A month ago, the government’s top infectious disease expert, Anthony S. Fauci, appeared on the “Today” show to discuss the ongoing coronavirus pandemic.

Asked whether experts were overstating the number of deaths from covid-19, the disease caused by the virus, Fauci said that was unlikely. He was then asked whether the deaths were being undercounted.

“I think there’s more of a chance of missing some that are really coronavirus deaths that are not being counted,” he said, “but I don’t think that number is significant enough to really substantially modify the trends that we’re seeing at all.”

That was April 9, when there were about 16,000 deaths nationally, nearly half of which were in New York state.

A version of the same question was posed to Fauci during his testimony before the Senate on Tuesday.

“The official statistic, Dr. Fauci, is that 80,000 Americans have died from the pandemic,” Sen. Bernie Sanders (I-Vt.) said, referring to the current death toll. “There are some epidemiologists who suggest the number may be 50 percent higher than that. What do you think?”

“I’m not sure, Senator Sanders, if it’s going to be 50 percent higher,” Fauci replied. “But most of us feel that the number of deaths are likely higher than that number.”

The reason was straightforward. During the peak of the outbreak in New York City in particular, Fauci said, “there may have been people who died at home who did have covid, who are not counted as covid because they never really got to the hospital.” By now, the death toll in New York state is near 27,000.

“I think you are correct that the number is likely higher,” Fauci added. “I don’t know exactly what percent higher, but almost certainly it’s higher.”

That statement undercuts a broad range of theories which have asserted that the number of deaths from the virus is being overstated. The purported reasons for the overstating vary, from claims that deaths from other causes are being misattributed, to insistences where the number is being intentionally inflated. But the evidence has consistently suggested that the number is too low, not too high, as Fauci stated.

Consider his point about the number of deaths in New York. The Post looked at one way of evaluating the full death toll from the virus last month, comparing the number of reported deaths from any cause through early April to historic figures for the same time period. Our research suggested that there were more than 15,000 more deaths than normal, twice what was attributed to the virus at the time. In New York, we counted about 8,000 more deaths than normal.

How would this happen? As Fauci said, it may be the case that people who were suffering from the virus died before being able to get to the hospital or before being tested. (There are certainly numerous anecdotal reports to this effect.) It may also be that the number of deaths increased as a function of other causes, as well, such as people deciding against seeking medical attention for urgent conditions, perhaps out of concern about contracting the virus.

More broadly, though, it’s always the case that the number of observed deaths is lower than the likely toll of an illness. Each year, for example, the Centers for Disease Control and Prevention tracks the number of people who are confirmed to have died of the seasonal flu. It recognizes, though, that this misses some cases of flu-related deaths, so it also generates an estimate of the actual death toll from the flu. That estimate is always significantly higher than the observed number of deaths.

Those 80,000 covid-19-related deaths are the observed deaths. It’s not necessarily the case that the gap between observed and estimated cases will be as broad for the coronavirus as for the flu, but it would be consistent with the CDC’s analysis that the estimate would be somewhat higher.

One way in which some critics have attempted to downplay the toll from the virus has been to suggest that those who died were often at risk of dying from other causes anyway. That those who die of covid-19 are disproportionately elderly is seen as evidence to that point, as is the extent to which covid-19 deaths occur in nursing homes and long-term care facilities.

Data compiled from the Kaiser Family Foundation through May 7 indicate that about a quarter of deaths from the virus occurred in nursing homes in states that report those data.

This data point has been used on some occasions to suggest that coronavirus was simply accelerating the inevitable.

This misses a few key points. The first is that pushing someone who’s standing at the edge of a cliff doesn’t mean you’re not a murderer. Even if patients are at risk from other conditions, their dying more quickly as a result of the virus is still a function of the virus. What’s more, this is an overly broad assumption. Studies indicate that those who have died of the virus might have been expected to have lived another 10 years on average.

What’s also often overlooked in considering nursing homes is that these are contained facilities of the sort where outbreaks have been common. A relatively densely populated community of higher-risk individuals is a key part of the problem.

It’s worth considering something else as well. It may actually be easier to track the death toll in a nursing home than in the public at large, given existing systems for tracking patients and the contained nature of the facilities. People who live at home and die alone are harder to track than patients in care facilities that are under scrutiny.

None of this is really new. We’ve known for a while that coronavirus deaths are probably higher than the current count indicates. What Fauci offered Tuesday is confirmation of that likelihood — a confirmation that he was more reticent to offer a month ago.