October 28: Case count grows; record 7 day count for VA
Daily Status, October 28
One thing that is immensely personal to me is I need to avoid COVID-19. Like, I do not know my mortality risk exactly, but last week, I thought my Infection mortality rate was 5-10%. With active cancer, that doubles, to 10-20%. So if you see me in town, wear a mask. I try to avoid challenging people, but maybe it is time. If I can see your nostrils, you are not wearing it right. If I can see a gap to your mouth, you are not wearing it right. I try to avoid confrontation with these people, because it goes south quickly, but I will not support any business that allows other customers to but me at risk.
Big picture: There is little doubt that the increase in cases viewed in the Commonwealth is real. This may be the beginning of the fall COVID-19 surge we were warned about.
Local: The caseload in Vienna doubled in the last month. Be careful and wear your mask.
In the last two days, VA reported 2,479 new cases of COVID-19 (1,134 on Tues and 1,345 today). The data were reported by VDH on October 28, 2020 were within one standard deviation above the three-week bias-adjusted two-day average of 2,305 cases. We had been seeing a caseload increase, but there is some indication that the trend has flattened which could be due to reporting issues. Based on prior experience, I am inclined to suspect the latter.
Viewed over the last seven days, VA reported 7,982 cases, or 1,140 cases per day, which works out to 13.7 cases/100K people. This is the highest verified weekly number during the pandemic, though it is likely the May numbers were significantly under-reported by a factor of four. There was a dip in hospitalizations that coincided with the September dip, leading me to believe that this reflected case count and not reporting issues. Furthermore, if we subtract the effect of the college outbreaks, there was a constant decrease from just before August 1st to Sept 25th. By October 3rd, there was a measurable increase.
The trends over the last three weeks now are showing a minimal increase at 0.6% per day or 4% per week.
By combining our current regional trends with the typical reporting for the day of the week, I expect about 1,237 cases tomorrow (Monday) with a 90% chance of the numbers falling between 946 to 1,619 cases. If tomorrow's numbers differ from that, we will have a very rare event. This is useful to consider if I am unable to report tomorrow, as you can compare the state reported numbers with this.
The testing numbers now show the percent positive to be just above the the 5% metric over the last week (5.28%) which is often used to indicate sufficient testing (e.g., it is safe to reopen schools). VA is testing about 2.8% of the population every two weeks
We will remain on NY’s quarantine list for at least one more week, and probably longer because we were well above their quarantine restrictions (10 cases per 100,000.) Over the last week, we were well above 10 cases per 100,000 to enter states like NY without quarantine—we are at nearly 13.7 cases per 100k.
When we look at the local ZIP code data, we see that the observed increases are almost universal across VA. I am comparing the current estimated % positive to that of one month ago. Note that almost all is a warmer color (further from blue and closer to yellow). This is an indication of the uniformity of the increase. As a practical matter, it means to be safe and careful, no matter where in VA you live, there is risk.
In the spring, COVID-19 in VA was primarily a concern in the DC suburbs. Over three to four weeks, (from late May to early June, NOVA recovered and for about a month the disease was under control to the point that restrictions were eased. Unfortunately, in eastern VA/Hampton Roads, the easing of restrictions resulted in a surge in cases which peaked just before August 1st resulting in stricter restrictions in that area. Since then, with the exception of growth on college campuses, the disease has been stable, excluding the rural parts of the state where safeguards (social distancing and masks) are largely ignored. Starting in October our weekly case count has been increasing throughout the Commonwealth, particularly in NOVA, and SWVA.
Looking at the weekly case count, we see that the numbers are more in line with where we were at the beginning of August.
Regional growth rates (in fraction per day) continue to show degradation over the past three weeks. It is easier to show a decline when the prior numbers increased. The current growth rates for the different regions are shown below.
The entire state is increasing with Rt=1.005
The following table shows the number per 100K for each region. Again, NOVA and Eastern VA are doing the best, and the mountainous regions in NW & SW continue having more cases. The concerning aspect is that in all regions the numbers for last week are significantly higher than the preceding three weeks. What is most concerning is that SW & NW VA are hitting significantly higher caseloads, in spite of VA Tech and Radford doing better.
