BOSTON — When state public health officials publish Monday’s report about the latest COVID-19 impacts on Massachusetts, the cumulative death toll through two years of the pandemic suddenly will stand about 15 percent lower.
The Baker administration will start using a new public health surveillance definition next week, narrowing the window of time between a confirmed COVID-19 diagnosis and death required for the fatality to get attributed to the highly infectious virus.
Saying the Bay State’s earlier methodology led to a “significant overcount of deaths,” officials said Thursday that they will adopt a new system recommended by the Council of State and Territorial Epidemiologists.
And in a step that could reshape understanding of the pandemic’s impact on Massachusetts, the administration will apply the new method retroactively, resulting in 4,081 deaths once linked to the virus being recategorized as stemming from other causes, and about 400 others newly being labeled as COVID-19 deaths.
It is not clear how many of those deaths, if any, took place in the Berkshires or how the county’s total fatality count could shift. As of Wednesday, the state had reported 408 COVID-19 deaths in Berkshire County.
“We think this is an absolutely critical step in improving our understanding of who COVID has impacted most significantly during the pandemic,” said state epidemiologist Dr. Catherine Brown. “We believe that this will provide us a much more accurate picture of who has died associated with a COVID infection in Massachusetts, and it will also improve our ability to compare our data with data from other jurisdictions.”
For the duration of the pandemic, state officials have deemed a fatality COVID-related if it met at least one of three criteria: if a case investigation determined that the virus “caused” or “contributed” to the death; if the death certificate listed COVID-19 or an “equivalent term” as the cause; or if state public health surveillance linked a confirmed COVID-19 diagnosis to a Bay Stater’s death.
The first two measures remain unchanged since the earliest days of the crisis, but the third already has been updated once and is set to evolve again Monday.
From March 2020 to March 2021, the Department of Public Health counted the death of any person who previously had tested positive for COVID-19 as a COVID-related death, regardless of how much time elapsed between those two events. Even if someone contracted the virus in March and died in a car crash in July, they were added to the ongoing tally of pandemic deaths for that first year.
“This strategy worked well at the beginning of the pandemic, and in fact, a paper was published last summer in the Journal of the American Medical Association, which lauded our efforts here in Massachusetts in counting deaths that occurred during the first wave of the pandemic as opposed to several other jurisdictions,” said Public Health Commissioner Margret Cooke.
“But, over time, our approach proved to be too expansive and led to a significant overcount of deaths in Massachusetts. People who had gotten COVID earlier in 2020 and died for other reasons ended up still being included in COVID-associated death counts.”
The department updated its approach for the third criterion in April 2021, officials said Thursday, keeping the death investigation and death certificate triggers in place. Under that method, officials counted only those who died within 60 days of a COVID diagnosis as deaths related to the virus, unless their death clearly was linked to another cause, such as trauma.
That system remained in place for most of 2021 and will be replaced in Monday’s daily report by the new definition, recommended in December by the national consortium of state public health leaders after months of study.
The new method suggested by the Council of State and Territorial Epidemiologists calls for counting deaths within 30 days of a COVID-19 diagnosis where “natural causes” is labeled on a death certificate as attributable to the virus, half as long a time frame as under the most recent definition in Massachusetts.
Brown said the update will “make sure that what we are capturing is the acute impact of COVID.”
“People who are seriously ill and hospitalized for longer and end up dying after that 30 days have almost invariably had COVID listed on their death certificate, so, they end up being counted under another method,” Brown said.
As has been the case throughout the pandemic, if an official death investigation determined that the virus caused or contributed or if a death certificate lists COVID-19 or an equivalent term, that fatality will add to the pandemic death toll.
Brown said the vast majority of the 4,081 deaths that no longer will be deemed COVID-related, about 95 percent, occurred from May 2020 to May 2021, covering the tail end of the state’s first surge and its second surge that winter. Most of the approximately 400 deaths that will acquire a COVID label also happened during that span, Brown said.
Taken together, the removals and additions net out to a reduction in the cumulative COVID-19 death toll of about 3,700 people, more than the entire population of Provincetown.
The DPH does not expect to have a new tally for the number of COVID-19 deaths in Massachusetts until it publishes new data on its COVID-19 dashboard about 5 p.m. Monday. Back-end work to merge datasets will take place over the weekend, according to Brown.
It also remains unclear if the change in Massachusetts will send out ripple effects across the country.
Asked if other states planned to adopt the national council’s recommended methodology as well, Brown said that the new definition planned for rollout in Massachusetts is “actually much more consistent with what many other jurisdictions are already using.”
“This is a recommended guidance definition, and it is designed to help improve comparability across jurisdictions, across states. But, we have also heard from a few jurisdictions that they are not planning on updating the way they count deaths,” she said. “What’s really important is that this change to the definition will actually increase the ability to compare the counts in Massachusetts with other jurisdictions because it will be more similar to what most other jurisdictions are using.”
The Center and Disease Control and Prevention’s online tracker on Thursday listed 959,533 total COVID-19 deaths across the country since Jan. 21, 2020. Massachusetts had the 13th-most total deaths among states and the 11th-highest rate of deaths per 100,000 residents, both of which likely will change when the updated death toll is published next week.
As of 5 p.m. Wednesday, Massachusetts health officials had recorded 23,708 confirmed and probable COVID-19 deaths since the outbreak first began, so, that figure is likely to drop to about 20,000 on Monday.
The new methodology also will apply to weekly reporting about COVID-19 cases in vaccinated individuals, the next version of which is set for publication Tuesday.
Brown added that preliminary analysis did not show any significant changes to the distribution of deaths by age group, gender and race or ethnicity once the new definition was applied.
“While we absolutely acknowledge that we’re moving to a more accurate and appropriate way to count deaths, it doesn’t change our understanding, it does not alter our understanding, of who has died from COVID and where the most disproportionate impacts have been,” Brown said.
The Baker administration appears not to have made as public an announcement about the first change to its statewide COVID-19 death definition as the latest update. Officials said in April 2021 that they would change how deaths were counted specifically in long-term care facilities to align with the CDC’s national definition, but made no mention in that news release of the broader change imposing a 60-day limit on the span between an infection and death to count in some cases.