Playing with Statistics

Playing with Statistics

4/14/2020 ©Moira Dolan, MD

Is CoVID claiming some victims that, in a typical year, would have fallen ill with the flu, leading to a so-called “CoVID divergence”?

The CDC reported that positive tests for influenza virus were down 6.9% in the US in the week ending March 21, 2020.

As of the week ending April 4, 2020 the CDC reports there have been 39 million flu illnesses this season so far.

In the past 2 decades, US flu incidence has varied from a low of 9.3 million to a high of 45 million cases per year.

Deaths are another matter entirely.

As of the week ending April 4, 2020 the CDC reports there have been 24,000 deaths in the US due flu this season. This number has fluctuated over the last 2 decades from 12,000 up to 61,000 flu deaths in the US per season.

The World-o-Meter website reports that as of 4/14/2020 in the US, there have been 591,066 CoVID cases and 24,600 associated deaths.

That amounts to 74 deaths per 1 million population in the US.

It is crucial to know that this is counting all deaths in persons who died with a positive coronavirus test as a “CoVID19 death”, as required by the CDC. This is required regardless of what the physician assesses as the cause of death.

In Israel, where doctors are still allowed to state the medically probable cause of death, they officially report 14 CoVID deaths per 1 million population.

But with the US and many other national governments requiring that all deaths in a person who tests positive for CoVID19 having to be counted as a CoVID19 death, both of those US death statistics (flu and CoVID) have to be viewed with some suspicion.

It is interesting to get some perspectives by looking at other causes of death.

For example, in 2017 in the state of New York alone, 44,092 people died of heart disease, at a rate of approximately 121 per day.

Cancer claimed 95 lives per day in NY, while 21 per day died in accidents in NY.

What will happen to some of these death rates after the pandemic has eased off?

Due to stay-at-home mandates, the 2020 stats are likely to show lower accident rates.

There may be some higher cancer rates due to delays in treatment, but this will be offset to an unknown degree by fewer deaths due to cancer treatment itself.

The lull in cancer treatments (those treatments that can be delayed with little risk) will provide a unique opportunity to evaluate the relative excess deaths due to cancer treatments, an historically under-documented statistic.

Similarly, the delay of non-emergency surgeries will be an opportunity to evaluate a difference in deaths due to post surgical sepsis.

Heart disease stats may rise if economic stress plays a role. Could heart health actually improve in the long run if the closure of restaurants caused a majority of citizens to take a prolonged break from junk food?

Cigarette smoking is estimated to be the biggest cause of death in high-income countries.

Maybe cigarettes will be considered an unnecessary luxury on a budget, such as when smoking rates decreased by up to 20% in the 2008 recession.

Death from all causes in the US in 1929 was 11.9 per 1000. The top 10 causes that year were infectious diseases.

In 2017 the US death rate was 730.9/100,000 or 7.3 per 1000.

The 10 leading causes of death in 2017 were heart disease, cancer, unintentional injuries, chronic lower respiratory diseases, stroke, Alzheimer disease, diabetes, influenza and pneumonia, kidney disease, and suicide.

The excess deaths due to coronavirus will be absorbed into the “influenza and pneumonia” statistic when the numbers for 2020 are published.

It all reminds one of that wisecrack attributed to Mark Twain:

“Facts are stubborn things, but statistics are pliable.”

Stay Informed. Join the Mailing List.

Close Menu