How we source PIP statistics.

Where the numbers come from

Every figure on the /pip-statistics/* pages is drawn from one of two official sources:

• DWP Stat-Xplore, the department's live statistics database. PIP Disability is coded at roughly 536 condition categories, updated each quarter.
• GOV.UK Personal Independence Payment statistics, the quarterly release page carrying the official summary tables and the next-release calendar.

A note on the source links. The underlying tables are queried through the Stat-Xplore Open Data API, which needs a free DWP account, so a direct Stat-Xplore link would land a reader on a log in screen. The published statistics collection linked from each figure is openly accessible and is where DWP releases the same underlying data.

Primary vs secondary labels

Each figure is tagged either primary or secondary.

• Primary: taken directly from a DWP or HMCTS release with no derivation on our part.
• Secondary: a value we have calculated from a primary source (for example, a percentage derived from two published totals). The underlying primary source is always linked.

Refresh cadence

DWP publishes PIP statistics quarterly, in January, April, July and October. When a new release lands we update the data file manually against the new tables, refresh the release date, and re-verify the source links. There is no automated pull: this is a citation asset and a bad number here is worse than no number, so a human runs every refresh.

What we deliberately do not publish

DWP does not publish some figures at the per-condition level. We do not extrapolate to fill the gap.

• Per-condition clearance and processing times are not published by DWP. They are not shown here.
• Per-condition tribunal outcomes are not published by HMCTS. Any tribunal figure shown here is all-conditions and labelled as such.
• Any figure without a working DWP or HMCTS URL is omitted. Missing data is missing, not a zero.

Independence and scope

Finally Seen is an independent UK service. It is not affiliated with the DWP, the NHS, HMCTS or any local authority. These pages are general information, not benefits advice. Corrections are welcomed at hi@finallyseen.org.uk and fixed within 48 hours.

The Believability Gap study

Our flagship per-condition study at /pip-statistics/believability-gap uses two award-rate definitions in parallel:

• rateAllTime: Awarded ÷ (Awarded + Disallowed) on new-claim decisions across all months since 2013. Withdrawn claims are excluded.
• rate12m: the same calculation restricted to decisions made in the most recent 12 months. Suppressed (falls back to rateAllTime, marked †) where fewer than 300 decisions were made in the window.
• Editorial classification. Grouping conditions into "objective", "invisible" and "mixed" is our judgement about how the condition typically presents at assessment, not a DWP field. The numbers are DWP's, the grouping is ours, and we list every condition's group on the study page.
• Regional analysis. Stat-Xplore supports splits by nation and Local Authority. A regional layer is now published for chronic fatigue syndrome at /pip-statistics/me-cfs/evidence, covering ten regions across England and Wales on a three year window, with cells below 200 decisions omitted and Scotland excluded because it has migrated to Adult Disability Payment. This is one condition, not the whole 165-condition study. Finer geography per condition is limited by small cell sizes rather than by availability.

The Postcode Lottery study

Our constituency study at /pip-statistics/postcode-lottery covers all 632 Westminster constituencies (2024 boundaries).

• Window. May 2023 to April 2026. Award rates are shown only where at least 500 new-claim decisions were made in the window; low-volume seats are listed without a rate rather than shown with a misleading one.
• Scotland. Scottish constituencies are listed for completeness but shown without award rates or rank, because Scotland has migrated to Adult Disability Payment and DWP no longer takes new PIP claims there. 575 England-and-Wales seats are ranked.
• Electorates. Electorate figures for the per-1,000-electors calculation come from the mySociety 2025 constituency dataset.

Mandatory reconsideration data

Per-condition MR change rates are drawn from DWP Stat-Xplore and cover April 2013 to July 2021. DWP stopped publishing MR outcomes at the per-condition level after July 2021; the department's current national aggregate (28%, May 2021 to April 2026) is used for the all-conditions figure. We label the older per-condition period on every figure so the two are never confused.