Our clients ask, how do we…
- Integrate record level data across organisations?
- Forecast the healthcare needs of our population?
- Simulate the impact of an intervention on performance and quality?
While a vast amount of healthcare data exists, a significant proportion is still on paper and data systems are operating in silos. Consequently, opportunities are being missed to improve efficiency, patient flow and experience.
Human health data is also inevitably complex and messy, meaning tolerances need to be built into the analysis. Nonetheless, rich insights and robust predictions can be generated and there are effective benchmarks organisations can use. We work with clients at all stages of their data and analytics journey to:
- Digitise paper-based data
- Establish data collection and reporting frameworks
- Develop data sets to inform understanding of the status quo
- Locate external benchmarks to inform service and system design
- Apply advanced analytics techniques to deliver new knowledge
- Integrate data from merging or partnering organisations
- Develop a data culture
Common challenges include organising complex data to be useful and comparable, aligning disparate sources of data, establishing analytics skillsets within managerial and clinical teams, and agreeing what needs to be measured going forward.
- Identify Establishing where improved analytics will provide value and to whom
- Data Assessing, extracting, aggregating and structuring data for analysis and improved data sharing
- Analysis Using advanced analytical techniques including machine learning to draw insights out of data and predict future trends
- Visualisation Crystallising analysis in clear charts that provide the “single version of the truth”
- Synthesis Synthesising insights to facilitate understanding of current and future scenarios
- Engage Working with system leaders to drive action and behavioural change based on robust, data-driven evidence
We worked with a sustainability and transformation partnership (STP) to analyse how mental health resources were being consumed. The aim was to expose variation in effectiveness and efficiency and develop a more joined up approach to providing care.
Generating detailed insights depended on in-depth analysis of populations needs, outcomes and complexity. The population in question lived across a wide geographic area with contrasting environments, where demographic characteristics and health needs differed significantly. We created a complexity index tailored to the region’s specific characteristics, which contributed further understanding of what was driving mental health needs, resource consumption and how best to organise care.
Triangulating complexity, spend and outcomes enabled us to identify the differences in practice and value for money, highlighting where areas of similar levels of complexity had markedly different outcomes. We could draw on locally successful models for future ways of working across the region.
Patients were broken out by supercluster and severity to reveal relative resource consumption. Costs of care for each group were estimated based on the calculation of an average spend per type of patient. Quality of services was assessed using a bespoke clinical standards survey, co-designed with local clinicians. This highlighted key areas where services were not meeting current standards and guidelines.
By bringing together the data on mental health in a structured fashion, we tested whether mental health was underfunded, shifting the focus to improving care. By comparing to peer CCGs with similar levels of complexity, our client concluded they should protect the growth in allocation earmarked for mental health.
Our analysis identified steps that are now being taken to reduce variability, increase reliability and thereby improve the efficacy of spend by £6.4m-8.4m, equivalent to 4-5% of mental health spend, over and above 2% provider efficiency.