7. Background and further information
Outcomes for cancer patients in Scotland lag behind those of our Northern European counterparts. Scotland has some of the best health service data in the world. Few other countries have information which combines high quality data, consistency, national coverage and the ability to link data to allow patient based analysis and follow up. This competition asks the question “How can data science be applied to existing NHS Scotland data to improve cancer patient care and outcomes?”
Scotland provides a unique test bed for exploration into innovative technological solutions to accessing large volumes of data pertaining to patients’ demographics as well as their pathways to diagnosis, treatment and long term follow up. The intelligent analysis of this wealth of information could be used to deliver benefit at several levels:
- Clinical benefit to individual patients, informing clinical decision making and prognostication
- Clinical benefit to populations, informing vital information on the efficacy of treatment options
- Resource utilisation, informing service delivery, service planning and service development
- Academic benefit of informing high quality research to inform efficacy of current and future treatments.
- Economic benefit of driving company growth and inward investment as the Scottish ecosystem leads to rapid commercialisation of medical technologies within a global market.
Data already held in primary, secondary and research databases relating to patients’ demographics and pathways to diagnosis, treatment and follow up offers a real opportunity to map closely individual patients. The ability to identify, share and learn from variation in clinical practice will undoubtedly change treatments for patients in the future: as well as contributing significantly to overall clinical care and service planning, as progress towards ‘personalised medicine’ is made, it is vital first to collate variables unique to individual patients and then apply these variables to the experiential learning from similar clinical cohorts. Data provides the key to unlock this process!
Non-clinical data such as administrative data sets such as the Scottish Cancer Registry, prescription data, acute admissions can also offer analytical insights that could lead to service improvements and predictive insights that can improve cancer outcomes.
The competition is looking for solutions that are well designed, simple and accurate within the confines of a modern healthcare system, respecting clinical confidentiality whilst providing a degree of accuracy which has the confidence of clinicians, patients and health service managers alike.
Any solution must integrate with components of the NHS eHealth architecture. Critically, a key goal will be to provide a solution that can be implemented to benefit directly Scottish patients and the wider service.
The goal of this challenge is to develop solutions using existing NHS Scotland data to improve cancer patient care and outcomes in Scotland. Proposed solutions should achieve at least one of the following broad objectives:
- Enable analysis of unstructured data (e.g. clinical notes, medical imaging)
- Enable data driven clinical decisions
- Enable data driven service improvement in the NHS
- Enable data driven recruitment for clinical trials
- Enable the adoption of precision medicine approaches
Innovative approaches to these challenges will be expected to incorporate data science techniques from fields such as:
- Predictive analytics
- Visualisation
- Machine learning
- Natural language processing
- Processing of structured and unstructured data
The ultimate aim of the challenge is to fundamentally change the way data and analytics are used to drive improvement in cancer outcomes in Scotland