.avif)
Agree and track key measures of success
Creating a meaningful measurement framework
While agreeing and tracking prevention measures sounds straightforward, it represents one of the most complex elements of successful implementation. Recent advances in analytics and evaluation techniques have enhanced authorities' ability to predict what would have happened without preventative intervention, strengthening confidence in impact assessment.
Given this complexity, many authorities engage academic or central government evaluation experts to develop robust approaches. Effective measurement frameworks typically address two core elements:
Measuring reduction in need
The principal aim for most prevention programmes is to deliver measurable impact on need and associated expenditure. The fundamental challenge in prevention evaluation is establishing what would have happened without intervention—the "counterfactual."
Local authorities typically consider three main approaches, each with distinct advantages and limitations:
- Approach: Compare cohort service usage post-intervention to pre-intervention patterns
- Advantages: Relatively simple to implement using existing person-level data
- Limitations: Fails to account for age-related increases in service usage and wider contextual factors like economic conditions or service changes
- Approach: Compare outcomes for supported individuals against those who declined intervention but consented to data tracking
- Advantages: Provides a practical comparison group without randomisation
- Limitations: Inherent differences between those who accept or decline support may influence outcomes independently of the intervention itself
- Approach: Randomly assign eligible individuals to receive or not receive intervention
- Advantages: Randomisation helps isolate intervention impact from other trends
- Limitations: Some authorities raise ethical concerns about denying potentially beneficial support based on random selection
More advanced approaches include propensity score matching, which uses statistical techniques to create comparison groups that closely mirror intervention recipients across multiple characteristics, and Cluster Randomised Controlled Trials, which randomise at the level of communities rather than individuals.
Establishing and tracking holistic leading indicators
Since impacts on long-term care demand may take time to materialise, leading indicators are essential to build confidence in the model and enable evidence-based improvements. Comprehensive frameworks typically include:
Advanced authorities use a logic model approach, commonly applied in public health, to show how different measures connect to create impact. This approach maps the relationship between:
- Number of staff in post within the preventative model
- % of staff who have received training in the model
- Number of people proactively outreached to
- % of residents working towards preventative SMART goals
- Attendance at MDT meetings
- Resident feedback on their experience
- Number of residents completing the preventative process
- Drop-out rates at different points of the process
- % of residents with SMART goals met
- Self-reported wellbeing and resilience measures
- Usage of acute services compared to the control group (e.g., social care front door contacts, A&E attendances)
- Long-term care starts or increases compared to the control group
By integrating measures into this structured approach, authorities build increasing confidence in their prevention model through evidence-based links between leading indicators and target outcomes. This evidence then informs business case iterations when considering expansion to new cohorts.
Practical measurement implementation
Authorities implement these measurement frameworks through several practical mechanisms:
- Embedding tracking in case management systems to capture operational measures and intervention progress
- Establishing feedback mechanisms for residents and staff
- Leveraging existing activity and finance datasets for system-level outcomes
- Creating integrated dashboards that bring multiple data sources together to support holistic analysis
While authorities adapt measurement approaches to local context and capacity, partnership working in establishing and tracking measures is essential to ensure evaluation remains holistic and avoids adverse system impacts.
