneficial effects; however, further studies are needed to better understand the benefits, mechanisms, and role of TC in the prevention and management of OA.
We found that most studies lacked other objective outcome measures, including exercise performance (e.g., 6-min walk distance), quality of life, body mass index, muscle strength, immune function, and survival, which would result in more reliable and convincing evidence of the effects of TC in patients with OA. Furthermore, comparing TC with general forms of exercise, such as jogging and motion or flexibility exercises, would be better, but this method has not seen much use in clinical research. Therefore, focusing on these additional interesting clues may be useful for future research on the topic. In addition, future researchers should attempt to understand the relationships among impairment, functional limitations, and disability.
Finally, we found no significant side effects or adverse events associated with TC, and participants had relatively high adherence in most studies, indicating that TC is safe and has satisfactory compliance. Given no special setting, no additional costs, independence from weather conditions, and multiple benefits to the body, TC should be an alternative to other exercise training and be incorporated into rehabilitation programs as a potential non-pharmacological treatment for patients with OA.
This study had numerous limitations. First, our analysis is based on seven RCTs, all of which had a small sample size. Overestimation of the treatment effect is more likely in smaller trials compared to larger trials. Although we performed a funnel plot for the outcomes, the limiting RCTs make it difficult to interpret the result of publication bias. Moreover, a major limitation of our subgroup analyses is that some (<12 and >12 weeks) are based only on 2 to 3 studies; thus, the conclusions about the duration of TC exercise should be interpreted with caution. Next, the targeted population varied greatly (e.g., patients of different gender, ethnicity, and duration of OA). The adopted TC protocols differed. These factors may have a potential impact on our results. Finally, some missing and unpublished data may lead to bias.
In summary, the positive findings of this study suggest that 12-week TC has beneficial effects on the management of knee OA, including reduced pain and stiffness and improved physical function. As an alternative, effective, inexpensive, and accessible approach, TC should be available in rehabilitation programs. However, given the heterogeneity among study designs and small RCTs, additional larger scale RCTs are needed to substantiate the current findings and investigate the long-term effects of TC in patients with knee OA.