The individual line charts show the unfiltered data per day, coupled with the trend lines.
The trend lines show the different periods of growth.
Early in the pandemic, the different parts of VA were functioning largely independently, with NOVA mimicking the northern states, and Hampton Roads mimicking the southern states. Since September 1, the regions have trended together. Starting in late September, NoVA and SWVA diverged from the rest of the state, a trend that continues today.
Currently SWVA has the greatest number of cases even though they have half the population of NOVA.
Note that the effects at both ends of the chart are probably artifacts of the (seven-day polynomial filtering I use for averaging); the filter is poorly constrained in the first and last few days of the time history. For those technically inclined, the filter is called a Savitzky-Golay filter, basically a moving window polynomial filter. At the edge (first and last days of the time series), the filter will over-compensate for the trend as it is unconstrained. I recommend the Wikipedia article if anyone is interested in more information, or contact me.
After the early peak in May (~1000 cases per day), NOVA saw a sharp drop in all COVID metrics, reaching a broad valley in mid-June (~200 cases/day), which lasted until around August 1st. By Sept 1, NOVA increased to 300. September showed notable caseload drops to about 150 by late in the month. Since then we have had a steady increase averaging about 250+ per day.
Fairfax Co.: 1.006
Arlington Co.: 1.014
City of Alexandria: 1.012
Prince William Co.: 1.001
Loudoun Co.: 0.996
The number above is Rt: Rt is an exponential time constant, where the number of cases in a time segment is approximately, n=Ao Rt ^ t, where Ao is the number of cases at the start of the segment, Rt is the exponential growth rate, and t is the number of days since the start of the segment. So, if Rt is greater than 1, it is growing exponentially, if it is less than one, it is decreasing each day.
The difference in the colors (contrast) in the NOVA map is increasing. In addition, the NOVA map is warming (as is happening throughout the Commonwealth). At this point, it seems likely that this is related to the fall surge others had predicted.
Most localities in NOVA have case counts near or above 10/100K/day. In Vienna, for example, we were under five in late September but are now at 9.1/100K/day.
Growth rate (%/day)
Recently, the trends show two groups of rates for Fairfax County: McLean is doing the best. All the other locations are grouped at a higher number -- almost every region is seeing small upticks in the last few weeks.
In Vienna the average daily case count has doubled ; I do not why (I have not been out and about). I know that youth sports are active although I have no reason to expect that is causing the increase.
I am attempting a different means of discussing risk. Risk is very personal-- people will respond to COVID differently. However, we know how many people have died (and how many cases) for each age group in VA. So, we can get a case fatality rate for age groups. Also, based on serology studies, we know the case numbers are low by a factor of 2.4, so we can get an infection fatality rate (IFR). We also know how comorbidities play in.
In the following tables, I have combined the current probability of a person being infected around Vienna to compute the probability that a person in a specific crowd is infected.
Assuming one interacts with everyone in the crowd (big assumption), I assume if you interact with an infected person, you have a 50% chance of getting infected. I do not know what that rate is; it will be a function of how long you interact, how close, and if masks are used, etc. This is the big unknown.
That gives me a probability of being infected based on the number of interactions. I then combine that with the IFR to estimate the risk of dying by age and number of interactions for people with and without co-morbidities.
For comparison, the risk of dying in a car accident in per day is about one in three million. That is a baseline but none of us likel knows people who will not get in a car because of the risk of dying.
