COVID-19 News: “Batteries Not Included”

“Batteries Not Included.” 

We’ve all read these disappointing words on the package of a newly purchased product.  It is especially unsettling when the package contains a toy, and your child or grandchild has just unwrapped it excitedly on Christmas morning.

Friday, I “opened” a different kind of “package”—an article reporting new COVID-19 cases in Ashland County here in Northeast Ohio.  With all due respect to its author(s), I selected this article because it resembles many other articles I could have chosen to illustrate the points I hope to make here. 

Like a toy that comes with “batteries not included,” this Ashland Source article comes without an important “power source”—good journalism.  But there’s also a second “battery” missing— “good science.”  The third “battery” must be supplied by educated readers who can exercise sound critical reasoning.

Good Journalism

I’m not an expert in journalism.  However, I located a list seven qualities of a good news story provided by pivotcomm.com.   A well written story should include good sources (e.g. data from “good science”) and provide the necessary focus and context that enables well-educated readers to reach accurate conclusions.  With these criteria in mind, note that the headline of our “example article” reports the occurrence of 5 new positive results of COVID-19 infection in Ashland Co.  This heading would grab attention and maybe even elicit fear among those who think a second wave of the virus is likely.

The Ashland Source article goes on to explain that 4 of the 5 COVID-19 cases “were exposed to the virus at mass gatherings – a wedding, church service, and auction – where safety precautions were not observed.”  Can you identify at least one “red flag” here?  If so, you may be asking how the 4 individuals could have known the exact place and person responsible for infecting them.  (My record at doing this is dismal.)  And finally, who verified that safety precautions were not observed at each of the events--wedding, worship service, and auction?

In spite of limited information, the article includes a stern reprimand from Heather Reffett, Ashland Co. Health Department (ACHD) Commissioner:  “If safety precautions had been in place…at least four of these new infections could likely have been avoided.”  There is no mention of how individual #5 was infected. 

Clearly, the ACHD is faithfully conducting its civil duty.  But in so doing their message through this article suggests that the citizens of the Ashland area should remain isolated and locked down much as they were in April.  But does the larger context of the pandemic in Ashland Co. support this policy?

The article would have been more helpful if it had informed the readers that the 5 new cases bring the total number of cases in Ashland Co. to 27 as of June 12.  According to the last paragraph of the article, of these 27 cases, 22 have recovered and 5 are still recovering.  Although the article dutifully lists the dates, beginning March 18, when each COVID-19 positive case was reported, it does not mention that there have been no deaths in Ashland Co. to date.  No deaths in Ashland Co!  This statistic is quite striking considering that, according to my research at the Johns Hopkins Coronavirus Resource Center, neighboring counties ranged from 1 to 61 deaths (average = 21) attributed to COVID-19.  And what about death rates statewide in Ohio?

We would expect the number of COVID-19 cases in Ohio to increase now that increased testing is occurring.  But if the virus is posing a threat to Ohioans sufficient to justify a total lockdown for nearly two months, we would expect to see a significant rise in total deaths in 2020 compared to deaths in previous years.  But, according to Dr. Mary Kate Francis, interim medical director at the Ohio Department of Health, quoted in the Akron Beacon Journal, “Despite the climbing death toll, though, total deaths are up by 1.2%, or 680 people, in 2020 compared to the previous five-year average, according to state mortality data.  It’s “hard to guess as to what could be the case or the cause” for a lack of an overall surge, though several factors have probably contributed, Francis said.

One factor to account for the absence of the predicted increase in deaths this year is the fact that epidemiological models were very misleading.  Based on modeling data, in early April, Gov. DeWine had predicted as many as 60,000 new cases in Ohio per day!  However, at the COVID-19 peak in Ohio, around April 19, less than 1,300 new cases were reported per day.  It is hard to account for this discrepancy (prediction was off by a factor of 46-to-1) based entirely on good public compliance to social distancing that “flattened the curve.”  Here is where we need “good science.”

“Good Science” Needed
I have been particularly interested in how the scientific data on COVID-19 is obtained and interpreted.  See “Pandemic—or Policy without “Good Science?” Our example article from Ashland Source refers to the basic reproduction number (or reproductive ratio, R_o) of a virus.  Please don’t panic here when you see the symbol, R_o. What follows is not beyond your ability to understand if you read slowly and stay with me.

