Identifying generative AI use among genuine responders in online survey research.

Chronic pain affects approximately 1 in 5 adults worldwide, with substantial health care costs attributable to its persistence, functional impairment, and biopsychosocial complexity. Despite advances in understanding nociceptive, nociplastic, neuropathic, and central sensitization mechanisms, estimating long-term clinical evolution remains challenging. Mobile health applications offer a unique opportunity to capture high-resolution longitudinal data on symptoms, functional impact, and psychological states in real time, enabling detection of subtle intraindividual fluctuations that traditional clinic-based assessments cannot detect. Previous longitudinal cohort studies using infrequent assessments have been limited by small sample sizes, single-pain focus, and inadequate handling of missing data and attrition bias. Attrition is particularly problematic in mobile health studies, where engagement is voluntary and often declines rapidly. Recent advances in trajectory modeling, such as transformer-based approaches, rely on population-scale cohorts with dense multimodal histories, conditions not typically available in general population mobile health or routine pain-clinic settings. Thus, identifying distinct trajectory subgroups and assessing the value of routinely available baseline clinical data in estimating trajectories are further challenges in chronic pain management that this study aimed to address. This study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology () reporting guidelines for cohort studies. It is a secondary analysis of the eDOL mobile health cohort, a web platform and mobile application for the long-term follow-up of patients with chronic noncancer pain, conducted in French tertiary pain clinics from September 2019 to September 2025. All participants provided written informed consent; the study was approved by the relevant research ethics committee and the French Data Protection Authority and was registered at ClinicalTrials.gov (), in accordance with the Declaration of Helsinki. Eligible participants had at least 6 months of follow-up and baseline data available within the first month (full details are shown in the eMethods in ).