|Title||Health Data and Methods for Evidence based Decision Making|
There is an increase in the volume and types of data across health to support healthcare decisions. An increasing number of clinicians and allied healthcare scientists are looking to integrate data to generate hypotheses and to support clinical and operational decisions. A different set of methods and technologies is used each to time to be able to address the identified health problem. The role of the Clinical Scientist in HSST is to be at the forefront of this and to support decision making for healthcare delivery by deriving understanding and significance from data through clear analysis and interpretation of results (to a range of stakeholders) as well as being able to apply tools for integrated working and data visualisation.
By the end of this module the Clinical Scientists in HSST will be able to analyse, synthesis and apply their knowledge and understanding of health data analysis to inform clinical and operational decision making. They will be able to integrate public domain and patient data to support a range of clinical decisions, as well as creating visualisations and responding to feedback from the intended audience. In addition, qualitative research methods that determine how and why the decision was made will be covered. The Clinical Scientist in HSST will also be expected to consistently demonstrate the attitudes and behaviours necessary for the role of CCS.
Knowledge and understanding
By the end of this module the Clinical Scientist in HSST will be able to analyse, synthesise, evaluate and critically apply their expert knowledge of health data and methods to support evidence-based decision making, including:
- The available data sets (health data and data captured for non-health purposes but impacting on health) and discuss their advantages and disadvantages for use in evidence- based decision making.
- How data sets are created and are used to support planning, commissioning, research and direct care.
- The structure of health data sets and how they are accessed.
- How to process health data for use in evidence-based decision making (e.g. data pre- processing, including cleaning, visualisation and integrity checks).
- The methods for data integration and the advantages and limitations of different strategies for data integration.
- The steps of how to convert data and a health problem into an appropriate modelling strategy.
- The steps required to analyse health data.
- Statistical/mathematical methods to analyse health data (e.g. supervised learning methods including goodness of fit, risk/loss functions (correlation, Root Mean Square Error, accuracy sensitivity, specificity); classification (e.g. including neural networks, decision trees) and regression, data mining and unsupervised learning (e.g. Principle Component Analysis, hierarchical clustering, partitional clustering).
- Qualitative research methods for data collection and analysis in the context of clinical and operational evidence-based decision making.
- The range of qualitative research methods available to understand and evaluate the decision-making process.
- Data analysis (e.g. R, Stata) and visualisation software (e.g. tableau) and how to use them effectively.
- Data presentations available and how to select which is the most appropriate for a specified audience.
- How to devise, evaluate and use workflows to enable appropriate metadata capture from domain experts in a clinical setting.
- The risks and opportunities of selectively creating new data sets from collating/merging/extracting from existing and new data sources.
Technical and clinical skills
By the end of this module the Clinical Scientist in HSST will be able to critically apply their knowledge and understanding of health data and methods for evidence-based decision making, critically reflecting on their performance, and apply in practice a range of technical and clinical skills, and will be able to:
- Perform analysis in a statistical software package (e.g. R, Stata) and interpret results to support evidence-based clinical decision making .
- Advise on how research and development protocols can be translated into practice for different clinical scenarios.
- Create effective infographics of a range of different data sources to show to different audiences, including the local trust board/directorate and the public.
- Elicit requirements from various stakeholders, use them to create a visualisation and respond to feedback.
- Use graphing tools, visual exploratory analysis tools and interactive tools to support decision making and an understanding of the underpinning science, in particular by the patients and the public.
- Critically appraise health data/information visualisations in literature and in other sources, and suggest improvements.
- Generate ethical committee permission applications for data use, including expression of evidence of informed consent.
Attitudes and behaviours
This module has no attitude and behaviours information.