Pet drinking-water evidence: what the studies show
By Alexander Snyder, Founder & Water Quality Data Lead · Study-level review completed August 8, 2026.
These five studies support narrow conclusions: private-well metals can exceed human regulatory comparators, and pets can reflect exposure in a documented contaminated setting. They do not establish a universal pet-safe drinking-water threshold, prove that a contaminant caused disease, or make a ZIP-associated record a result for a pet's water. No pet-safety conclusion should be inferred from a ZIP-level report.
The evidence is mixed by design. A negative canine bladder-cancer study and a feline kidney-disease association that did not survive multivariable adjustment belong beside the positive sentinel studies. Each summary below states the population, comparison, and limitation that control its use.
Heavy metals in dogs' well water (Dog Aging Project)
114 of 178 private-well samples (64%) exceeded at least one EPA MCL or guidance comparator
In a Dog Aging Project pilot, 114 of 178 returned private-well samples (64%) had at least one analyte above an EPA maximum contaminant level or guidance level. Those were human regulatory or guidance comparators used by the study, not pet-specific safe levels. The study did not test municipal tap water or establish that an exceedance caused disease in a dog.
Univariable only: tap-water OR 3.43 (95% CI 1.08–11.45); not retained in the final multivariable model
In a small Bangkok case-control study (101 CKD cats and 29 controls), tap water (OR 3.43) and filtered water (OR 0.13) were associated with CKD status only in univariable analyses. Neither water variable remained independently significant in the final multivariable model; only commercial dry food did. The study did not prove a water effect or identify a contaminant.
Disinfection byproducts and bladder cancer in dogs
100 cases / 100 controls; no association detected with increasing chlorinated-water exposure
A case-control study of 100 dogs with bladder cancer and 100 controls detected no association between increasing years of chlorinated-water exposure and cancer risk. Cases had higher estimated total-THM exposure, but the difference was not statistically significant. The authors cautioned that dogs may receive less household tap-water exposure than people, so this study does not establish dogs as sentinels for this pathway.
Dogs and horses in a documented PFAS-contaminated community
31 dogs and 32 horses in a community with documented drinking-water PFAS contamination
Rock 2023 measured serum PFAS and clinical chemistry in 31 dogs and 32 horses from households in Gray's Creek, North Carolina, with documented drinking-water PFAS contamination. The study supports dogs and horses as exposure sentinels in that defined setting; it did not study cats, pair pets with human serum, or establish a pet-safe water threshold.
Wang 2018 compared serum from older Northern California cats with human group data from the same region and time periods; the samples were not paired by household. In the later cat group, summed PFAS was higher in hyperthyroid cats, but only PFOA was individually significant. The authors described a possible link and called for larger studies.
A Connecticut ZIP lookup provides background records. Identify the serving utility or well and use a certified sample from the relevant tap before drawing a conclusion for a person or pet.
These studies report measurements or associations in specific animal populations; they are not a diagnosis, a veterinary drinking-water standard, or evidence that an area record applies to a property. Health-protective EWG/OEHHA human references, EPA legal limits, and veterinary evidence remain separately labeled. Discuss animal-health questions with a veterinarian. Last updated August 2026.