Knee Osteoarthritis Improvements Using Telehealth Exercise Programs
Overview
This study aimed to investigate the factors that mediate the impact of two 6-month telehealth-delivered exercise programs, with and without a weight-loss diet, on pain and function improvements in individuals with knee osteoarthritis (OA). The secondary analysis involved 345 participants from a 3-arm randomized controlled trial, comparing exercise alone (Exercise program), exercise with a weight-loss diet (Diet + Exercise program), and information only (Control program). The outcomes measured were changes in pain (using an 11-point numeric rating scale) and function (assessed by the Western Ontario and McMaster Universities Osteoarthritis Index, score range 0–68) at the 12-month mark. Potential mediators considered included changes at 6 months in attitudes toward self-management, fear of movement, arthritis self-efficacy, weight, physical activity, and willingness for knee surgery.
For the Exercise program versus the Control program, possible mediators included a reduction in fear of movement (associated with a –1.11 unit improvement in function) and an increase in arthritis self-efficacy (linked to a –0.40 unit reduction in pain and a –1.66 unit improvement in function). In the Diet + Exercise program versus the Control program, potential mediators included a reduced fear of movement (–1.13 unit improvement in function), increased arthritis self-efficacy (–0.77 unit reduction in pain and a –5.15 unit improvement in function), and weight loss (–1.20 unit reduction in pain and a –5.79 unit improvement in function). Notably, weight loss was identified as a mediator for the Diet + Exercise program versus the Exercise program (–0.89 unit reduction in pain and a –4.02 unit improvement in function).
In conclusion, this study suggests that increased arthritis self-efficacy, reduced fear of movement, and weight loss may partially mediate the effects of telehealth-delivered exercise programs, with or without diet, on pain and/or function in individuals with knee OA. Additionally, weight loss was identified as a partial mediator for the combined diet and exercise program compared to exercise alone. These findings contribute valuable insights into potential factors influencing the outcomes of telehealth-delivered interventions for knee osteoarthritis.
Introduction
Clinical guidelines universally recommend education, exercise, and weight loss as primary strategies for managing knee osteoarthritis (OA). Specifically, for overweight or obese individuals, the combination of a weight-loss diet with exercise demonstrates superior outcomes compared to individual treatments. However, the understanding of the mechanisms through which exercise and diet impact pain and physical function in knee OA remains limited. This gap in knowledge hinders the optimization of treatment programs and their potential to alleviate the substantial burden of this condition.
Despite the prevalent recommendations for diet and exercise, their clinical benefits for knee OA are only modest. To enhance the effectiveness of interventions, a comprehensive understanding of the underlying mechanisms is crucial. Causal mediation analysis, utilizing data from randomized controlled trials (RCTs), offers a robust method to investigate the causal links between intermediate variables (mediators) and the impact of interventions on outcomes. Previous studies on nonsurgical interventions for OA have suggested potential mediators such as reduced inflammation, body weight, increased muscle strength, and heightened self-efficacy. Notably, while some studies have explored the physical aspects of exercise, psychosocial factors like self-efficacy or fear of movement remain understudied.
In a recent RCT employing telehealth-delivered programs, both exercise with and without a weight-loss diet exhibited improvements in pain and physical function over 6 and 12 months compared to a control group. The combined diet and exercise program showed additional benefits over the exercise-only group. However, the specific mechanisms driving the effectiveness of these interventions remained unclear. To address this gap, the authors conducted causal mediation analysis using RCT data to identify potential mediators influencing the effects of exercise programs on pain and physical function in overweight or obese individuals with knee OA. This research aims to provide valuable insights into the nuanced pathways through which exercise, with or without diet, impacts the symptoms of knee OA.
Methods
This secondary analysis draws from a randomized controlled trial (RCT) evaluating the impact of exercise, with and without a weight-loss diet, on knee osteoarthritis (OA) in participants recruited through Australia’s largest private health insurer, Medibank Private. The study, approved by The University of Melbourne Human Research Ethics Committee, involved participants meeting specific criteria, including age, OA symptoms, knee pain severity, and body mass index.
The Control group accessed an informative website covering OA-related topics. The Exercise group underwent a 6-month program with physical therapist consultations, personalized exercise plans, and supporting materials. The Diet + Exercise group received additional videoconferencing consultations with a dietitian, guiding them through a ketogenic very low-calorie diet (VLCD).
Outcomes, self-reported via online questionnaires at baseline, 6 months, and 12 months, included knee pain and physical function measured by an 11-point Numeric Rating Scale (NRS) and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC OA Index), respectively. Six potential mediator variables assessed at baseline and 6 months comprised attitudes toward self-management, fear of movement, self-efficacy for arthritis symptom management, self-reported body weight, physical activity levels, and willingness for knee surgery.
This detailed methodology provides a comprehensive understanding of participant recruitment, interventions, and outcome assessments in the context of an RCT investigating the effects of exercise and diet on knee OA. The inclusion of specific measurements for potential mediators ensures a thorough examination of factors influencing the primary outcomes, contributing valuable insights to the field of OA management.
