The United States and Russia have the two highest numbers of cumulative, confirmed coronavirus cases in the world.

As of May 20, the United States also has the highest official death toll.

Russia, however, is ranked 19th for its reported number of coronavirus fatalities, and its statistics are increasingly under scrutiny from experts who suspect something's not quite right with Moscow's methodology.

Is the United States overcounting its coronavirus deaths? Is Russia undercounting?

Most countries around the world try to adhere to the guidelines set out by the World Health Organization (WHO) for classifying coronavirus deaths: "a death due to COVID-19 may not be attributed to another disease (e.g. cancer) and should be counted independently of preexisting conditions that are suspected of triggering a severe course of COVID-19."

In other words, if a person has a heart condition, then acquires the coronavirus, and later dies, it should be classified as a coronavirus death. The same for diabetes or respiratory illnesses like pneumonia.

"When studying the pandemic, you should count as much as possible, then you can classify the cases otherwise later on," said Steven Van Gucht, a virologist at Belgium's main public health institution, Sciensano.

Experts say there is a mix of explanations why coronavirus fatality statistics vary, some a function of long-standing, institutional practices and traditions within countries, some potentially a function of politics.

According to Johns Hopkins University, the U.S. institution whose database on infections is considered one of the most authoritative, differences in mortality numbers can be caused by things like the number of tests performed in a population: The more testing there is, the more people with milder cases are identified, and this then lowers the ratio of cases to fatalities.

Also, the older or sicker a population is, the higher mortality rates are likely to be. And each country's health-care system is a factor. The number and quality of hospitals and doctors, for example, affects how infected patients are treated and whether they recover.

In the United States, the Centers for Disease Control and Prevention -- the lead government agency for health and disease responses -- says its statistics are based on numbers entered into the National Vital Statistics System, from all 50 U.S. states.

Among the statistics' shortcomings are the lag time: Several weeks can elapse before a COVID-19 death will be processed, coded, and tabulated, and then be reflected in national figures.

Call It COVID-19

In Russia, there's a more fundamental issue, with mounting evidence that the deaths of many people infected with the coronavirus have been attributed to other diseases or conditions.

Yelena Malinnikova, the Health Ministry's chief of infectious diseases, argued on May 4 that the low mortality was due to testing and quick detection.

Russia has been praised for its wide national testing program, with more than 7.5 million tests conducted.

But Russia's official figures, already under scrutiny, drew more attention earlier this month when news media including The Moscow Times, The New York Times, and the Financial Times, examined preliminary fatality rates for Moscow for April, and discovered they were markedly above average. That has prompted angry denunciations from the Foreign Ministry.

Russia's WHO representative has also downplayed doubts about the country's tallies.

Mikhail Tamm, an associate professor of Moscow State University and Moscow's Higher School of Economics, told RFE/RL's Russian Service that there was a twofold discrepancy in such deaths in some Russian regions, as local agencies tally and report statistics.

In Moscow, Tamm noted, the city Department of Health said that more than 60 percent of patients infected with the coronavirus were not included in the death statistics. The coronavirus was considered only a "catalyst" for the development of other diseases.

Whether that is a deliberate political decision -- perhaps to minimize the perception that Russia is suffering disproportionately -- is an open question.

"We can say that there are several factors" in Russia's unusually low fatalities, Tamm said, "but only one of them, which the Moscow authorities openly wrote about, allows us to assess how great an understatement [it] is."

Death Rates, Death Counts

Other countries whose overall coronavirus numbers have come into question include Iran, which as recently as early April was one of the leading countries for confirmed cases. Last month, the country stopped publishing provincial figures, even as authorities noted "a rising trend or the beginning of a peak" in eight regions.

As of May 20, the country had reported 7,119 deaths, according to the Johns Hopkins University database, which relies on reports from countries' governments for its figures.

On the flip side are countries like Belgium, which is only seventh in the ranking of deaths, with 9,108, but instead has the world's highest fatality rate: 79.50 per 100,000 people.

By comparison, the U.S. rate is about one-third that: 27.61. Russia's is 1.88 per 100,000 people.

A Belgian government spokesman, Yves Van Laethem, earlier this month tried to dispel perceptions that the country was exceptionally ravaged by the disease. The reason is that authorities track "excess deaths" for the period that the coronavirus has been in the country.

That figure is the number of deaths in excess of what would otherwise be expected for the same period, based on past statistics – similar to what reporters found for Moscow in April.

"Our way of counting things is the most scientifically correct and honest," Van Laethem was quoted as saying on May 15.

The Belgian virologist, Van Gucht, said that, because the coronavirus that causes COVID-19 was so new when it emerged, many countries did not know how best to respond or how to classify and tally deaths. That's changed as more information and data have been generated.

"It's a bit silly that countries are worried about their image, because a virus. It's nobody's fault. It's a force of nature. You try to deal with it," Van Gucht told RFE/RL. "It's become too much of a competition between countries, about their image.

"It's a bit of a political issue, because we, as scientists in public health, we don't care if one country looks better or worse than other countries," he said. "We just much want as much data as possible, as soon as possible."

Virus Politics?

And then there are countries whose statistics are beyond implausible: Turkmenistan to this day has not reported any coronavirus cases at all, despite being surrounded by countries that have.

The World Health Organization itself has come under severe criticism from U.S. President Donald Trump's administration and from other governments. Much of the criticism has focused on the organization's interactions with China, where the disease first emerged in November and December, and on the perception that it has provided political cover for authoritarian governments.

Its local representatives have also come under fire -- for example, in Tajikistan, where authorities insisted there were no cases for weeks, a position endorsed by the WHO's representative there, and its president flouted warnings from international experts to order social-distancing restrictions or other measures aimed at curtailing any spread of the disease.

On April 30, however, Tajik authorities announced the country's first cases; as of May 19, there were nearly 2,000 confirmed cases and 41 deaths.

James Aldworth, a spokeswoman at WHO's global headquarters in Geneva, told RFE/RL that the variation in data is partly explained by differences in reporting methods and testing strategies from one country to another.

"It is therefore not unusual to see the data on severity vary from one area to another as the local context plays a role in the spread and scale of infection of the disease," he told RFE/RL in an e-mail.

"Many countries are in fact struggling to capture the deaths that are occurring due to COVID-19," he said.

Aldworth did not respond to questions about whether WHO has in some cases provided political cover for authoritarian governments concerned about negative media coverage on the coronavirus.