Lead can sicken people of any age, but young children are at greatest risk; hundreds of thousands of them in the U.S. have elevated levels in their blood, according to the National Center for Health Statistics. The Department of Housing and Urban Development considers paint lead levels starting at 5,000 parts per million (ppm) or 1 milligram per square centimeter (1 mg/cm2) high enough to require evaluation in federally funded or aided housing.

We hired a licensed lead inspector to scan for lead in three pre-1960 homes owned by Consumers Union staffers. The homeowners then used seven do-it-yourself lead test kits, costing $8 to $30 each. Most of the kits quickly indicated whether lead was present.

The kits we tested detected lead levels as low as 2,000 ppm in our home-based tests. In our lab tests, some lead test kits detected lead at levels below 1,000 ppm. None of them falsely identified paint in a Consumers Union lab painted in 1990 as having lead. Some kits can be a reasonable first step in detecting lead, as long as you follow instructions precisely. Here’s what else our tests revealed:

Two kits were problematic. Pro-Lab’s $30 Lead-in-Paint-and-Dust kit (not in our Ratings) doesn’t give instant results. You mail the sample to an EPA-certified lab. Although the results we got from the kit were close to our lab results, Pro-Lab’s report tells customers that lead levels below 5,000 ppm are “safe,” rather than warning that 5,000 ppm is the government’s action level. Homeowners with a lead level of 4,500 ppm could be misled into thinking no further action is necessary, when experts believe they should consider remediation. Another kit, the SenSafe, $15, had confusing directions in tiny type, which made mistakes more likely.

All lead test kits require practice. Exposing the layers of old paint took strength, dexterity, and lots of practice. Home lead test kits use one of two chemicals to detect lead by color change. But correctly “reading” color changes when lead levels were low also took practice.