Statistical Analysis
The statistical analyses for this exploratory study were conducted using Stata, version 16.1, based on complete case data, which excluded participants with any missing baseline, outcome, or potential mediator data. The decision to use complete case data was justified by the comparability of the characteristics of the complete case and omitted sample.
To assess the treatment effect on potential mediators at 6 months, linear and logistic regression models were employed for continuous and binary mediators, respectively, in each treatment group comparison (Diet + Exercise vs. Control and Exercise vs. Control). The analysis included change in weight for Diet + Exercise vs. Exercise. Results were presented as estimated mean differences and relative risks.
Causal mediation analyses were then performed to explore if mediation was present, with regression models estimating direct and indirect effects for each outcome (change in knee pain and physical function) and each potential mediator. The total effect was decomposed into direct and indirect effects, where the latter was calculated as the product of pathways A and C. Assumptions for causal mediation analysis were outlined, including the random allocation of treatment and the assumption of missing data being completely at random. Covariates such as history of knee surgery, baseline mediator and outcome scores, and the stratification variable were included in the regression models.
The “paramed” function and Stata’s “medeff” function were utilized for causal mediation analysis, providing estimates of the proportion of the total effect mediated through the potential mediator, presented as a percentage with a 95% confidence interval. The final step involved estimating the association between each potential mediator and change in symptoms using linear regression models, incorporating relevant covariates.
These rigorous statistical analyses aimed to elucidate the causal relationships and mediating effects within the study parameters, ensuring a comprehensive understanding of the data.
Results
In this secondary analysis involving 415 participants in a randomized controlled trial for knee osteoarthritis, 345 participants with complete case data were analyzed to explore the mediators of two 6-month telehealth-delivered exercise programs, with and without a weight-loss diet, compared to an information-only control group.
The baseline characteristics were similar between groups, with the control group having a higher proportion of female participants.
The analysis assessed the effects of the Exercise program and Diet + Exercise program on potential mediators, such as fear of movement, self-efficacy, weight, physical activity, and willingness for knee surgery, compared to the Control program. Both exercise programs led to improvements in various mediators, with the Diet + Exercise program resulting in additional improvements in attitudes toward self-management and a reduction in weight compared to the Exercise program.
Causal mediation analyses were conducted to explore the mediating effects of potential mediators on outcomes, including knee pain and physical function. For the Exercise program versus Control, an increase in self-efficacy partially mediated the reduction in knee pain, while changes in fear of movement and self-efficacy partially mediated the improvement in physical function.
In the comparison of Diet + Exercise versus Control, an increase in self-efficacy and a reduction in weight partially mediated the reduction in knee pain, while changes in fear of movement, self-efficacy, and weight partially mediated the improvement in physical function. The mediation analysis for Diet + Exercise versus Exercise revealed that changes in weight partially mediated the effects on both knee pain and physical function.
Overall, the findings suggest that improvements in self-efficacy, fear of movement, and weight contributed to the positive outcomes in knee osteoarthritis for participants in the exercise programs, with the addition of weight loss further enhancing the effects. These results provide valuable insights into the mechanisms underlying the benefits of telehealth-delivered exercise programs for knee osteoarthritis management.
Conclusion
This study investigates the mediators influencing the impact of two telehealth-delivered exercise programs, one including a weight-loss diet and the other without, on pain and physical function in individuals with knee osteoarthritis (OA). The findings suggest that reduced fear of movement and increased arthritis self-efficacy play roles in mediating the effects of exercise and diet-exercise combinations on pain and physical function. Additionally, weight loss emerges as a mediator for the effects of diet and exercise on these outcomes, compared to exercise alone and information-only approaches.
The study expands on limited existing research in this area, with few previous studies exploring mediators of exercise and diet programs for OA. Notably, this research contrasts with prior findings on self-efficacy as a mediator, possibly due to different measures employed. Regarding combined diet and exercise interventions, the results align with previous studies indicating that self-efficacy and weight loss mediate effects on physical function and pain.
Notably, fear of movement emerges as a novel mediator, indicating a need for further research to validate this observation. However, attitudes toward self-management, physical activity, and willingness for knee surgery did not exhibit evidence of mediation for either diet and exercise or exercise alone, suggesting a potential lack of contribution to symptom improvement.
Implications for future exercise and diet programs for knee OA highlight the importance of incorporating elements targeting self-efficacy and fear of movement. Behavioral and psychological interventions focusing on these aspects could enhance program effectiveness. Weight loss, found to mediate positive outcomes, suggests that future interventions may benefit from strategies facilitating weight reduction alongside behavior change support.
While the study provides valuable insights, limitations include the absence of certain mediators like joint loading and muscle strength, potential interactions between mediators, and the assumption of data missing completely at random. Generalizability is limited to those with private health insurance in Australia and may not extend to programs lacking similar behavior change support. Overall, the study underscores the multifaceted mediation involved in telehealth-delivered exercise programs for knee OA, emphasizing the nuanced interplay between self-efficacy, fear of movement, and weight loss in achieving positive outcomes.
Oncology Related Tools
Other
Latest Research
