Weight-loss and exercise for communities with arthritis in North Carolina (we-can): design and rationale of a pragmatic, assessor-blinded, randomized controlled trial

Worldwide, only half of the population seeking osteoarthritis (OA) health management receive appropriate care. This is due, in part, to limited prescription of non-pharmacologic interventions, the rise of pharmacologic prescriptions for symptom management, low patient health literacy, and barriers impeding lifestyle modifications [1, 2]. Knee OA is the most common and persistent cause of mobility dependency and disability; its prevalence is estimated at over 250 million, or 3.6% of the world’s population [3, 4]. OA is chronic – patients live with the symptoms for an average of 26 years [5]. With no effective disease-modifying interventions, treatment focuses on pain relief, but the safety concerns associated with many pain medications highlight the need for safe, effective non-pharmacologic interventions. Clinical guidelines strongly encourage the use of non-pharmacologic exercise and diet to relieve pain and improve function [6, 7]. Unfortunately, only 25% of the patients requiring advice concerning these treatments (BMI ≥ 27 kg/m² with diagnosed knee OA) receive the necessary care [8]. Moreover, these interventions, proven effective under highly controlled conditions, have yet to show long-term improvements in community settings [9‐11].

The association between obesity and knee OA was first documented in 1945 and has been widely verified [12‐15]. Two in three obese people may develop symptomatic knee OA in their lifetime [16]. Obesity is the most modifiable risk factor for knee OA, and weight loss is a safe, effective non-pharmacologic intervention to improve clinical outcomes [9, 10]. Recently, our Intensive Diet and Exercise for Arthritis (IDEA) trial compared a diet-induced weight-loss (D) of 10% of baseline weight and exercise (E) interventions, separately and in combination (D + E), across an 18-month period in 454 overweight and obese older adults with knee OA. The primary mechanistic outcomes, knee joint loads and inflammation, were reduced in D and D + E compared to E [10]. An intent-to-treat analysis also revealed that while D and E successfully reduced pain after 18 months (D: 25%, E: 28%), the combination of D + E was twice as effective with a 51% decrease from baseline. The D + E group was also superior to the E group on other clinical measures including physical function and walking speed. IDEA, with its 10% weight loss, produced superior results to our previous trial (ADAPT) of exercise and 5% weight loss that resulted in a 30% reduction in pain [9]. Taken together, these trials provide strong support for our model (Fig. 1) indicating that weight loss plus exercise improves clinical outcomes by affecting both the biomechanical and inflammatory OA disease pathways, decreasing knee joint loads and pro-inflammatory cytokine activity.

IDEA was an efficacy study implemented in a university setting under rigorously monitored conditions that included highly educated and well trained staff, and excellent, well-maintained facilities. Clinicians have expressed concern about the lack of practical means to implement and sustain a similar program in a community-based environment. Thus, determining whether we can adapt this successful program to a pervasive public health problem in real-world clinical and community settings is a clinically important question that has public policy implications.

Approximately 25% of the United States population lives in rural communities [17‐19] with most having fewer services and resources than urban communities [20]. Rural dwellers report poorer health-related quality of life [21], reflecting higher prevalence of many disorders, including knee OA. Delivering proven health interventions to communities with limited healthcare access is a national public health priority [22]. Weight-loss and Exercise for Communities with Arthritis in North Carolina (we-can) is the first long-term trial of diet-induced weight loss and exercise in older adults with knee OA implemented under pragmatic conditions in both rural and urban communities. Building on the results of IDEA, the intent of the we-can trial is to translate this highly beneficial, long-term intervention for a major chronic health condition to diverse community settings.

