What’s the true covid death toll in Pa.? New analysis sheds light

For many reasons, it’s hard to know the full impact of the covid-19 pandemic. Some deaths may have gone uncounted for reasons such as the person died at home and was never tested. A person may have died from an existing disease exacerbated by covid-19. Or a person may have died from a more indirect result of the pandemic, such as disruption in drug treatment that led to overdose.

Scientists have a method to better understand the full impact of events such as pandemics. They calculate “excess deaths” — the number of deaths in a given year beyond the number predicted based on past history.

A new analysis concluded Pennsylvania saw 19,776 excess deaths in 2020, the first year of the pandemic, and 18,880 in 2021.

“Excess deaths is a nice, simple, tractable way for even the lay public to understand the magnitude of what we’ve just experienced. This is as simple as, last year should have been about the same as this year, and it wasn’t. It was bigger by this number,” said Matthew Ferrari, the director of The Center for Infectious Disease Dynamics at Penn State.

“These excess deaths are above and beyond what we would have expected in the absence of a pandemic, whether the actual attribution of any individual death is due to covid infection, secondary infections during or following covid infection, or just reduced health care capacity to deal with conventional health problems because of the weight of the pandemic.”

Anna Ssentongo, a professor of trauma surgery and public health at Penn State College of Medicine, said the data “describes the indirect and downstream effects of a disaster.” It further shows things we should focus on in preparing for future public health emergencies.

Ferrari, who was not involved with the analysis, found it noteworthy how similar the excess death finding is to the official covid-19 death toll in Pennsylvania, which was about 42,000 as of last Monday. About 16,000 of the deaths came in 2020.

Dr. John Goldman, a UPMC infectious disease specialist, said it “tells us what I think we already kind of knew, which is for all the talk of over counting covid-19 deaths, the evidence is that we probably undercounted.”

Ferrari similarly said the true covid-19 death toll may be higher, since other contributors to the normal annual death toll, such as the flu and car accidents, were lower during the pandemic.

The analysis was done by Pennsylvania’s Independent Fiscal Office, considered a trustworthy source of information by both Democratic and Republican lawmakers.

One of the values of calculating “excess deaths” is that it helps to understand the magnitude of deaths that may be indirectly related to covid-19, such as those resulting from suicide or drug overdose or skipping medical care.

Regarding those, the analysis zeroed in on five types of deaths that may be indirectly related to covid-19: overdoses, homicides, suicides, vehicle accidents and non-vehicle accidents.

It found significant increases in several: a 44.4% increase in homicides, a 19.7% in overdose deaths and an 11% increase in non-vehicle accidents.

Goldman pointed to the rise in homicides, which had declined in the three years prior to 2020. He said it may be related to stress from being cooped because of lockdowns and restrictions. “The number of homicides is not surprising to me because a huge percentage of homicides are people killing their partner,” he said.

Ssentongo, who was involved in research which showed a rise in gun violence during the pandemic, wasn’t surprised the rise in homicides.

“While stay-at-home orders helped to reduce mortality from covid-19 and lessen the strain on hospitals, there were consequences, some fatal consequences. Domestic violence, child abuse, increased use of drugs and alcohol, lack of social support, lack of access to healthcare, and more,” she said.

The rise in overdoses, which had also been declining in the years leading up to the pandemic, seems another logical consequence of factors such as disrupted or limited access to drug and mental health treatment, Goldman said.

However, Goldman found it surprising that suicides declined, given the disruptions to mental health care, and said that should be a topic of further study.

The analysis begins with official death statistics for 2017, 2018 and 2019.

It concluded an average of 118,500 people per year died from diseases and medical causes in Pennsylvania. Therefore, that’s how many would be expected to die in 2020 and 2021 had they been normal years.

But the actual number of deaths from those causes, 138,276 in 2020 and 137,380 in 2021, was higher — the logical consequence of the covid-19 pandemic. (Some data for late 2021 wasn’t available, so the analysis used estimates.)

On a human level, the analysis underscores that 38,000 “family, friends and colleagues are not with us anymore, and that loss is exceptional compared to what we would have expected before covid-19,” Ferrari said.

Ssentongo said, “we, as a society, need to invest more in public health preparedness so that we can adequately prepare for disasters. This includes both the direct and indirect effects.”