Another Increasingly Bad Covid-19 Week in Kentucky. Is it Grim Yet?

There is not much positive to say about last Saturday’s KHPI end-of-week update. New reported daily cases set new daily record highs four days in a row. Discounting the approximately 1500 Fayette County cases that had been unreported prior to last week, a new weekly high was also reached. This rise in new reported cases can also be visualized by inspection of the 14-Day rolling average of daily cases– also at a new high. (The 7-Day rolling average of new cases was so distorted by low reporting over the 3-day holiday weekend and the bolus of 1500 delayed cases, that this previous helpful metric was invalid.) Reported deaths last week reached the highest levels since the week ending September 12. That latter week reflected new infections from July suggesting we should anticipate increasing deaths in the coming weeks independent of additional new cases. Confirming that the increases in new cases and deaths are real, reported hospital and ICU admissions for Covid-19 disease are at their highest since the beginning of the epidemic. All of these markers of disease spread and impact are going in the same (wrong) direction.

You can inspect a full panel of interactive data visualizations on KHPI’s Covid-19 Update portfolio which is generally updated daily on Tableau Public.

New reported cases of Covid-19 through 10/17/20.
Weekly reported cases of Covid-19 through 10/17/20.
7-Day and 14-Day rolling averages of daily new cases of Covid-19 through 10/17/20.
Weekly reported deaths of Covid-19 through 10/17/20.
Hospital and ICU utilization by cases of Covid-19 through 10/17/20.

I cannot bring myself to simply repeat all of my earlier musings. We are not doing very well despite the best honest efforts of Governor Beshear and his administration. It doesn’t help that so many elected state and federal (almost entirely) Republican party officials are doing their best to torpedo sound health policy approaches to combating our current lethal epidemic. As a former state and federal lobbyist, I learned early the maxim that the only thing better than doing something effective is to make your opponent look bad. Through my lens as a physician and medical scientist, I can only appreciate political selfishness.

What will it take to turn this ship around? Those seeking political advantage attempt to convince us that the application of public health measures means the same thing as comprehensive lockdowns. That is manifestly not the case. As a community we must be willing to tackle the spread of disease locally where we live and to accept that no one’s personal beliefs place them above the law with respect to their actions.

Today is the first day that Frankfort will report the percent-test-positive rate in a new way. The justification of this change seems very reasonable to me. I will address this topic additionally at a later time.

Peter Hasselbacher, MD
Emeritus Professor of Medicine, UofL
October 19, 2020

Another Bad Week in Kentucky’s Covid-19 Epidemic Compounded by a Major Reporting Failure.

We were given plenty of warning that this week was going to be the worst yet of our first 32 weeks of Covid in Kentucky. It was– by far. With 7444 new confirmed cases, we appeared to leapfrog over the 5965 cases of the previous weekly high that ended October 3d. The height of the jump was artifactually magnified by an unexpected and inexcusably delayed reporting of at least 1472 old cases from Fayette County on October 7. Perhaps this was related to the large numbers of positive tests among students at the University of Kentucky. I am unaware of how far back these cold cases stretched. On the one hand, these strays confuse the significance of the new current state “spike.” On the other, depending on which days the backfill should have been reported, it means that rather than hovering about a possible new higher plateau that would have been the lesser of two evils, that we have been continuously escalating the rate of growth of Kentucky Covid for some weeks. That is not a comfortable possibility. The state deserves better. If Fayette County with all its resources cannot report its cases to Frankfort in a timely and accurate manner, what confidence can we have that the 119 other Kentucky counties, distributed as they are among an assortment of semi-independent public health departments, are not also having difficulties in managing their own local information?

I do not have the information needed to venture much more commentary. I am not aware such has been made public. I offer below a limited assortment of KHPI data visualizations. The full workbook of Covid-19 Tracking graphics is available on KHPI’s Tableau Public website, along with a separate workbook of individual county Covid-19 epidemiologic curves as described in the most recent articles in this series. The following overview graphic shows as well as the more complex visualizations the relentlessly increasing impact of this virus.

Overview of Confirmed Cases and Deaths. Covid-19 in Kentucky 10-10-20.

The following are bar charts of daily new confirmed cases, and of the weekly cases back to March.

Daily confirmed cases of Covid-19 in Kentucky as of 10/10/20.
Weekly confirmed cases of Covid-19 in Kentucky as of 10/10/20.

This plot of both 7-Day and 14-Day rolling averages of new cases makes clear the dramatic rise in the rate of growth of the epidemic in October. Even before the October 7th Fayette report, the 7-Day average was rising faster than the 14-Day average signifying an expansion of the rate of new infections.

7-Day and 14-Day rolling averages of new confirmed cases of Covid-19 in Kentucky as of 10/10/20.

