Remote measurement technology (RMT) has transformed longitudinal remote monitoring of research participants and offers unique translational potential. This project uses data from our large RMT study on adults and adolescents with attention deficit hyperactivity disorder (ADHD) that involves active and passive remote monitoring over a period of 12 months.
Aim 1: To use RMT to identify longitudinal patterns in physical activity and other healthy lifestyle behaviours in people with ADHD. We will investigate the degree of stability or fluctuations in such behaviours over a period of 12 months, and whether sub-groups can be identified within the individuals with ADHD.
Aim 2: To identify predictors of increased physical activity and other healthy lifestyle behaviours in people with ADHD. We will explore a wide range of different potential predictors, including co-occurring clinical symptoms (e.g. autism, aggressiveness, irritability), socio-demographic factors, age, gender, cognitive abilities, physiological measures and engagement with clinical services.
Aim 3: To use machine learning on the remote monitoring data to build real-world ADHD prediction models that further assess complex temporal relationships over time between a range of measures. These models take findings from the initial analyses (aims 1-2) forward and incorporate the possibility that some factors can be both predictors of certain outcomes while also representing outcomes of other predictors. For example, lifestyle factors such as physical activity may predict changes in clinical symptoms, while also representing consequences of other predictors, such as excessive online activity.
Training on all aspects of the project, including advanced analyses, will be provided.