A Pragmatic Approach to Early Quantitative Evaluation of an Integrated Neighbourhood Model
No Thumbnail Available
Authors
Emma Wells
Contact
Check for full-text access
Issue Date
06-May-26
Type
Conference Abstract
Language
Keywords
Neighbourhood health & place-based working , Rapid evaluation
Alternative Title
Abstract
Recent NHS reforms have prioritised neighbourhood‑scale integration as a route to improving access, continuity and prevention, with integrated neighbourhood teams central to this shift. From September 2024, Camden's East Integrated Neighbourhood Team (EINT) piloted new interdisciplinary, place based ways of working in a test and learn phase. Alongside a qualitative evaluation, system leaders wanted early quantitative insight into how the EINT was shaping system functioning. As population level outcomes change slowly, EINT impacts were unlikely to be measurable at this stage, creating a need for a pragmatic early phase monitoring approach.
To design a simple, quantitative way of spotting where short-term “signal” indicators might have shifted sooner than resident outcomes, and to help system partners notice early positive, negative or unintended effects of the EINT model.
We developed a set of signals covering system flow, workforce pressures and resident experience. Most signals were positioned close to the workforce and point of intervention, on the premise that changes in how staff work are likely to move before downstream outcomes for residents. We included indicators with and without an expected direction of change to capture potential improvements and unintended consequences. Statistical Process Control (SPC) charts were used to distinguish normal variation in signals from meaningful change, and trends in the East neighbourhood were compared with the four other Camden neighbourhoods.
Across 18 indicators, no distinct or statistically meaningful shifts were observed in the East. Where change occurred, it generally reflected borough-wide patterns.
The absence of early signal change does not necessarily mean the model is having no effect: early monitoring in complex system change is exploratory and helps indicate where deeper qualitative inquiry is needed. The approach provided a practical way for neighbourhood partners to track emerging implementation effects and any unintended consequences. Overall, this was a useful interim approach while longer-term outcomes mature, at which point a robust evaluation using causal methods would be more appropriate.
Challenges included the breadth of signals and the difficulty of selecting indicators from an outcome framework developed after EINT activity was underway, which did not always reflect how the intervention worked in practice or how staff understood its aims.