"Blood Tests Cannot Predict": One Position, Three Speakers, and What It Leaves Out

On 23 September 2026, 49 Thornton-Cleveleys residents received PFAS blood test results arranged by the law firm Leigh Day. Reports say 69.4% were above 20 ng/ml, the higher of the US guidance thresholds (blood test results). Between 22 and 24 September, three bodies advised against individual PFAS blood testing, in similar words.

The same position, three times

The UK Health Security Agency, 22 September 2026, the day before the results were handed out. Its general information page on PFAS, first published that morning, says:

“PFAS biomonitoring is not currently recommended for individuals through the NHS or private blood tests. At this time, such tests are unlikely to provide clear or helpful information for individuals and can sometimes create unnecessary worry. PFAS blood testing also cannot be used to predict future health problems or determine the cause of any existing ill-health, and there is no evidence that reducing PFAS in the body on an individual basis will improve health outcomes.”

AGC Chemicals Europe, 23 September, the company that has operated the Hillhouse site since 1999, in a statement to ITV News:

“We would emphasise that the UK Government’s position is clear: PFAS blood testing does not predict future health problems and can sometimes create unnecessary concern. For these reasons PFAS biomonitoring is not currently recommended through the NHS or private blood tests.”

Lancashire County Council’s Director of Public Health, 24 September, speaking to BBC Radio Lancashire, said he was concerned that residents had been offered blood tests that “cannot show where the exposure occurred, whether PFAS has caused a health condition or what their future health risk may be”, and that:

“PFAS bio-monitoring is not recommended through the NHS or private blood tests.”

AGC referred explicitly to the government’s position, and its second sentence closely follows UKHSA’s wording. There is one difference of substance: UKHSA says testing is not recommended “for individuals”, and AGC’s version does not include those two words. The Director of Public Health’s version does not include them either, or the word “currently”. These statements do not show how any of them was prepared.

The company whose site is the subject of the investigation, and the public bodies responsible for residents’ health, are making the same point in similar words. The rest of this post sets out what that point leaves out.

What the position gets right

A blood test cannot tell one person whether they will become ill. It cannot say where their exposure came from, and it cannot say whether PFAS caused an illness they already have. This site does not claim otherwise (Common criticisms).

What it leaves out

1. “Cannot predict” is not the same as “tells you nothing”. The US National Academies’ 2022 guidance uses blood levels to place people in risk bands. Above 20 ng/ml it describes an increased risk, with additional screening advised (NASEM, 2022). A result can inform what a doctor looks for without predicting what will happen.

2. UKHSA distinguishes testing individuals from measuring exposure across a population. The same page says the agency “is developing capabilities to assess human exposure to PFAS” and has access to population samples of blood and urine from the Health Survey for England. The 49 results were also reported as a group: the share above each threshold, and higher averages within 1 km of the site. The 49 were not a random sample of the community (why that matters, and why it does not make the results meaningless). But group results are a different kind of information from one person’s test, and the individual-level objection does not address them.

3. Other countries have offered this testing. Around 3M’s plant at Zwijndrecht in Belgium, everyone living within 5 km was offered a free blood test, and 8,568 people took part in 2025. At Parkersburg in the United States, 69,030 people joined the C8 Health Project in 2005 and 2006 (what authorities did around other sites).

4. The evidence review published the same day was narrowed. Alongside the general information page, UKHSA published a rapid review which concluded that there is “inadequate evidence to determine if there are potential adverse effects of chronic PFAS exposure from local industrial and manufacturing sites” on nearby communities. Its methods section records that:

• the protocol was amended “to account for time constraints”;
• papers on “biomarkers of effect”, measurable biological markers associated with health outcomes, were excluded;
• only cohort studies were included. The review itself says that “by excluding other study types, such as case-control studies or cross-sectional studies the review may not have captured the full breadth of available evidence”;
• where a single paper covered an outcome without a statistically significant primary result, the review treated the evidence as inadequate “regardless of confidence rating of the paper, magnitude of the effect reported, or consistency of results in the paper”.

The same review notes that associations have been reported between PFAS exposure and “lower birthweight, certain cancers (particularly testicular and kidney), and increased cholesterol levels”, and that “[a] more comprehensive review using evidence from a wider range of study types would be beneficial”.

The questions this leaves

• For residents above 20 ng/ml, who provides the screening that the US guidance advises, and the “specialist clinical advice and oversight” the Director of Public Health says any treatment would need?
• If individual testing is not recommended, will UKHSA or the council measure exposure in this community at population level, of the kind UKHSA describes developing for the general population?
• Will the fuller review that UKHSA’s own report calls “beneficial” be carried out, and will it include the study types this one excluded?

The quotations above were checked against the published pages on 27 September 2026. The UKHSA publication dates are taken from the pages’ own metadata.