Every intelligence tool in the Assistiv Cloud suite has a dedicated methodology page: the rationale, the data and proxies with justification, the method, the signal weights where they apply, and, stated plainly, the weaknesses. A score that cannot be interrogated should not move public money.
For the cross-cutting methodology, the FEP composite, normalisation, the economic model and overall limitations, see the master methodology. The pages below go deeper on each individual tool.
Every tool below is built the same way: open public data is turned into proxies for frailty (prescribing, deprivation, isolation, access), those proxies are combined as a weighted sum, and the result is ranked relative to the other Kent and Medway districts. This is a deliberate, defensible design, but it means two things a reader should hold in mind on every page. First, the uncertainty compounds: each proxy is imperfect, and stacking several into one composite can move a ranking more than any single "stated limitation" box suggests in isolation. Second, the engine is not yet validated against observed outcomes, so the figures it produces are modelled, not measured. The "Live" badge means the data feed updates automatically; it does not mean a tool has been formally validated. Closing that gap, by scoring each month of predictions against what actually happens when lagged NHS outcomes publish, is the active priority, and these pages will say so plainly until it is done.
Live live data feed, not yet outcome-validated · Prototype working prototype, not yet validated · Simulation illustrative or simulated data
Layer 2
LiveLayer 2 · scoring engine
LiveLayer 2 · LSOA hotspots
LiveLayer 2 · dual intelligence
LiveLayer 2 → 3
LiveLayer 3
LiveLayer 2 · system pressure
Live2026-27 planning
PrototypeLayer 4 · screening
LiveFalls intelligence · SECAmb FOI
LiveLayer 2 · informal care