Researchers sampled 9 buildings with confirmed mold problems and 9 matched clean buildings, using open dishes and measured air samples side by side.
If you are thinking about mold testing with a DIY kit, the kind where you leave a dish of nutrient gel open in a room, know this first: the dish comes back positive in a clean house too. In a study of moldy and clean buildings, open dishes in fall averaged the same in both. Samples that measured a known volume of air did better.
Every set of dishes in the clean buildings grew fungal colonies. In fall, dishes averaged 25 colonies in buildings with confirmed mold problems and 26 in the clean ones.
The study
Anne Hyvärinen, Tiina Reponen, Tuula Husman, Juhani Ruuskanen and Aino Nevalainen of Finland’s National Public Health Institute and the University of Kuopio compared 9 buildings with confirmed mold problems against 9 reference buildings matched for age, site, use, construction materials and design. Twelve were homes and six were day-care centers. The paper appeared in Indoor Air in 1993.
Every problem building had a history of dampness and mold growth confirmed during site visits, caused by leaks, poor ventilation, bad drainage or condensation. The reference buildings were inspected too, and no signs of moisture or mold were found.
In fall 1990 and winter 1991, the team sampled the air in 2 to 4 rooms of each building two ways. Six open culture dishes sat on tables at about 1 meter for an hour. Separately, an instrument pulled a measured volume of air onto culture plates for 10 minutes. Both were grown and counted the same way.
What they found
Open dishes, fall: 25 colonies in the problem buildings, 26 in the clean ones.
Geometric means per six dishes per hour. No statistically significant difference in the total or in any type of mold.
In winter the dishes read 17 against 7. That points the right way, but the difference in total was not statistically significant. Only Penicillium differed significantly.
Measured air, fall: 250 against 160 colony-forming units per cubic meter.
A significant difference. In winter, 120 against 58, significant at a stricter level.
In winter the measured air samples also found Stachybotrys, Acremonium, Oedocephalum and Botryosporium only in the problem buildings. In the authors’ words, the measured-air instrument “gave most clearly the difference between mold problem and reference buildings.”
Why it would happen
An open dish catches only what happens to fall on it, from an amount of air nobody measured. The authors note that the dish method “overestimates larger particles.”
Season matters too. Outdoor air in fall ranged from 37 to 11,000 colony-forming units per cubic meter, and the authors write that “the outdoor air had a clear effect on indoor air fungi.” When outdoor spores are pouring into a building, a dish fills up whether or not the building has a problem. That last point is our reading, not something the study tested.
The limits, stated plainly
Eighteen buildings in subarctic Finland, sampled more than thirty years ago with culture-based methods. Not every sampling method in use today was tested.
The authors are kinder to the dishes than our headline is. They write that the dish results “supported” the measured-air results, pointing to the winter trend.
Group averages say little about any one house. Even with measured air, the ranges overlapped widely: 19 to 7,900 in the problem buildings in fall, against 40 to 580 in the clean ones.
The team used six dishes for an hour. A kit with one dish would likely be noisier still. That is our inference; this study did not test retail kits.
What survives: a positive dish shows only that spores reached the room, which happens in almost every building.
What gets lost in the DIY kit marketing
A positive dish gets sold as an answer. In this study, the dishes were positive in every clean building. EPA says no federal limits have been set for mold or mold spores in the air, so no dish result can be checked against a standard either.
EPA’s mold testing page (last updated January 14, 2026, read October 9, 2026) says sampling “should be conducted by professionals who have specific experience in designing mold sampling protocols, sampling methods and interpreting results,” with lab analysis following methods from professional bodies such as the American Industrial Hygiene Association. The same page says that in most cases, if visible mold growth is present, sampling is unnecessary.
What the data supports
- Skip the open dish. In this study, the dishes grew colonies in every clean building.
- If you test, measure a known volume of air, and compare indoors with outdoors on the same day.
- Sample more than one room, including the one that worries you.
- Read the result by type of mold, not only the total.
- Can see mold? Skip the test, fix the water source, and deal with the growth.
Bottom line
A DIY mold dish comes back positive in clean houses too. In fall, open dishes averaged the same in moldy and clean buildings; samples that measured a known volume of air did better. If you test, test the air properly.
First step: note where and when the musty smell is strongest. That is where sampling should start.
illumenair does mold testing and indoor air quality testing in Portland and Vancouver homes.
We measure the air and explain what we found. Mold inspection in Vancouver, WA and Portland, Oregon
971-363-5626
Sources: Hyvärinen A, Reponen T, Husman T, Ruuskanen J, Nevalainen A. “Characterizing Mold Problem Buildings: Concentrations and Flora of Viable Fungi.” Indoor Air, 1993; 3: 337-343. US EPA. “Mold Testing or Sampling.” https://www.epa.gov/mold/mold-testing-or-sampling, last updated January 14, 2026, read October 9, 2026.