If you are confused by all the talk about “game-changing” coronavirus tests that are fast, cheap, and convenient, it’s no wonder.

President Donald Trump, governors, public health experts, and the media often have done a lousy job of explaining the nature and limits of these recently introduced antigen tests, and how they differ from better-known molecular diagnostic tests.

Antigen tests can signal a potential new infection even before it causes symptoms, by detecting a protein the virus starts making soon after it invades the body. The cell-phone-size device gives results in 15 minutes wherever the test is used, with just a swab taken from the front of the nose.

The U.S. Food and Drug Administration says antigen tests can be used for “rapid detection” and “diagnosis” of coronavirus infections, but the agency cautions that these tests aren’t as reliable as lab-based molecular diagnostic tests, which detect viral genetic material. Positive antigen test results often need to be confirmed by a molecular test.

In August, for example, Ohio Gov. Mike DeWine tweeted that he had tested positive for the coronavirus and would be quarantining for 14 days. Hours later, he tweeted that a “second” test had come back negative.

DeWine first had an antigen test. Then he had two molecular tests — the long-swab-way-up-the-nose kind that have been a technical, logistical, and public health nightmare in the United States.

Experts say the lower accuracy of antigen tests is not necessarily a problem. The challenge is to find a balance between convenience and accuracy. If antigen tests are given often enough to a group of people — say, college students or nursing home residents — then the frequency will offset any inaccuracy.

That means, for about $5 a day per person, antigen tests could fill the gaping need for a coronavirus surveillance tool, says Michael Mina, a Harvard Medical School epidemiologist and testing expert.

He has been advocating that approach since before the first antigen test, made by Quidel, was authorized by the U.S. Food and Drug Administration in August. He points out that molecular tests can be too sensitive, giving positive results days after a person is no longer spreading infectious virus. Antigen tests, in contrast, only work during that early transmission period, when the coronavirus’s telltale protein surges and then plummets.

Officials almost invariably lump molecular and antigen tests together, perhaps to make their numbers look better. This week, Trump, New Jersey Gov. Phil Murphy, and Assistant Health Secretary Brett Giroir (who is also Trump’s coronavirus testing czar) heralded the announcement that the federal government would distribute 150 million antigen tests made by Abbott Laboratories to the states.

» READ MORE: Chester County will notify thousands of ‘questionable’ COVID results, in wake of Inquirer story

But using antigen tests effectively will require careful deployment strategies, and standardized rules for reporting and classifying the data — things that are sorely missing now, the Washington Post reports.

If the challenges can be overcome, Mina, for one, believes antigen testing really could be a pandemic game-changer.

Using the technology to screen millions of people every few days in hot spots would be “more than enough to suppress the outbreaks that are happening and make the country much safer,” Mina said during a recent lecture posted on YouTube.

Three types of COVID-19 tests

All of these tests are on the market, and all serve specific purposes. Here’s a rundown:

Molecular diagnostic tests (a.k.a. PCR or viral tests): the long-swab-up-the-nose type.

Pro: They are the most accurate because they detect actual viral genetic material, not just proteins made by the virus or the immune system.

Con: Must be analyzed on machines usually found only in high-complexity labs. Results typically take at least a few days. Only provides a snapshot in time. Cumbersome for screening.

Rapid antibody tests: These look for disease-fighting proteins, called antibodies, in the blood.

Pro: Antibody tests can help define the scope of outbreaks by measuring how many people in a particular population have antibodies. In New York City, for example, this “seroprevalence” was a whopping 23% in May. (But “herd immunity” requires about 60% seroprevalence, so we’re far from that.)

Con: The tests aren’t very accurate, and no one knows if immunity lasts or what antibody level is protective. Hopes that antibody tests could identify people who are truly immune to the virus have faded.

Antigen tests: Can flag a new infection, even in people with no symptoms, by detecting proteins the virus makes soon after it invades the body.

Pro: Inexpensive and gives results that can be read on the spot in minutes. Can be used on a regular basis to screen groups such as students or nursing home residents.

Con: Not as reliable as molecular tests, and provides only a snapshot in time, so negative results offer only limited reassurance. Positive results may need confirmation with molecular diagnostic tests.