Few trials are purely pragmatic; most fall on a continuum between pragmatism and efficacy [23]. The pragmatic components of we-can include: (1) large sample size (N = 820); (2) broad inclusion criteria; (3) patient-centered outcomes; (4) conducted in established community facilities rather than a referral or university center; (5) commonly available community-based staff to deliver the standardized treatment; (6) non-physicians trained by our physicians to perform a knee exam based on American College of Rheumatology (ACR) clinical criteria; and (7) diverse means of communication with participants (e.g., phone, email, study website, Facebook, text messages). Our experience with ADAPT [9] and IDEA [10] indicates participants need regular interaction with study staff to achieve significant weight loss. Hence, the primary efficacy component of our design is the use of diverse methods of contact by staff who are not physicians but trained community interventionists to maintain participant interaction. The pragmatic nature of many aspects of this trial combined with cost effective methods of staff-patient communication is more likely to improve health care that can be generalized and applied in varied community settings [24].

Exercise and weight loss have level 1 evidence of efficacy [6] for pain relief and the combination is most efficacious [9, 10], yet health practitioners often do not refer patients to community programs, either because they don’t exist or because non-pharmacologic alternatives are not part of the treatment plan [2]. Intensive weight loss combined with exercise reduces abnormal stress by decreasing knee joint loads and reduces abnormal physiology by lowering inflammation [10]. This results in less pain and less disability. Losing 10% of baseline body weight is necessary for a moderate to large clinical effect. We found a dose response to weight change with pain and function, independent of group assignment; IDEA participants who lost more than 10% of baseline body weight had greater reductions in pain and improvements in function than participants who lost between 5 and 10% and less than 5% of baseline body weight [10]. Hence, our minimum weight loss goal for the we-can diet and exercise group is 10% of baseline body weight.

Losing weight and maintaining weight loss are both difficult and often unsuccessful. Biological changes fight attempts to lose weight; the body acts in starvation mode increasing feelings of hunger, it suppresses satiety, slows metabolic rate, and attempts to defend higher body weights [51]. We exceeded our mean weight loss goal of 10% in IDEA, in part because participants received regular attention. A similar pragmatic, community based trial of diet and exercise for obese women free of knee OA found that limited patient contact (≤4 h/year) during the 2.5 year intervention period resulted in less than 20% of the D + E group achieving their weight loss goal of ≥ 5 kg or ≥ 5% of baseline body weight (mean weight loss < 1 kg) [52, 53]. Indeed, a possible consequence of a pragmatic approach with little or no participant contact is low adherence. The challenge in we-can is to maintain the regular patient contact utilized in IDEA in a less structured, community-based environment. The use of multiple communication techniques with participants is critical in achieving compliance and retention rates that approach those achieved in IDEA. These include, but are not limited to, phone calls, text messages, email, Facebook, LinkedIn, Skype, Twitter, and YouTube. Participants will select their preferred method of communication.

Maintaining intervention fidelity is critical. Fostering close relationships with the community centers that serve as intervention sites is important in maintaining an open line of communication. Site interventionists maintain personal relationships with participants both face to face and electronically. The Coordinating Center staff monitors the quality of the intervention being delivered with periodic face-to-face observations, biweekly Adherence and Retention Committee meetings, and monthly Nutrition and Exercise Committee meetings. Including underserved rural communities such as Haywood and Johnston counties in knee OA clinical trials is unique; however, recruitment and retention may be difficult as communication and transportation concerns are complicated by lower socioeconomic conditions and a less densely populated recruitment base. The recruitment goals for Johnston and Haywood counties have been adjusted to reflect these anticipated impediments.

To meet the chronic disease burden the Centers for Disease Control and Prevention launched several strategic initiatives of which one is to establish community-based programs to support healthy behaviors [22]. Indeed, the physician’s dilemma is the lack of practical means to implement and sustain a diet and exercise program in a community-based environment. we-can is designed to test the effectiveness of a community program that will serve as a blueprint and exemplar for clinicians and public health officials in urban and rural communities to implement a weight loss and exercise program designed to reduce knee pain and improve other clinical outcomes in overweight and obese adults with knee OA. We envision a national implementation of this systematic program that will serve as a model for healthcare professionals on implementing a platform that is accessible to consumers and clinicians, and is of value to insurers because it can be sustained long-term and at a reasonable cost.