Localising clusters of pain contributors in chronic back pain with machine learning techniques
Principal Investigator: Professor Daniel Belavy
Approved Research ID: 55843
Approval date: December 6th 2019
Globally, back pain (BP) is the leading cause of years lived with disability. Chronic BP (CBP; lasting for more than three months) contributes to majority of the costs and disability of the condition. Up to 90% of CBP cases are non-specific (no clear cause), meaning there is a need to better understand the underlying causes and improve our ability to screen which patients will benefit the most from which kind of treatment. Brain changes, mental illness, social support, physical factors and genetics have been associated with CBP, however the contributions of each is poorly understood. It may be possible that there are varying contributing factors in different individuals with CBP that sub-groups exist. These sub-groups may benefit from more targeted treatment types. Therefore, the aims of this research are to determine the main contributing factors and attempt a sub-group approach using 'big data' analysis techniques for individuals with CBP. This research project is expected to take up to four years. This project will have a substantial public health impact as current recommendations for managing CBP treat 90% of sufferers as though they have the same underlying causes. The novel approach of identifying the relative contribution of potential factors and sub-groups in CBP using a data science approach will enable the field to move towards a targeted and individualised management stream. With refined screening procedures and tailored intervention pathways for CBP, this may also assist with reducing disability and health care costs while improving quality of life for these individuals. Given the limited efficacy of current generalist approaches, there is an urgent need for this kind of paradigm change.