How Reliable Is the Data on Microplastics in Drinking Water?

Microplastics have been found in tap water, in bottled water, and in the rivers and groundwater that feed them. That much is settled. What is far less settled is a question that matters more for your health: how much is actually there, and can we trust the numbers?

A 2019 review led by scientists at the World Health Organization and Wageningen University set out to answer exactly that. Instead of running new samples, the team audited the existing science — grading 50 studies on microplastics in drinking water and its freshwater sources against nine quality criteria. [Koelmans et al., 2019] The result is a rare, honest look at how shaky the evidence base still is.

The headline finding is sobering: only four of those 50 studies passed every quality check. [Koelmans et al., 2019] If you have read alarming statistics about plastic particles in your water, this is the context those numbers were missing.

Why measuring plastic in water is so hard

There is no standardized method for sampling, extracting, and identifying microplastics in water. [Koelmans et al., 2019] Every lab makes its own choices: how much water to collect, how to filter it, whether to digest organic matter, and how to confirm that a suspicious speck is actually plastic. Those choices are not cosmetic — they change the answer.

The review built a scoring system adapted from earlier work on microplastics in biota, assigning each study a score on nine criteria: sampling method, sample size, sample processing and storage, laboratory preparation, clean air conditions, negative controls, positive controls, sample treatment, and polymer identification. [Koelmans et al., 2019] Each criterion earned 0, 1, or 2 points, for a maximum “Total Accumulated Score” of 18.

10 orders of magnitude

That span — from roughly 1 × 10⁻² to 10⁸ particles per cubic metre — is not just natural variation between clean lakes and dirty influent. [Koelmans et al., 2019] A large part of it comes from the fact that different studies targeted different particle sizes and used incompatible methods, making direct comparisons treacherous.

The scorecard: only four studies passed clean

On average, studies earned about half the available points — 8.41 out of 18. [Koelmans et al., 2019] That mirrors the score researchers had previously given studies on microplastic ingestion by animals, suggesting the problem is field-wide, not confined to drinking water.

Only four studies received non-zero scores on every single criterion: a surface water study by Wang and colleagues, a bottled water study by Mason and colleagues, and two wastewater studies by Ziajahromi and by Hendrickson. [Koelmans et al., 2019] All four were published in 2017 or 2018, which the authors read as a sign that methods are improving.

Certain steps were weak across almost the entire field. The average score was below 1 (out of 2) for five of the nine criteria: sample treatment, polymer identification, laboratory preparation, clean air conditions, and — worst of all — positive controls, which averaged just 0.21. [Koelmans et al., 2019] Positive controls check how many particles are lost during processing; without them, a study cannot know if it undercounted.

Bottled water scored best — for a mundane reason

Not all water types are equally hard to analyse. Average scores were highest for bottled water (13.7) and treated tap water (11.5), and lowest for surface water (7.9) and wastewater (7.3). [Koelmans et al., 2019]

Before reading too much into that, note the reason. Bottled and tap water don’t need a digestion step to strip away organic matter, so those studies automatically earned full marks on the sample treatment criterion. [Koelmans et al., 2019] Cleaner water is simply easier to analyse well — the higher scores reflect the difficulty of the matrix as much as the diligence of the researchers.

There’s another catch worth naming: only two treated tap water studies, two drinking water treatment plant studies, three bottled water studies, and a single groundwater study existed to review. [Koelmans et al., 2019] With samples that small, the averages are fragile.

What the numbers do tell us

Despite the quality limitations, some patterns held up. Surface waters showed the lowest concentrations of any water type — likely because most surface water studies chased only larger particles, while smaller particles are more abundant. [Koelmans et al., 2019] Wastewater treatment plant influent showed the highest median concentrations, reflecting direct domestic inputs, with effluent lower because treatment plants retain a share of the particles. [Koelmans et al., 2019]

On polymer types, 32 of 55 records identified the plastics they found. The most frequently reported, in order, were polyethylene and polypropylene (roughly tied), then polystyrene, then PVC, then PET. [Koelmans et al., 2019] The authors explain this with two factors: global plastic demand, and density. Polyethylene and polypropylene float; PVC and PET, at 1.3–1.7 g/cm³, tend to settle out of the surface waters most studies sampled — which likely suppresses their apparent abundance. [Koelmans et al., 2019]

Fragments and fibres were the most commonly reported shapes, followed by film, foam, and pellets. [Koelmans et al., 2019] And a consistent thread ran through the size data: studies that hunted for smaller particles generally found higher particle counts. [Koelmans et al., 2019] How small you look determines how much you find — which is precisely why loose comparisons between studies mislead.

The honest limit: no risk assessment yet

Here is where the review refuses to overreach, and where it earns its credibility. Human health risk depends on exposure, and exposure depends on trustworthy concentration data. [Koelmans et al., 2019]

It’s also worth being clear about scope. The review covers drinking water and its sources only — it does not account for microplastics people take in through food or by breathing. [Koelmans et al., 2019] So even a perfect drinking water dataset would capture just one slice of total exposure.

This is the nuance that most headlines strip out. “Microplastics found in bottled water” is true. “Microplastics in your water are harming you by X amount” is a claim the underlying data cannot yet support.

What this means for you

The practical takeaway is not to panic, and not to dismiss. Both reactions overreach the evidence.

• Treat single scary statistics with skepticism. A number pulled from one study — especially one that skipped positive controls or clean-air handling — may be off by orders of magnitude. The tenfold-times-ten spread in reported concentrations is a warning about method, not just reality. [Koelmans et al., 2019]
• Understand that “detected” is not “quantified.” The presence of microplastics in tap and bottled water is well established. The precise dose you ingest is not. [Koelmans et al., 2019]
• Watch for better data. The four reliable studies all appeared in 2017–2018, and the authors expect quality to keep improving as methods standardize. [Koelmans et al., 2019] Future numbers will be more trustworthy than past ones.

Reducing your own plastic contact is a reasonable, low-cost choice. But do it because it’s sensible, not because a specific health threat from water has been proven — because, as this review makes plain, it hasn’t been yet.