On average, people had antibodies against 10 species of virus. There were a few people who'd been infected with many different viruses—five people had antibodies against 62 species, and two had been infected with 84 different species.
It's well known that viruses can cause disease. The human papilloma virus (HPV), for instance, causes cervical cancer, as well as head and neck cancer and cancers of the genitals. Epstein-Barr virus, which infected 88 percent of the volunteers, is linked to lymphoma and is suspected of causing some cases of stomach cancer. It's also been linked with multiple sclerosis.
Read MoreEbola survivor's eye teeming with virus
Enterovirus D-68, a distant relative of polio, concerned doctors when it made an unusual resurgence last year and appeared to cause a mysterious polio-like syndrome in a few children.
Having a test that can look for all sorts of viral infections at once can help researchers tighten up these links, and also help them find possible other links between viruses and long term health.
It's not perfect. Elledge said some viruses didn't turn up as often as they expected.
"For example, the frequency at which we detect influenza (53.4 percent) and poliovirus (33.7 percent) is lower than expected given that the majority of the population has been exposed to or vaccinated against these viruses," they wrote.
Read MoreUS: Bird flu vaccine not good enough for outbreak
It also only found that about 24 percent of the people tested had antibodies to chickenpox, also called varicella, which is far fewer than expected.
But when they tried the test against people who knew they had HIV and hepatitis C, the test was 95 to 100 percent accurate. "We didn't falsely identify people who were negative," Elledge added. "That gave us confidence that we could detect other viruses, and when we did see them we would know they were real."
Rare viruses were appropriately, well, rare.
"Although we detected antibody responses to rare and highly virulent viruses such as Marburg and bat lyssavirus, they were found in less than 0.4 percent of the population," they wrote.
Elledge's team says the test could easily be expanded to include other human pathogens such as bacteria, fungi, and protozoa. The test might also be used to find out if there are consequences of being infected with two or more particular viruses.