In simply terms, R_o is the number of people that one infected person will infect in a population.  So, if one person develops the infection and passes it on to two others, the R₀ is 2.  The “0,” or naught, indicates a reproductive number (R) for a new virus in a population with zero immunity. Jeffrey K. Aronson, et al. of the Centre for Evidence-Based Medicine , University of Oxford, explain in more detailed terms that R_o is “the number of cases that are expected to occur on average in a homogeneous population as a result of infection by a single individual, when the population is susceptible at the start of an epidemic, before widespread immunity starts to develop and before any attempt has been made at immunization.”

The article refers to estimates of R_o reported by Ohio Governor Mike DeWine.  According to the governor, “Ohio’s R₀ number was as high as 1.9 in March but as of June 1, it was slightly lower than 1.”  Gov. DeWine has attributed the decrease to the interventions implemented by his administration to “flatten the curve.”  Ohio’s Region 5, which includes Ashland currently has an R₀ of 0.72.

By this time, you may realize that, by definition, R_o applies only at the start of viral infection of a population.  In contrast, the  effective reproductive number, is the number of people that can be infected at a given time during the progression of the virus in a given population.
The R_e “changes as the population becomes increasingly immunized, either by individual acquired immunity following infection or by vaccination, and also as people die.”  Gov. DeWine and the ACHD seem to be misusing the R_o value (applies only at the start of viral infection of a population).  As noted in the previous paragraph, where Gov. DeWine reports changes in R_o, in all likelihood he is referring to R_e.

The estimated R_o for COVID-19 (or SARS-CoV-2) is given in the adjacent figure from CEBM.  Note that COVID-19 to date is not significantly more deadly than several prior influenza viruses.  Yet, because we adopted the policy of broad separation and economic lockdown, this pandemic may end up being much more devastating because of economic disruption, deaths from delayed health procedures, suicides, and drug abuse.

We should note that estimates of R_o, are based on complex mathematical modeling which incorporates estimates of susceptibility, infectivity, and rate of removal of individuals by either recovery through acquired immunity or through death.  As I have already stated, considerable variability has occurred among the epidemiological models largely because they are predictive and based on limited data.  Some of the modeling error may be attributed to the unexpected compliance of our culture to social distancing guidelines.  But there is still reason to doubt their accuracy.  George Box, the noted British statistician, has quipped: “Models, of course, are never true, but fortunately it is only necessary that they be useful. For this it is usually needful only that they not be grossly wrong.”  See “Pandemic—or Policy without “Good Science?”

In conclusion, I have noted the importance of good journalism, good science, and an educated population that can think critically to read and analyze articles such as the one from Ashland Source on the topic of the COVID-19 pandemic in Ashland Co., Ohio.  So far in Ashland Co., there have been no deaths and only 27 reported cases as compared to over 5,000 cases and 303 deaths in Cuyahoga Co. (Cleveland).  Obviously, reported cases and deaths vary greatly among different Ohio counties depending on population density.  However, in no case did number of cases and deaths reach those predicted by the models used to justify the total lockdown. Today, a more recent report in Ashland Source hints that the county's low number of cases (28 as of today) and their very well executed Continuity of Operation Plan (COOP).

Nationwide, it remains to be seen whether broad isolation instead of protection of vulnerable individuals was worth the great cost of economic disruption, unemployment, interruption of routine health care, and termination of most social and cultural events, classroom education, and spiritual gatherings. 

It is not uncommon that both toys and news stories come with “batteries not included.” But let us pray that our nation doesn’t suffer from news reporting that is poorly communicated, lacking in “good science,” and even biased to serve dishonest political or economic gain.

More Questions to Consider:
1.  Can you see how important it is that we exercise caution and think critically, especially when the journalism sometimes lacks sufficient accuracy, context, and helpful assistance to critical reasoning on the part its readers. 
2.  Ohio residents, how many COVID-19 cases have currently been recorded for your county.  You can go to various sources such as Akron Beacon Journal.  Here, scroll down to the Ohio map and click on your county for statistics.  Readers from outside Ohio can find similar data at the Coronavirus Resource Center under your state.
3.  How do deaths from COVID-19 compare to other causes of death?  You can look up a listing of DEATHS BY DIFFERENT CAUSES IN YOUR STATE.  Then, observe how the average deaths from each cause for the past few years compares to the deaths in 2020 (adjusted for the fact that we are only about six months into 2020).  Particularly, how do COVID-19 deaths compare to deaths from other influenza viruses?