Correction note (23 July 2026): this article has been revised. Earlier versions carried figures that did not survive a re-verification of our dataset; every number below is reproduced from the published sources named, and the full method sits on our mental health map page.
Your latest wellbeing survey shows most employees rate their mental health as good, and the leadership team breathes a sigh of relief. But here is what the survey did not capture: a Bradford office sits in an area where 26.1% of adults reported high anxiety in 2022/23, above the England average of 23.3%. A strategy built only on internal survey data is built on half the picture.
We analysed 15,308 rows of public health data across 184 English areas from the government's OHID Fingertips database. In 2022/23, the share of adults reporting high anxiety ranged from 10.4% in Enfield to 33.8% in Southampton, a 3.2 times gap. Eighty-three of the 150 upper tier areas with published 2022/23 figures sit above the England average, and the spread between areas dwarfs the year-to-year movement in any national figure. You can check any area on our interactive mental health map of England.
Why your wellbeing surveys miss the bigger picture
Internal surveys have three familiar blind spots. People who are struggling most are often the least likely to fill them in. Social desirability skews answers, especially in small teams where anonymity feels thin: when your finance team has eight people, everyone recognises everyone. And a survey captures a moment in time inside work, while a large part of what shapes how people are doing sits outside it.
None of that makes surveys worthless. It makes them one instrument rather than the whole dashboard.
What population health data tells you that surveys cannot
Population data describes the places your people go home to. The ONS asks adults across the country how anxious they felt yesterday, and OHID Fingertips publishes the results by local authority precisely so services can be planned around them. The differences are not subtle: Blackpool reported 29.1% of adults with high anxiety in 2022/23 while Wokingham reported 18.2%, in the same country in the same year.
These are population figures for residents of a place, not scores for any employer's workforce, and they predict nothing about an individual. What they describe is context: the baseline conditions around a workforce, which differ far more from place to place than most benefits decisions assume.
The compliance and duty of care implications
Employers must assess workplace risks, and a reasonable risk assessment takes account of context that is publicly known. If many of your people live in areas with well-documented higher anxiety rates and your mental health provision has never considered that, it is hard to argue the assessment was thorough. This is not about liability scares; it is about being able to show your provision was designed against evidence rather than habit.
How to use population data alongside internal insights
Map where your people live against the published area baselines, at area level, never as individual surveillance. Then use the baseline to interpret your survey. If the published context suggests elevated anxiety but your survey reads calm, ask why: are people comfortable disclosing, and is the survey asking the right questions? A quiet survey in a high-baseline area is a prompt for curiosity, not celebration.
Building a complete wellbeing intelligence system
The strongest approach reads several sources together: population baselines for context, an anonymous regular check-in for how people say they are doing, absence patterns for impact, and benefits usage for what support people actually reach. Each covers a blind spot of the others.
The geography alone can be striking. Wakefield reported 30.3% of adults with high anxiety in 2022/23; Leeds, in the same commuting belt, reported 20.1%. Two workforces drawing on the same labour market live in measurably different contexts, and identical provision will land differently in each.
Your next steps
Keep your survey; it carries individual voice. Add the baseline: look up the areas your people live in on our interactive mental health map of England, every figure sourced to OHID Fingertips with the method published. Then compare what the context suggests with what your survey says, and design support around the gap. Wellbeing is not shaped in a vacuum, and neither should your strategy be.