Possible answer: When new cases and the Percent Test Positivity Rate progressively increase for 12 days or more.
I do not have much more to add to my comments of November 13 but will allow the data to speak for itself. Yesterday’s announced 2195 new cases is the highest daily count since October 14. Yesterday’s case count was higher than any of the previous 4 Wednesdays. Given that this week’s 3 remaining weekdays through Saturday comprise those with historically higher counts, it can be projected that the total number of new cases this week will exceed those of the previous three or even four weeks. Since the number of daily tests has been fairly stable, increased testing does not explain the increase in case numbers but the progressively increasing Test Positivity Rate helps. Hospital utilization is creeping up and along with ICU and ventilator utilization and deaths will eventually follow wherever new cases go. Elsewhere in the world and in some other US states, that place is another winter surge.
I am personally being very careful to follow good public health measures including getting a booster vaccination, wearing a mask indoors in public, and limiting my exposure to groups. I recommend the same to everyone else.
Since mid-October there were suggestions, and then evidence, that the rapid decline in new cases of Covid-19 in Kentucky that began in September was beginning to slow towards a potential “plateau” in the shape of the epidemic curve. By early November, there were suggestions that Kentucky’s epidemic was actually getting worse. As of yesterday (Friday), the number of new cases reported supports that an expansion is real– at least in the short term. Kentucky is not alone among the states or nations in which epidemic surges are occurring. We in the Commonhealth are not immune.
Increase in Weekly Total Cases. KHPI defines a week as the days from Sunday to Saturday and new cases as the daily increment in total cases. Last week’s total case-counts include those of Sunday, Monday, and a federal holiday– days for which counts have been historically lower due to issues of data collection. Yesterday’s final count for the week-ending Saturday, November 13 will not be available until next Monday evening. Even without Saturday’s count, the total number of new cases for the first 6 days of last week is only 253 fewer than the previous week’s. There is no doubt that with the addition of today’s cases, the total last week will exceed that of the previous week by several hundred. A reversal in the decline of weekly case counts has not occurred for the previous 10 weeks since its peak of the fall surge.
New Cases on a Given Weekday Higher Than in Previous Weeks. Last Monday’s count was higher than the previous 2 Mondays; Tuesday’s and Wednesday’s than the previous 3 weeks; and Thursday’s (Memorial Day) was greater only than the previous week. Friday’s new case count was lower than the previous week’s, but it was post-holiday.
Both 7-Day and 14-Day Rolling Averages of New Cases Are Rising. The tremendous volatility in day-to-day counts of cases and deaths plays havoc with our ability to plot our course in the short run. For this reason, virtually all trackers of the epidemic nationally rely on averaging counts over a week or more of days to better judge longer term trends. When the 7-Day average is rising faster than the 14-Day one, the epidemic can be said to be expanding. This has been the case in Kentucky since November 9.
Semi-log-Plot of New Daily Cases No Longer Falling. When dealing with exponentially changing numbers it can be tricky to judge the rate and direction of change to allow insight into the future. One solution is to is to plot the logarithm of case numbers against time in Semi-log plot. I have found this method to be a useful tool in this series of articles. When the linear trend line in such a plot slopes in a downward direction, it can be said that the epidemic is slowing down. A horizontal trend line indicates a steady state. At best, since mid-October that is where we have been. However, a case-count in excess of 1000 per day is not a happy place to be.
Test Positivity Rate (TPR) Has Been Rising. If no testing were being done, many fewer cases would be identified and reported. There has been a trend for fewer tests over the past two months, but testing remains at a high level. Kentucky calculates a 7-Day average of the both number of tests performed and the percent of those reported electronically that are positive for the virus. The TPR is one indicator of the prevalence of infection in the community and has tracked closely with both new cases and hospital utilization. Since November 1, a progressively falling TPR has been rising in parallel with the 7-Day case-rate which most recently is 5.53%.
Summary of New Case Indicators. I began this series of articles in March 2020 seeking a way to best understand visually our path through what has become by any measure a global catastrophe. I believe the measures I have been using correlate well with those of both Kentucky and the nation. Predictions to which the data have led me have by-and-large been on target. These current indicators lead me to conclude that for the past two weeks the direction of our local epidemic has changed in the direction of expansion. The most favorable alternative interpretation is that we have entered a steady-state situation of endemic infection albeit at a troublingly high level.
Other Indicators. It is also a little disconcerting to see that over the past 4 days and for the first time in many weeks, the number of people currently in-hospital or ICU has been ticking up. Also rising progressively towards its previous peak in early September is the percent of new cases in school-age individuals 18 years old or younger which is now at a 7-Day average of 29.6% of total cases. As expected, the 7-Day average of deaths reported daily is now at 30 and is certain to rise several-fold, the consequence of September and October’s infections.
