The broad objective guiding our research is to conduct analyses that will provide critical insights to clinicians and decision makers to optimize knee OA pain management while reducing opioid use in persons with OA and major comorbidities including mental health disorders and morbid obesity. Symptomatic knee osteoarthritis (OA) affects over 14 million Americans and accounts for $27 billion/year in healthcare expenditures. Traditional views of knee OA pain as nociceptive have been challenged by evolving evidence that nervous system alterations often result in sensitization and neuropathic-like symptoms. Many OA patients have comorbidities including depression, which leads to worse knee pain and complicates pain management, especially regarding opioid use. Medications are only modestly efficacious, in part because they are not tailored to pain mechanisms. Obesity, especially morbid obesity, further complicates both pharmacologic and surgical OA pain management. The diversity of pain mechanisms and frequency of comorbidities have begun to reframe knee OA as a syndrome comprised of multiple phenotypes, wherein a single treatment strategy does not fit all. The challenges in OA pain management posed by multiple pain phenotypes and comorbidities, coupled with the threat of the opioid epidemic, are further exacerbated by the gap between `what we know' and `what we do.' While the efficacy of pain phenotype-based pharmacotherapy, weight management and exercise have been established in RCTs these approaches have not translated to routine care. PA is essential to managing OA pain, yet most OA patients are inactive. Strong evidence suggests that exercise and physical activity (PA) are as effective as analgesic medications, but implementation of exercise programs is hindered by lack of infrastructure and funds. Comorbidities augment the risk of physical inactivity, despite strong evidence that PA is efficacious in OA patients with comorbidities. Evidence of efficacy is not sufficient to facilitate the implementation of these programs into clinical practice. Implementation requires investment; and knowledge of the cost-effectiveness and budgetary impact of these programs will help to translate research findings into day-to-day clinical management. Decision analysis is an important methodology that helps to evaluate the value of programs that have been shown to be efficacious. We propose to use a validated computer simulation model of knee OA (OAPol) to narrow the gap between evidence and practice by assessing the value of three major therapeutic strategies in managing pain in knee OA patients with comorbidities: 1) tailored pain management according to pain phenotypes to optimize pharmacologic regimens; 2) weight management in morbidly obese persons to improve outcomes of OA-focused treatments; and 3) PA programs as non-pharmacologic pain reduction regimens.

Symptomatic knee osteoarthritis (OA) is prevalent, disabling, and costly and is worsened by concomitant comorbidities that lead to greater pain and functional limitations. There are several unresolved questions regarding the incorporation of pain phenotyping, weight management strategies (both non-surgical and surgical), and physical activity programs into the care of those with OA and additional comorbidities. This project combines the analysis of large datasets with a computer simulation model of the lifetime clinical and economic consequences of knee OA, and it will establish the value of treatments strategies for knee OA pain in persons with comorbidities.