The state has changed how it classifies coronavirus cases. It could lead to a spike in numbers

The change may create an ‘iceberg phenomenon,’ revealing more cases than previously reported.

Nurses Jennifer Stevens (left) and Katie Dalton of ConvenientMD secured a patient swab at a drive-through COVID-19 testing site at Harvard Pilgrim Health Care’s Quincy headquarters’ parking lot.Craig F. Walker/Globe Staff/file

As the coronavirus pandemic stretches into a third month, state health leaders are reclassifying the way they count COVID-19 cases, a change that’s sure to lead to an increase in the number of reported victims and have a profound effect on our understanding of the devastating impact of the disease in Massachusetts.

The Massachusetts state health department last week issued new directives on what counts as COVID-19 to hundreds of local health boards based on guidance from a low-profile, but influential national group of disease trackers.

Officials are now classifying as infected those who were likely stricken by the virus but did not have the diagnosis confirmed through a laboratory test. The state also plans to take a closer look at death certificates — as far back as March 1 — hoping to uncover the true number of people who may have been felled by the virus.

“This will certainly change the number of detectable cases and deaths over the outbreak,” said Nadia N. Abuelezam, an epidemiologist and assistant professor at Boston College’s Connell School of Nursing.

Milder cases that had gone unaccounted and unnoticed will now be surfaced and counted, what researchers call a “clinical iceberg phenomenon.”

Public health officials have struggled to get an accurate count of infections, and experts say the current tallies vastly underestimate the virus’s spread. A chronic shortage of tests, unequal access to health care, and the asymptomatic nature of so many apparent cases have been major factors in hiding the true enormity of the virus’s impact here.

A true accounting is critical because it helps public health experts better understand the disease and how it spreads. It also can shape public policy — and opinion — as well as the allocation of resources.

Exactly how many people will be added to the state’s growing count of infections under this new way of measuring cases is unclear. The state health department said that it intends to publicly release the numbers using the new criteria, but could not say when that might happen.

The state’s new approach follows the April 5 recommendation of the Council of State and Territorial Epidemiologists, a nonprofit organization that called for health leaders in each state to use a standardized definition for coronavirus infections. The group also suggested officials reexamine earlier deaths that had the hallmarks of COVID-19.

The US Centers for Disease Control and Prevention concurred with the recommendations days later and suggested states follow suit.

New York City took an early lead, with health officials announcing on April 14 that the city would start including in its daily count of deaths people who had never tested positive for the virus but were presumed to have died of it. With that change, the city’s death toll soared in one day by more than 3,700.

Clarity about the extent of COVID-19 has been sorely needed, said Jim Collins, coauthor of the Council of State and Territorial Epidemiologists’ guidance to state officials.

“When I say it’s a [COVID-19] case in Michigan and somebody says it’s a case in Massachusetts, now we’re speaking the same language and that’s important for surveillance activities,” said Collins, director of the Communicable Disease Division of the Michigan Department of Health and Human Services. “That way we can create a bigger pool of information to inform our description and understanding of the outbreak and the transmission of the outbreak.”

Under the new rules, probable COVID-19 cases will include people who have experienced at least two specific symptoms from a list that includes: fever, chills, shivering with chills, headache, sore throat, or the sudden loss of taste and smell. To be officially classified as a probable case, a person with symptoms must also have been in close contact with someone confirmed to have had the disease by a lab test, or with someone who is a probable case. Alternatively, a person with symptoms must have traveled to, or live in, an area with “sustained, ongoing community transmission” of the virus.

All communities in Massachusetts currently meet that widespread transmission status, state health officials said.

Close contact is defined in the new guidance as “being within 6 feet for at least a period of 10 minutes to 30 minutes,” but the group of epidemiologists that established the guidance acknowledged it needs more data before that definition can be made more precise.

The council of epidemiologists also recommended that states add COVID-19 to the list of diseases required to be reported to state health departments. The Massachusetts health department said it already required doctors to report confirmed and suspected cases of COVID-19, but that cases in which infection is merely suspected or probable “get reported much less frequently.”

It said it will not require doctors and other health care providers to report probable COVID-19 infections from earlier in the outbreak.

“Although it would be valuable information, the additional burden placed on healthcare facilities and providers to report retrospectively is too great at a time when their attention is needed elsewhere,” the health department said in a statement.

But the department has pledged to review death certificates going back to March 1 or earlier to ensure a more complete accounting of COVID-19 deaths. It did not give a timetable to complete that review.

The Globe recently reported that the number of people killed by coronavirus in Massachusetts in the early days of the pandemic is likely much higher than reflected by the official death toll, according to the newspaper’s analysis of preliminary state death records from March.

Total deaths in Massachusetts soared by 11 percent in March compared to the March average for the last 20 years, a statistically significant increase that far exceeded the expected swings from year to year.

The March spike in deaths eclipsed the number of fatalities attributed to the coronavirus in official counts, and suggests that fatalities began to climb before the full extent of the outbreak was clear.

Now, with states using the same barometer to measure confirmed and probable cases, officials will be able to more accurately track the speed of the spreading virus and the effectiveness of interventions, such as social distancing and isolating probable cases, health experts said.

A recent study published in The Lancet Public Health illustrates just how much of an impact a change in definition can have on the number of reported infections. Researchers found that China’s health leaders changed the case definition of COVID-19 seven times between mid-January and early March, widening the definition over time to include milder cases.

The researchers concluded that had China used the fifth of those definitions throughout that period, it would have recorded more than four times the number of infections than it reported.

“As a result, when case definitions were broadened, more mild cases could be detected,” the researchers wrote. This, they said, suggested a “so-called clinical iceberg phenomenon” because so many previously uncounted cases had been hidden from view.