To Conclude. My prayer is that the apparent current reversal in epidemic indicators summarized above is a fluke. Our public health reporting systems here and nationally have certainly revealed to us their limitations. They beg to be improved to allow us to deal in a more timely and effective way to the inevitable next epidemic that turns our lives upside-down. My hope is that we Kentuckians and Americans will believe, as I do, that we as a community are no healthier than the sickest among us, and that in our interconnected world diseases like Covid that are anywhere will be everywhere.
Over the past century or more, public health measures have done more to improve our health and prolong our lives than pills or surgeries. Therefore stay healthy!
Peter Hasselbacher, MD Emeritus Professor of Medicine, UofL Saturday, November 13, 2021
[Addendum added November 6, 2021: Yesterday’s (Friday) reported Kentucky Covid-19 data was the last such available until Monday evening. There were 1,421 new cases, 53 new deaths, and 726 individuals currently in the hospital. Kentucky’s 7-Day calculated Test Positivity Rate increased to 5.19%. On Friday, 30% of all new reported cases were in people 18 years old or younger. The current hospital census continues to decrease. These new numbers do not change the tentative conclusions offered in the article below. While the number of new cases was less than for recent Fridays, is seems more certain that the rate of decrease is declining. While there may be other reasons for this, the increase in Test Positivity Rate offers a potential less benign explanation! KHPI’s Tableau Public website has been updated.]
As of Thursday evening, the number of new cases of Covid-19 in Kentucky continues to decline but at a lower velocity. Wednesday’s count of 1614 new cases was only 59 fewer than the previous Wednesday’s. The seven-day average of new cases was 1191, but a steady-state above 1000 per day may approaching. The reported KY test positivity rate is no longer falling as it was. According to other reports, this is a trend that is occurring nationally.
Deaths are now beginning to decline predictably with a one-month delay following the peak of new cases during the first half of September. The current 7-day average of reported deaths is 24 per day.
Hospital and its closely related ICU bed utilization continue to fall from their record high peaks on September 10 but remain significantly elevated.
The semi-log plots of both new cases daily and their 14-day average are not trending downwards as rapidly as they were a month ago. (A horizontal plotline indicates a steady state of new cases.)
The number of new tests per day has been slowly drafting downwards with a current 7-day average of 22,100 total reported tests per day.
The largest number of daily new cases has historically been reported on Thursdays and Fridays. Recall that it took a while for last winter’s surge to settle down to its nadir in June’s good weather. Individuals of school age under 19 still make up 27% of new reported cases. Winter is coming. Whether we have relaxed our vigilance too much remains to be seen. No one precaution, be it vaccination or a combination of traditional public health measures are perfect, but we ignore either approach as our communal peril.
Peter Hasselbacher, MD Emeritus Professor of Medicine, UofL Kentucky Health Policy Institute 4 November 2021
As of Wednesday, October 20, the number of new cases of Covid-19 and the test positivity rate continue to fall sharply. The 7-Day New Case number was 1590 and the 7-Day Test Positivity Rate 6.83. Both of these are about where we were last February when we were halfway down from winter’s peaks, but still considerably above last summer’s nadirs of 142 cases and 1.79 percent of tests.
Hospital bed and ICU utilization (which track each other closely) are also clearly easing but not to the same extent as new cases. For example, ICU bed occupancy this week is nearly as high as the peaks of December and January. (One could speculate that this may be connected to the enhanced virulence of the Delta variant or perhaps even better reporting.) If we continue to see fewer new cases at the same rate since September 8, we could be below 500 daily by mid-November.
We should expect to see the number of deaths reported daily to begin to fall also– but we are not there yet. A substantial proportion of individuals still on respirators will not survive despite optimal treatment. It takes weeks or months to clear out a backlog of reported death certificates. It will take many months to arrive at an estimate of the number of “hidden” Covid deaths due to long-term systemic damage caused by Covid-19, or to delayed or absent care for unrelated illnesses.
Altogether, this week continues the decidedly good news we have been seeing for the past 4 weeks. This does not imply that we can back off of the public health measures that brought us to this point. The majority of Kentucky counties are still in the red in terms of high-risk case incidence rates. As is the Game of Thrones, “Winter is Coming” and with it a season of indoor activities and respiratory illnesses. The big-time holidays of November and December will soon be upon us.
Schools will be in session. Since August 18, the 7-Day average of new cases of Covid-19 in individuals 18 years old or younger has usually exceeded 25%. The good news is that vaccinations are currently available for children 12 and up and we are on the brink of having a vaccine approved for children as young as 5. The rate of vaccinations in these groups remains to be seen, but schoolrooms are obviously mixing bowls where students, teachers, staff, and their families are exposed to Covid-19 infection.