Is the fast-spreading Omicron milder than previous COVID-19 variants? Not according to a new large-scale study in Massachusetts, which suggests its apparent mildness comes down to vaccination uptake and other factors.

The study examined hospital admissions and death rates throughout various COVID waves, covering the experiences of 130,000 patients.

It has not yet been peer-reviewed, but it appears to show that—despite appearances—the essential threat of hospitalization and death stayed consistent across the different variants causing those waves.

"Our analysis suggests that the intrinsic severity of the Omicron variant may be as severe as previous variants," wrote the researchers from Massachusetts General Hospital, Minerva University, and Harvard Medical School.

The data used in the study showed the Omicron wave generated fewer hospital admissions and deaths than previous waves had—however, when adjusting for demographics, vaccination status, and comorbidities, the researchers found Omicron to be no less dangerous in itself.

So how does this square up with earlier findings that Omicron was milder, in particular those coming from South Africa, where the variant was first identified?

There, the West's initial disbelief about the data indicating mildness was a major source of irritation for figures such as Shabir Madhi, a professor of vaccinology at the University of the Witwatersrand and a top World Health Organization adviser.

On May 5, Madhi told Fortune that he agreed Omicron was "not necessarily less deadly than earlier variants, as evident from the experience in Hong Kong."

Nonetheless, in the South African context, the Omicron wave caused around 7% of all COVID deaths since the start of the pandemic, as opposed to the preceding Delta wave, which caused around 45%.

"The reason for [Omicron] being 'milder' was because of the evolution of immunity mainly inadvertently from natural infection and complemented by modest vaccine uptake, rather than necessarily from intrinsic lower virulence of the variant," Madhi said.

The World Health Organization said on May 5 that about 14.9 million people around the world were likely to have died from COVID by the end of 2021—nearly three times the official death toll, which was around 5.4 million.

The WHO based its calculation on "excess mortality" statistics that count how many more people died during the pandemic than could otherwise have been expected to die.

The toll includes not only those who directly succumbed to COVID, but also those who died because overwhelmed health care systems could not sufficiently treat them for other conditions.