Is bottled water safer than tap water?

Who sets the rules. Public tap water falls under EPA standards. The FDA regulates bottled water as a food, with its own rules, and for each EPA tap-water standard it either adopts the limit for bottled water or finds it unnecessary¹. Some limits differ on purpose: for lead, the EPA's action level at the tap is 15 parts per billion (ppb), while bottled water is held to 5 ppb because it does not travel through old pipes¹. A 2009 review by the U.S. Government Accountability Office found other gaps: bottlers did not have to use certified labs or report test results to the FDA, and, unlike utilities, they had no duty to send customers an annual quality report⁶. Some bottled water is simply treated tap water¹.

Where tap falls short. Tap water is not uniformly safe. An analysis of 17,900 community water systems found violations were more common in rural areas, with hot spots in the Southwest². Your own utility's annual report shows where you stand.

What bottled water adds. A systematic review of 12 studies found microplastics (plastic fragments under 5 mm) in both tap and bottled water, with maximum counts of 628 particles per liter for tap and 4,889 for bottled; studies counted different particle sizes, so numbers are not directly comparable³. A 2024 imaging study that could see particles down to about 100–200 nanometers estimated roughly 240,000 plastic particles per liter in three bottled-water brands, about 90% of them nanoplastics (smaller than 1 micrometer)⁷. Whether these amounts affect health is unknown; the WHO judged in 2019 that, on limited data, microplastics in drinking water do not appear to pose a health risk at current levels, while calling for more research⁸. Bottled water can also contain PFAS: 39 of 101 U.S. products had detectable amounts, mostly below 5 ng/L⁵.

Heat and storage. In lab tests of eight U.S. brands, 60 °C heat plus shaking raised nanoparticle counts 9.29-fold⁹. Antimony from PET bottles stayed far below the EPA limit at room temperature but would reach it in about six months at a constant 60 °C¹⁰.

Mineral water. In half of 21 North American tap supplies, 2 L a day supplied 8–16% of the calcium and 6–31% of the magnesium reference intake, while a liter of moderately mineralized European bottled water supplied 20–58% and 16–41%¹¹. FDA mineral water must contain at least 250 ppm of dissolved solids from the source¹.

When bottled is the right call. Boil-water and do-not-drink advisories and emergencies¹². For the environmental and cost angle, a reusable bottle filled from the tap avoids both.

Regulatory comparison. The GAO's 2009 report (GAO-09-610) compared the two systems and concluded that FDA protections for bottled water were often less stringent than EPA protections for tap water⁶. Bottlers were not required to use certified laboratories or to report test results to the FDA, had no equivalent of the 24-hour, 30-day or one-year public notification duties placed on water systems, and had no counterpart to the annual consumer confidence report the EPA has required since 1998. FDA gave bottled-water plants low inspection priority, and states performed about 70% of inspections in fiscal years 2000–2008⁶. These are 2009 findings. The FDA now says bottled-water facilities register as food facilities and follow preventive-controls rules¹, and we did not find an updated head-to-head audit.

Tap-water violations. Allaire and colleagues analyzed 17,900 community water systems from 1982 to 2015. Health-based violations clustered in rural systems and in hot spots, particularly in the Southwest, with repeat offenders; private ownership and purchased water were associated with compliance². Violation data describe a system, not the water at one tap, where home plumbing matters.

Plastic particles: three different kinds of study. Mason and colleagues tested 259 bottles from 11 brands bought in 19 locations across nine countries: 93% showed microplastic contamination. The average was 10.4 particles larger than 100 µm per liter, and 325 per liter when particles of 6.5–100 µm were included, with a range from zero to more than 10,000. Polypropylene, the cap polymer, was the most common type (54% of confirmed particles)⁴. Qian and colleagues used stimulated Raman scattering imaging on three brands and estimated (2.4 ± 1.3) × 10⁵ particles per liter, about 90% of them nanoplastics, roughly 10 ng/L by mass. Only about 10% of imaged particles matched their seven-polymer library, and they noted the count could reach 10⁶ per liter if every organic particle were plastic⁷. The systematic review of 12 studies could not pool results because of heterogeneity (I² above 95%), and the smallest particle size analyzed (1 to 100 µm) strongly shaped counts; PET and polypropylene were the most common polymers³. The cautious reading is that the highest counts reported are higher for bottled than tap water, that bottle and cap polymers show up in the particles, that totals depend on the size range measured, and that no study links these amounts to disease.

PFAS. Chow and colleagues screened 101 uniquely labeled U.S. bottled-water products for 32 PFAS. Thirty-nine had detectable PFAS; among those, totals ranged from 0.17 to 18.87 ng/L (median 0.98), and 97% of all products were below 5 ng/L. Spring waters had more than purified waters, which the authors linked to reverse osmosis treatment in 25 of 35 purified products versus 1 of 45 spring products. No enforceable PFAS rule existed for U.S. bottled water at the time⁵.

Heat. Westerhoff and colleagues tested nine bottled waters in Arizona: antimony was 0.095–0.521 ppb against the EPA limit of 6 ppb and did not change over three months at 22 °C. Release rose with temperature, and their model predicted the limit would be passed after 176 days at 60 °C, 12 days at 70 °C and 1.3 days at 85 °C¹⁰. Hussain and colleagues exposed eight U.S. brands to 60 °C, shaking and 15-day temperature cycling; nanoparticle counts rose 9.29-fold with heat plus shaking, with particles from both bottle and cap⁹. These are lab simulations of worst-case storage, not typical shelf conditions.

Minerals. Azoulay and colleagues compared published mineral analyses from 21 North American cities' tap water with bottled-water data and dietary reference intakes. Mineral levels varied widely among both, and the authors advised physicians to encourage patients to check the mineral content of their drinking water, tap or bottled¹¹. This is a calculation of potential intake, not a trial of health outcomes.

Null and conflicting results. We found no trial comparing health outcomes in people who drink tap versus bottled water. The WHO's 2019 view that microplastics do not appear to pose a health risk at current levels rested on limited data⁸, and it predates the 2024 estimate of nanoplastics in bottled water⁷, so “no apparent risk” is a statement about what has been measured so far, not proof of safety.

Bottom line (confidence 3 of 5). Bottled water is probably not safer than U.S. tap water. Both can fail, tap water comes with a free annual report, and bottled water adds plastic particles whose health effects are unknown. Bottled water is the right choice during advisories and emergencies. For daily use, filtered tap water chosen to match your utility's report is the more transparent option.