We suggest a statistical test for underdispersion in the reported Covid-19 case and death numbers, compared to the variance expected under the Poisson distribution. Screening all countries in the World Health Organization (WHO) dataset for evidence of underdispersion yields 21 country with statistically significant underdispersion. Most of the countries in this list are known, based on the excess mortality data, to strongly undercount Covid deaths. We argue that Poisson underdispersion provides a simple and useful test to detect reporting anomalies and highlight unreliable data.
Irregular statistical variation has proven a powerful forensic tool for detecting possible fraud in academic research, accounting statements and election tallies. Now similar techniques are helping to find a new subgenre of faked numbers: covid-19 death tolls.
That is the conclusion of a new study to be published in Significance, a statistics magazine, by the researcher Dmitry Kobak. Mr Kobak has a penchant for such studies—he previously demonstrated fraud in Russian elections based on anomalous tallies from polling stations. His latest study examines how reported death tolls vary over time. He finds that this variance is suspiciously low in a clutch of countries—almost exclusively those without a functioning democracy or a free press.
Mr Kobak uses a test based on the “Poisson distribution”. This is named after a French statistician who first noticed that when modelling certain kinds of counts, such as the number of people who enter a railway station in an hour, the distribution takes on a specific shape with one mathematically pleasing property: the mean of the distribution is equal to its variance.
This idea can be useful in modelling the number of covid deaths, but requires one extension. Unlike a typical Poisson process, the number of people who die of covid can be correlated from one day to the next—superspreader events, for example, lead to spikes in deaths. As a result, the distribution of deaths should be what statisticians call “overdispersed”—the variance should be greater than the mean. Jonas Schöley, a demographer not involved with Mr Kobak’s research, says he has never in his career encountered death tallies that would fail this test.
The Russian numbers offer an example of abnormal neatness. In August 2021 daily death tallies went no lower than 746 and no higher than 799. Russia’s invariant numbers continued into the first week of September, ranging from 792 to 799. A back-of-the-envelope calculation shows that such a low-variation week would occur by chance once every 2,747 years.
The state revealed the full count — which added thousands of additional deaths — only in January, after a report by the state attorney general suggested an undercount, and after a state court ordered the data be made public in response to a lawsuit filed by the Empire Center, a conservative think tank. As of this month, New York has recorded the deaths of more than 15,000 nursing home residents with Covid-19.
Melissa DeRosa, Mr. Cuomo’s top aide, tried to explain why the administration had withheld the data last year to state lawmakers in a conference call, saying she and others “froze” because of the federal request for data, which came in late August as the governor faced criticism over nursing homes.
Investigators have contacted lawyers for Cuomo’s aides, interviewed senior state Health Department officials and subpoenaed the governor’s office for documents relating to the alleged data coverup, the sources said.
When New York Gov. Andrew Cuomo’s administration learned last year that the federal government was about to release data on Covid-19 deaths in nursing homes, state officials were concerned: Would the federal numbers tell the public a different story than the state’s own?
….. No other state requested details about the data release, according to the official. The federal health officials on the call described the scope of the data, how it was collected and how it compared with the state’s data.
When it was released, the federal data included fatalities inside both nursing homes and hospitals. But the federal tally of 3,525 deaths was lower than the state’s total of 5,944 because nursing homes weren’t required to report deaths to the federal government from March and April, the deadliest period for the state in the pandemic.
What DeRosa told lawmakers had them aghast. Not only had Cuomo misled them; he had, in DeRosa’s telling, done it in order to keep relevant information hidden from U.S. investigators. If the latter were true, Cuomo administration officials could well be guilty of federal-obstruction and false-statements crimes. In other words, so shameful was their actual reason for covering up nursing-home deaths — namely, to make a wayward governor look like a fantasy hero — that Cuomo administration officials figured it was better to be seen as potentially felonious than to admit their crude political motivation.
As the New York Times reported on Thursday night, in the spring of 2020, DeRosa and other members of Cuomo’s inner circle, who have no public-health background, studiously purged the nursing-home death data from a report compiled by state health officials. The Justice Department was not eyeing them at the time. That happened months later, in August, when the feds began seeking information about the treatment of, and record-keeping about, COVID-stricken nursing-home residents by New York and three other states.
So what was going on at the time of the purge? Well — whaddya know! — it turns out that was just when Cuomo was quietly securing the state ethics approvals that would permit him to earn outside income from a book he’d decided to write. The book would inform the world about his unparalleled mastery of the COVID crisis — which, oddly enough, he contemplated as a work of nonfiction.