The project's objectives are
- to analyse prospectively collected data from cohort studies linked with health and administrative registries to determine how co-occurrent pain and psychological distress in adolescence and emerging adulthood is associated with academic achievement, income group trajectories and health trajectories
- to identify factors associated with co-occurrent pain and psychological distress by applying machine learning techniques, and develop prognostic models using machine learning for adulthood disability benefit and long-term sickness absence in conjunction with co-occurrent pain and psychological distress
Background
The project addresses several knowledge gaps in the field. It highlights the limited use of machine learning in musculoskeletal health research.
Most studies in this area have been conducted on small samples, underlining the need for larger datasets, such as national registries and population-based cohorts, to leverage machine learning effectively.
The project also underscores the importance of determining when machine learning is preferable to traditional statistical methods, especially in the context of prevention and registry data analysis.
Methods
Methodologically, the project involves data linkage of various sources, including the SHoT study and the Young-HUNT study, as well as health and administrative registries.
Overall, this project aims to advance our understanding of the complex interplay between co-occurrent pain and psychological distress from adolescence to adulthood, utilizing innovative methodologies such as machine learning and data linkage to address critical knowledge gaps in the field of musculoskeletal health research.