COVID-19 - Assessing the Data
Have we ever have ever experienced a worldwide data driven event like the COVID-19 situation unfolding in real time across our 24 hour news cycle and social media? Like most of you, I have been tracking COVID-19 on the following websites: John Hopkins, Centers for Disease Control and Prevention and the World Health Organization. I've lost track of time spent gawking at the news, researching articles and discussing with colleagues. The communicated data ranges from downright scary to ridiculously inaccurate:
A CDC report on March 13, 2020 states, "given the residual uncertainties, health sector decision-makers and disease modelers probably should consider a broad range of 0.25%–3.0% for COVID-19 case-fatality risk estimates".
The Ohio Department of Health believes 100,000 Ohioans are carrying coronavirus (3/12/20) - a staggering (inflated?) estimate as Ohio currently has 704 reported cases (3/26/20).
Trying to form a solid opinion is challenging as it is hard to consolidate and process all the information. I analyzed the John Hopkins data set at 3:13pm EST on 3/16/2020 and 10:02am EST on 3/26/2020. I am not an epidemiologist, nor do I have a public health background so I won't be making any proclamations on infection or mortality rates (if I happen to - please immediately assume they are incorrect).
John Hopkins Data
Below are the top 25 countries with the highest number of confirmed cases sorted from high to low. Case Fatality % measures the % of deaths of currently confirmed cases (Deaths / Confirmed).
A few observations:
Most concerning is the global case fatality % has increased from 3.95% to 4.51% in the past 10 days. Admittedly, I thought this rate would fall as more patients were tested.
Italy played a big part in driving up the global case fatality %. After adjusting for Italy, the rate drops to a range of 3.26% - 3.52%.
Overall confirmed cases increased about 175% from 179,103 to 491,623 (or an average of 34,000 new cases a day)
The below schedule shows the increase in confirmed cases by the top 15 countries (by number of cases). The jump in U.S. cases really stands out - as we are doubling roughly every 60 hours or 2.5 days. China cases have slowed to 75 a day and South Korea's 12% increase is notable as well.
I will make a projection that I believe is accurate - the United States' rate of infection won't continue! I ran the numbers and assumed the U.S. rate of confirmed cases would double every 3 days (see below). How do I know I'm correct? Per the U.S. Census, the current U.S. population is 329,438,793 (but it is increasing by the minute). By the way the number on 12/30/20 is a nonillion (I have plenty of time in my house now to do the extra research).
Below is the case fatality % for the top 15 countries (based on confirmed cases). Shaded in yellow are countries that had an increase of greater than 2% in its case fatality %. I found the difference between Germany and Italy's case fatality rate incredible (0.6% vs. 10.1%) and a perfect example of why you can't make conclusions using partial data sets.
Germany and Italy - A Deeper Dive
The CDC has reported that older adults and people of any age who have serious underlying medical conditions might be at higher risk for severe illness from COVID-19. Based on that, it is logical to make the conclusion that Italy's population has a larger elderly demographic than Germany. We will find that is a false conclusion.
Below is a chart showing Europe's population age structure by country (a bit of an eye chart so I've added a link to the site). Italy (if you squint hard) has the highest % of its population (compared to all other countries) over the age of 65 (22.6%). However, Germany is 3rd with 21.4% of its population over the age of 65. Germany and Italy has similar demographics from a statistical standpoint.
The chart below portrays Italy's death rate by age demographic:
Maybe respiratory disease is a bigger issue in Italy than Germany (or other EU countries). The chart below suggests that's not the case either as Italy's standardized death rates related to diseases of the respiratory system compares favorably to Germany (and the EU-28 average as well). The difference could be due to quality of care, bad luck that more high risk patients were infected or something different altogether. The early results show that it may be just bad luck or Germany did a good strong isolating their most at risk citizens as the average age of an infected patient in Germany is 47 years old vs. 63 years in Italy.
Does Gender Matter?
Several reasons for the disparity are being studied by researchers (taken from the Guardian):
Early on, smoking was suggested as a likely explanation. In China, nearly 50% of men but only about 2% of women smoke, and so underlying differences in lung health were assumed to contribute to men suffering worse symptoms and outcomes.
Some studies have shown that men are less likely to wash their hands (Yuck!), less likely to use soap, less likely to seek medical care and more likely to ignore public health advice.
Male immune systems may not initiate an appropriate response when it initially sees the virus.
Could be a combination of all the above or some other reason that is uncovered once we have a full data set after the 2019 COVID-19 pandemic is over.
We know that ensuring patient safety and maintaining data integrity are critical to conducting a successful clinical trial. The COVID-19 pandemic is causing disruption for clinical trials around the globe as many sites are closed or not enrolling new patients to ensure their safety (from COVID-19 infection). CROs and sponsors are collaborating in real time to amend project plans and put forward strategies to progress studies and keep patients safe.
Healthcare professionals around may be living out the largest "clinical" study of the 21st century. Physicians, nurses, et al are working around the clock to ensure patient safety, while epidemiologists are completing daily cuts of the data. Unfortunately, there is too much noise in the data to make valid conclusions.
Differing patient populations
Differing patient demographics
Quality of health care
Pandemic preparedness by country
Country data in the early stages vs more mature stages of the infection curve
Plus too many others to list
The world's population is roughly 7.8 billion strong, which means over 7.7 billion of us are not confirmed cases as of this writing. How the remaining 7.7 billion of us carry ourselves during this crisis will determine how quickly we flatten the curve, save lives and return to a normal state (the term COVID-19 beard is starting to spread inside Clinipace and I am guilty of being a participant). Right now, we should all follow the guidelines being communicated by the medical community and local authorities - no exceptions. Also accept that until we have a more robust data set on this disease, we will continue to have unanswered questions and likely more surprises in our future. Everyone take care, be vigilant and keep an eye out for the elderly and those with pre-existing conditions during this challenging time. I wish everyone great health as we win this battle!
Jason Monteleone is CEO of Clinipace & President at Pivotal Financial Consulting, LLC. Clinipace is a global mid-size CRO with operations in the Americas, Europe and Asia-Pac serving small and mid-size pharma and biotech sponsors. Pivotal provides Divestiture Assistance, Acquisition Advisory Services and Strategic Planning to the Pharmaceutical Outsourcing Industry. Jason can be reached at firstname.lastname@example.org, email@example.com. Follow me on Twitter @JMPivotal. Sign up for Jason's latest blogs and updates at www.pivotalfinancialconsulting.com.