That our state-wide communities are actually getting sicker is supported by the most recent reports of hospital and ICU utilization for Covid-19 patients. We are approaching new highs, but these hospitalizations reflect new infections of around one month earlier. We probably have not yet begun to see the effect of October’s new cases.

Covid-19 cases in KY hospitals and ICUs as of 10/10/20.

Kentuckians– we are our own best hope!
In a recent pair of articles in this series, I was able to document that the Covid-19 virus is alive, well, and reproducing in every county of the Commonwealth. An excuse not to wear masks or follow other basic public health measures because the epidemic is a problem only for somebody else as in big cities is no longer rational. I would like to believe that the inhabitants of every county, indeed every community of any size throughout the Commonwealth would feel some responsibility to take care of each other. Without such a sense of neighborly good will, I can only assume the worst is yet to come.

Peter Hasselbacher, MD
President, KHPI
Emeritus Professor of Medicine, Uofl
October 11, 2020

Serial Map Display of Spread of Covid-19 Across Kentucky.

I should be no surprise to readers that Covid-19 has entered a new and rapidly expanding phase of its presence in Kentucky. It is no longer (and never has been) solely a threat to the larger urban centers of the Commonwealth. Lack of a national strategy, conflicting advice, and poor example from some political leaders has made matters much worse than they had to be. I have been updating KHPI’s Covid-19 tracking page daily and will close out the past week of new record highs with additional commentary Saturday evening.

Watch Covid-19 spread ‘live.’
I want to take this occasion to augment yesterday’s offering to test a newly learned capability of the Tableau software I have been using for its powerful database, statistical, and data visualization capabilities. Using the New York Times Covid-19 County Database, I previously offered static maps that displayed the aggregate number of cases and deaths in each of Kentucky’s 120 counties at a single point in time.

What I have now been able to do is assemble a “filmstrip” of daily maps with a map for each of the 216 days since Covid-19 was first recognized in the state. The result is posted on the KHPI Tableau Public website. Using the date slider control in the panel on the right, the viewer can select any date interactively. There are also buttons (Back-Stop-Forward) to automatically transition one way or the other, but unfortunately on this browser-based platform, the rate of change is rather slow. (The image below is for illustration purposes only. Not a link.)

Control Panel for animated KY County maps.

An alternate way to view the workbook is available for those with Macs or PCs. At the bottom of the on-line workbook visualization are buttons for full-screen viewing, but also a ‘download’ option that I have enabled allowing the user to download the entire workbook both as it stands now and also after I add some additional tables to it. To view the packaged Tableau Workbook, the user will need to install the free application, Tableau Reader. (Tableau is a major international platform widely used by business, journalism, government, academia, and folks like me.) This will show additional buttons that allow the speed of transitions to be adjusted and offers the capability to investigate the underlying data more closely. I can use this download function to share this and other information of interest to Kentuckians more widely.

Frame from animated filmstrip of Covid-19 expansion in Kentucky. Image above is not an active link.

Let me know how this works for you. It is an experiment that I hope will be useful.


Peter Hasselbacher, MD
Kentucky Health Policy Institute
9 October 2020

Where Have Kentucky’s Covid-19 Cases Been Coming From?

Short Answer: From every single county and at similar rates!

From the first recognized case of Covid-19 disease in Kentucky on March 6 in Fayette County, our epidemic was most easily recognized in larger population centers where there were both more people and access. However, as early as April 1st, 69 of Kentucky’s 120 counties had at least 1 diagnosed case among its people. Only in 35 of those counties were there more than 2 cases. Only in 11 counties were there more than 10 cases. In one short month later on May 1st, there were only 6 counties without a recognized and reported case! By June 2d, only tiny Robertson County was officially reported as free of the disease. By July 1st, the disease was recognized in every county. The last holdouts were centered in the central/eastern part of the state. Of course, just as is true today, unrecognized cases of Covid-19 infection walked among us well before their presence was documented with a specific lab test.

The reader is invited to interactively explore for themself the tables, graphs, and maps that serve as the source of the material presented below. Filters in some of the visualizations allow the user to select a specific county, date, or range of dates to view the data. I will continue to tinker with and add additional visualizations in the online version.

Epidemiology curves display the spread graphically.
The first graph below depicts the total number of “cases” as of a given date for the Commonwealth as a whole. From numbers too small to show up on the plot, total cases rose at a very gradually increasing rate until an inflection point around the July 4th holiday when counts took off at a rate that shows no sign of stabilizing. Indeed, the rate may well be at a new and steeper inflection point as I write this. (See the footnote at the end regarding to the definitions of “Case”)

Cumulative case totals for all 120 KY Counties as of 10/3/20.
Continue reading “Where Have Kentucky’s Covid-19 Cases Been Coming From?”