Project Abstract Chronic nonbacterial osteomyelitis (CNO), also known as chronic recurrent multifocal osteomyelitis (CRMO), is a severe autoinflammatory bone disease of unknown cause that predominantly affects children and young adults with debilitating complications and poor quality of life. The incidence rate has risen with increasing awareness of the disease. First-line treatment of CNO is non-steroidal anti-inflammatory drugs (NSAIDs), although optimal duration of treatment has not been studied and long-term treatment with NSAIDs has the potential of serious side effects. There is a need to assess the impact of prolonged treatment on the disease. Patients who fail NSAIDs or who have high-risk skeletal site involvement most often require second-line treatment with 1) conventional synthetic disease modifying anti-rheumatic drugs (csDMARD), 2) tumor necrosis factor inhibitors (TNFi), or 3) bisphosphonates. Data from the prospective CHronic nonbacterial Osteomyelitis International Registry (CHOIR) and a survey of practicing pediatric rheumatologists showed widespread use of these three second-line therapies, yet comparative effectiveness of second-line therapies has not been formally evaluated. CHOIR has already enrolled 364 patients, including 297 patients from eight U.S. sites, which provides a comprehensive data collection and analysis platform for the comparative effectiveness research in this proposal. Several disease assessment scores have been proposed including a composite clinical disease activity score (CDAS) that was validated. Radiographic Activity Index (RAI) based on whole body MRI has been standardized and will be used as a secondary outcome. Using the CNO CDAS and CHOIR data we propose to: 1) determine the impact of prolonged treatment of NSAIDs on reducing flare rate, as defined by a new onset of clinical symptoms after temporary remission and after discontinuation of NSAIDs in children who are responsive to NSAIDs; 2) compare the effectiveness and safety of TNFi and bisphosphonates in patients who fail NSAIDs or have high-risk skeletal sites. We hypothesize that 1) prolonged treatment of NSAIDs is associated with reduced disease flare and, 2) bisphosphonates are non- inferior to TNFi (±csDMARD). Patients with active CNO will be prospectively enrolled into CHOIR. Detailed clinical, laboratory, and imaging data will be collected as standard care at baseline and follow up visits. Results from the proposed studies will determine the optimal duration of treatment with NSAIDs as well as the effectiveness and safety of current second-line treatments for patients with CNO.