For the cautious — or simply curious — homeowner, an at-home water testing kit may seem reassuring. But there are high levels of variability between test kits’ abilities to detect potential contaminants in water, a new study from the University of Massachusetts Amherst has found.
“People might be concerned about their drinking water, whether they’ve heard things in the news, or they notice it tastes different, or the color is different,” says Emily Kumpel, associate professor of civil and environmental engineering at UMass Amherst and senior study author on the new paper.
While water quality reports are widely available from utilities, they only pertain to people on city water, not well water. Also, sometimes the water source isn’t the problem. “Some of these issues, like brown water, can come from home plumbing, and that’s something that the utility doesn’t always know,” says Kumpel. “A test to understand more about your home plumbing can be very helpful.”
However, finding a kit that actually works is easier said than done. The researchers found that there are hundreds of kits on the market — the availability of which can change daily — and it’s an unregulated field.
The researchers selected eight kits that evaluated levels of iron, copper, manganese and fluoride. Overall, they found high variability between the kits — some worked well, while others didn’t.
Despite the mixed results, there were some general takeaways from the research. First, the type of kit matters, and there are essentially two types available: one that measures for a particular element, and one that supposedly can measure a dozen parameters at one time, Kumple explains.
Generally, single-parameter tests had more consistently accurate results than multi-parameter ones, compared to laboratory-measured results. None of the multiparameter tests detected low levels of iron, while three out of the four single tests did (though the results often over- or underestimated the presence of iron).
Many tests advertise that they can detect high levels of iron (20-100 mg/L). While the multiparameter tests performed better when measuring for high levels of iron versus low levels, most tests still underestimated the actual concentration present. Kumpel advises that kit users should interpret the results with a healthy dose of skepticism, especially if you’re testing to see if there are concerningly high levels of metals in your water. “[These tests] might be a good first cut on things, but it’s not necessarily telling you all the information you need,” she says.
Kumpel says that single tests often have a preprocessing step for water samples that improves result accuracy. When testing for iron, changing the pH of the water makes it easier for the metal to be detected by kits.
The study also found that testing instructions and result interpretation guidance were inconsistent. For instance, for iron, one kit informed users that 0-0.3 parts per million (ppm) was “OK” and 0.5-5 ppm was “high,” while other test kits said that 0 ppm was “ideal.”
“It really points to the fact that this is an unregulated space,” says Kumpel. “This shouldn’t just be on the homeowner. These tests should be better checked for how well they actually perform, particularly under real-world conditions. A lot of them perform perfectly fine in deionized water, but not so much once you get into real water, [as] normal levels of background minerals or organics can interfere with your testing.”
One piece of advice Kumpel offers that you likely won’t see in kit instructions is to think about when you test. Metals like copper or lead (not assessed in this study) likely come from a home’s pipes, not the upstream distribution system. “You want to do what’s called a first draw sample,where you’re getting the very first water that comes out of your system that’s been sitting there overnight.” If metal is leaching from the pipes into the water, this first draw will have the highest concentration, giving you the best opportunity to detect it.
“If you want to see what’s going on directly from your well or directly from the distribution system, then you want to do what’s called a flush, which is, you let the water run for a few minutes before collecting your water sample.”
For consumers who are looking for a water test and are looking for something more reliable than the home test option, Kumpel says that state departments of environmental protection or public health websites list locally or nationally certified labs.
Consumers who should consider testing their water are well owners, especially after there has been flooding; people who live in older houses that haven’t had plumbing updated or replaced in the last two decades; or after a disaster such as wildfires or floods.
“There’s widespread mistrust in tap water across the U.S.,” says Kumpel. “Having access to be able to test your own water and confirm that it is okay — which is the most common result that people would get by testing their water — is a really good thing. I think this could be a positive tool if we can get these to work [reliably] and get people to really understand more about their water.”