Why microplastics studies are hard to trust: the methods problem

The most useful thing to understand about microplastics research is that the measurements are hard, and many published ones are less reliable than their numbers suggest. When a 2026 team screened 133 datasets from 101 studies reporting microplastics in human organs, tissues and fluids against a set of basic quality criteria, not one study — and not one dataset — met all the essential ones [s1]. That is not a scandal about fraud. It is a young field wrestling with a genuinely difficult measurement, and knowing why is the key to reading every alarming headline that comes out of it.

The contamination trap

The central problem is that plastic is everywhere, including in the laboratory doing the measuring. Airborne fibres, plastic labware, sample containers, even researchers' clothing can shed particles into a sample. So when a study reports plastic in blood or tissue, the first question is not "how much" but "how much of it was actually in the body, versus introduced during collection and analysis." Distinguishing the two is the hardest part of the work, and the easiest to do badly.

Guarding against it requires specific controls: procedural blanks run alongside samples to catch background contamination, positive controls to confirm the method finds particles it should, and clean-handling protocols throughout. The screening tool built these expectations into fourteen criteria across three categories — sampling methods, particle characterisation, and contamination mitigation — and applied them to the published record [s1]. The 2026 screening found the common gaps fell in exactly these areas: weaknesses in sample representativeness, in the use of positive controls, and in the documentation of blank procedures [s1]. Without them, a reported concentration can be measuring the lab as much as the patient.

Identification is also uncertain

Even a genuinely internal particle has to be identified, and that carries its own error. Confirming that a speck is polyethylene rather than a mineral, a fibre or a biological fragment requires techniques that vary in how specifically they can fingerprint a polymer, and the smaller the particle, the harder the identification. A 2026 framework paper argued that different analytical methods offer different levels of certainty, and proposed classifying techniques by their specificity and requiring agreement between complementary, orthogonal methods before a particle is counted as confirmed [s2].

The framework's practical proposal is that a particle should not count as confirmed on one technique alone: it sets minimum requirements from orthogonal methods for identification at different confidence levels, so that a claim of "polyethylene detected" carries with it how firmly that was established [s2]. Its deeper point was about communication. The paper called for "transparent reporting of methodological limitations and uncertainties" and for a shared way to state how confident an identification actually is, so that a tentative detection is not read as a firm one [s2]. A number reported without its uncertainty invites exactly the overinterpretation that dogs this field.

Why this changes how you read the news

None of this means microplastics are not in the human body — the better-controlled studies show they are, and the screening exercise exists precisely to sort stronger work from weaker. It means several things about individual findings. A single striking concentration, especially for the tiniest nanoplastics, may not survive scrutiny of how the sample was handled. Numbers from different studies often are not comparable, because they used different methods, size cut-offs and definitions. And a first-of-its-kind detection is, by the screening framework's own description, "pioneering exploratory research" — valuable, but not a settled measurement [s1].

The 2026 authors went further than method-checking. They urged researchers to "explicitly reflect on the plausibility of their results," noting that even studies with relatively strong quality control sometimes reported findings that warranted a hard look at whether they were biologically plausible at all [s1]. When a reported tissue burden implies more plastic than the plausible intake could deliver, the measurement, not the biology, is usually the thing to doubt.

The standards are still being built

The reason no study passed is partly that the criteria are new: until recently there was no agreed quality standard for microplastics in human samples, which is why one had to be built and applied retrospectively [s1]. Both papers are, in effect, the field trying to grow up — writing the rules that will let future studies be trusted and older ones be weighted accordingly [s1][s2].

For a reader, the takeaway is a habit rather than a conclusion. When a new microplastics finding lands, the questions that matter are whether it controlled for contamination, whether the identification was confirmed by more than one method, and whether the amount is biologically plausible. Studies that clear those bars are worth taking seriously. Many that make headlines have not yet been shown to.

This article is informational and is not medical advice.

Sources

• A QA/QC screening framework to assess studies reporting microplastic presence in human organs, tissues, and bodily fluids — Environment International, 2026-07-22
• Communicating Confidence in the Reliability of Micro- and Nanoplastic Identification in Human Health Studies — Environment & Health, 2026-01-26