For objective 1, we have obtained data access for the register data and produced harmonized definitions. We have developed a common protocol and distributed network approach that allow us to perform cross-country pooling and comparisons. We have finalized or are working on several multi-site/country projects.
For objective 2, we have obtained data access for the register data and produced harmonized definitions for ADHD medication and the relevant outcomes (i.e. a common code book). We have developed a common protocol and distributed network approach that allow us to perform cross-country pooling and comparisons. We are working on several projects and have protocols ready for several cross-country studies. We have acquired real-world data using novel Remote Measurement Technology (RMT) including both active (smartphone active app) and passive (smartphone passive app and a wearable device) monitoring to capture real-time disease processes that cannot be captured by register-based studies, including ongoing real-world data regarding detailed aspects of ADHD medication treatment, physical activity and cardiometabolic risks, over a 12-month monitoring period. Baseline assessments have been completed following the conclusion of participant recruitment (304 participants in total). We have published the protocol.
For objective 3, we have estimated the expected treatment length of each medication dispensation using a validated machine-learning algorithm, taking advantage of the free-text prescription and package information on the prescription. We have developed protocol and analysis plans for two cross-country studies: i) clinical modifiers of ADHD treatment discontinuity and ii) ADHD treatment discontinuity in adults with co-occurring cardiometabolic disease. We have also finalized our genomic project (WP4) to advance the understanding of the biological mechanisms underpinning ADHD treatment discontinuation. As described in objective 2, we have collected data that will allow us to identify predictors and correlates of adult ADHD medication treatment discontinuity with a focus on day-to-day reported side-effects, using RMT ADHD Remote Technology (ART).
For objective 4, we have developed the relevant machine learning and deep learning models and shared codes in public repositories. We are now applying these models in large scale real-world datasets, optimizing the model performance in these datasets and evaluating the generalizability to reserved unseen data, or datasets from different cohorts. Our goal is to develop generalizable risk stratification tools that will be especially useful in primary care settings where ADHD itself, well as its comorbidity with cardiometabolic disease is typically under-identified in adults.
For objective 5, we have formed an Ethics and Data Management board as well as a Data Management Advisory Committee. We also oversee adherence to all existing ethical and safety provisions and we have also developed and updated Data Management Plan (DMP). Furthermore, we have overseen all ethics approval documents for all sites involved in newly recruiting patients. In addition, we have also overseen the collection of all approvals from the existing cohorts that are used in TIMESPAN. We have also made great progress in overseeing, supervising and implementing our plan for pre-registration, and the common protocols and distributed network approach.
For objective 6, many dissemination actions have been carried out towards scientists, the medical community (e.g. research articles, lectures at congresses, videos, infographics), patients and general audience (e.g. website, social media activities, infographics). We contributed to cross-project collaboration together with PRIME.
For objective 7, the TIMESPAN mentor mentee program, established in the previous reporting period, is well received and has identified and paired more Early Career Researchers with a TIMESPAN senior scientist as mentor during RP2. The ECRs developed personal portfolios specifying their personal training objectives and needs. We organized masterclasses for ECRs during our annual meetings. We created a training course in pharmacoepidemiological analyses using real-world data. We contributed to cross-project collaboration through active participation in the ECNP Course on Brain and Body.
