The 10-Year Plan is clear. ICBs are to commission outcomes, not activity. NNHIP is the delivery wedge. The measurement architecture is still built for activity. The gap between what is asked and what the system can produce is the editorial space that needs filling.

Read the 10-Year Plan in one frame and the ask is straightforward. ICBs are to move from commissioning activity (referrals, appointments, contacts) to commissioning outcomes (does the patient feel better, does the cohort move on a measurable health gradient, does the borough close an inequality gap). NNHIP Wave 1 is the structured experiment that proves this works at neighbourhood scale. The 2032 trajectory commitments are the audit horizon. £10M of Wave 1 money plus a £45bn neighbourhood-health architecture follows if the proof comes through.

The Plan is reasonable. The architecture to produce that proof is mostly absent.

What outcome commissioning needs that activity commissioning does not

Activity commissioning runs on data the NHS already has. Hospital Episode Statistics (HES). Secondary Uses Service (SUS). General practice extract from QOF. Mental health services data set (MHSDS). These count activity. They count it well. They have been tuned for two decades.

Outcome commissioning needs something different. Three things, ideally all three.

First, a patient-reported baseline and follow-up. EQ-5D-5L is the established universal anchor. It is in NICE technology appraisals. It is in trial inclusion criteria. It is short enough to capture in two minutes on a phone. It is sensitive enough to register meaningful change across populations. It is the closest thing the NHS has to a lingua franca for health-related quality of life.

Second, a domain depth measure that picks up what is specific to the condition or the cohort. For stroke, Stroke Impact Scale 3.0. For long-term-condition self-management, Patient Activation Measure (PAM). For mental wellbeing, WEMWBS. Each is well validated. Each has a published interpretation guide. Each gives the cohort-specific signal that EQ-5D-5L alone cannot.

Third, a credible signal-trust layer. Two named clinicians who endorse the methodology and the cohort. Reciprocal credit. The Editorial Validator model.

Why the system does not yet do this

Three structural reasons.

One. The data warehouses that hold HES and SUS are not built for patient-reported data. PROM data sits in fragmented silos. The Hip and Knee replacement PROM has its own pipeline. The IAPT-now-NHS-Talking-Therapies pipeline has its own outcome capture. The 10-Year Plan ask for outcome-driven commissioning at neighbourhood scale requires PROM capture at neighbourhood scale, which the data architecture does not currently support.

Two. The instrument selection is contested. Every condition area has a preferred measure. Every researcher has a preferred measure. Every commissioner has a preferred measure. Without a small canonical set, every neighbourhood produces incomparable evidence. The 10-Year Plan ask is for comparable evidence across boroughs and ICBs.

Three. The trust layer is missing. A new evidence pipeline that nobody senior has endorsed will not be cited by an ICB board even if the methodology is impeccable. Without an Editorial Validator equivalent, outcomes data is read as researcher noise, not commissioner signal.

One credible proposal currently on the field

RiDraw's PROM/PRIM v0.1 is one of the proposals. It is not the only credible proposal. It is one of the few that addresses all three structural reasons simultaneously.

The shape of PROM/PRIM v0.1.

Universal anchor. EQ-5D-5L on patient device. Apple HealthKit. Android Health Connect. The capture happens on the phone the patient already carries. Aggregation is at cohort level. No PII leaves the device.

Domain depth. One additional measure chosen per sector. For NNHIP neighbourhood-health cohorts, the depth measure is PAM. For neurorehab cohorts (the Chiltern Neurocentre is the first paid example), it is Stroke Impact Scale 3.0. For mental health, WEMWBS. The depth measure is captured alongside EQ-5D-5L. Two-minute total burden.

Sovereignty by absence. Because no individual record leaves the device, the data-protection question changes shape. The cohort aggregate is reported. The individual record is not transmitted. The patient retains the data layer the NHS does not currently hold.

Editorial Validators. Two named doctors per published outcome cycle. Reciprocal credit between the validator and RiDraw. The publication carries the names. The names carry the credibility.

It is one of the few outcome-architecture proposals on the field that ships now, with no new data warehouse, no new central infrastructure, and no patient-identifying data leaving the device.

What this asks of the reader

If you sit at an ICB, the ask is: read PROM/PRIM v0.1 as a candidate methodology for one neighbourhood cohort. The minimum viable commission is a 4-week capture cycle on 8 to 12 patients in one INT footprint. The bound A5 outcome book that follows is the first artefact the ICB can cite at board level.

If you sit at NHS England in the NNHIP programme team, the ask is broader. The 2032 trajectory commitments need outcome architecture, not just activity reporting. PROM/PRIM v0.1 is one of the proposals. RiDraw is happy to publish the comparable cross-pioneer cumulative evidence pattern across the 43 sites if a cohort of three to five ICBs commission cross-pioneer field readings; commercial terms are agreed off-publication.

If you sit at DHSC, the ask is structural. The outcome-commissioning ambition in the 10-Year Plan is correct. The architecture to produce that proof needs investment, not promotion. PROM/PRIM-equivalent instruments need air cover from the centre. Editorial Validator-equivalent trust mechanisms need air cover from the centre. Without these two, every ICB will build its own and the evidence will be incomparable across the country in 2032.

The window is also eighteen months

This is the same window as Why now No. 1. The 2032 trajectory commitments harden over the next eighteen months. After that, ICBs are accountable to an audit chain that does not yet have the right inputs. The instruments that get adopted in the next eighteen months are the instruments that NHS England will read for the next decade.

If PROM/PRIM or an equivalent is adopted in this window, outcome commissioning becomes the operating model. If nothing is adopted, ICBs fall back to activity reporting and the 2032 commitments are a paper trajectory.

Proof discipline