1 0.00e+00 3.14e-08 2.52e-07 1.10e-06 2.86e-06 7.92e-06 1.81e-05 2.89e-05 5.44e-05
5 0.00e+00 1.57e-07 1.26e-06 5.49e-06 1.43e-05 3.95e-05 9.02e-05 1.44e-04 2.72e-04
15 0.00e+00 4.68e-07 3.74e-06 1.64e-05 4.26e-05 1.18e-04 2.69e-04 4.30e-04 8.10e-04
50 0.00e+00 1.53e-06 1.22e-05 5.35e-05 1.39e-04 3.85e-04 8.79e-04 1.41e-03 2.65e-03
100 0.00e+00 2.97e-06 2.38e-05 1.04e-04 2.71e-04 7.49e-04 1.71e-03 2.73e-03 5.15e-03
500 0.00e+00 1.20e-05 9.58e-05 4.19e-04 1.09e-03 3.02e-03 6.88e-03 1.10e-02 2.07e-02
1000 0.00e+00 1.87e-05 1.50e-04 6.55e-04 1.70e-03 4.72e-03 1.08e-02 1.72e-02 3.24e-02
2000 0.00e+00 2.47e-05 1.97e-04 8.63e-04 2.24e-03 6.22e-03 1.42e-02 2.27e-02 4.27e-02
1 0.00e+00 3.14e-07 2.52e-06 1.10e-05 2.29e-05 3.96e-05 5.42e-05 5.78e-05 1.09e-04
5 0.00e+00 1.57e-06 1.26e-05 5.49e-05 1.14e-04 1.98e-04 2.71e-04 2.88e-04 5.43e-04
15 0.00e+00 4.68e-06 3.74e-05 1.64e-04 3.41e-04 5.90e-04 8.07e-04 8.60e-04 1.62e-03
50 0.00e+00 1.53e-05 1.22e-04 5.35e-04 1.11e-03 1.93e-03 2.64e-03 2.81e-03 5.29e-03
100 0.00e+00 2.97e-05 2.38e-04 1.04e-03 2.16e-03 3.75e-03 5.13e-03 5.46e-03 1.03e-02
500 0.00e+00 1.20e-04 9.58e-04 4.19e-03 8.72e-03 1.51e-02 2.07e-02 2.20e-02 4.15e-02
1000 0.00e+00 1.87e-04 1.50e-03 6.55e-03 1.36e-02 2.36e-02 3.23e-02 3.44e-02 6.48e-02
2000 0.00e+00 2.47e-04 1.97e-03 8.63e-03 1.80e-02 3.11e-02 4.25e-02 4.53e-02 8.54e-02
I am not updating this section for the time being, except for the charts. I will leave it here as is for a while longer, as at times it can be very interesting. This is particularly so when specific age groups do not follow other groups. For example, teens and 20-somethings surged in early September while the other age groups did not, due to the outbreaks at colleges.
Overview: Things are static at most colleges. My data mining by ZIP code requires that I group them together when they share a ZIP code.
My process combines the VA Department of Health data and what is reported by the colleges. The assumption is that students feeling ill are going to seek health services; however, we have all heard anecdotal reports of people not doing that because they did not want to quarantine. The colleges usually update the dashboards after this post, or on Monday/Tuesdays. Except for Radford, all weekly updates are complete at this time and the numbers are encouraging (except for UVA).
As the process has evolved, it was clear that I needed to show active cases (e.g., ten days) in addition to cumulative cases.
Also, I am assuming the colleges are promptly reporting their numbers to VDH. As it turns out, W&M is reporting the cases approximately two days after updating the dashboard.
Note: I have been finetuning the following table to improve the estimated number of cases (the order of operations has changed a bit). Also, I am adjusting the pre-student caseload: to estimate the number of cases from students, I subtract out the average number of cases prior to the student arrivals from the reported cases. If the numbers are large (UVA, VA Tech, JMU & Radford), the impact is minimal. This makes it harder, though, to identify cases in more urban settings, like ODU, VCU and CNU.
RED means there is clear evidence for community spread
YELLOW means there may be community spread; still ambiguous
GREEN means no evidence of community spread
BLACK means they went online.
Last 10 days
% that had COVID
* Estimated from the number of cases in the ZIP codes associated with the university, removing the pre-student arrival case rate
** Estimated number of cases is an attempt to normalize for testing limitations. Specifically, I assume at 5% positive, 100% of the cases would be caught, so I normalize it to that value. If the % positive is very high (>40%) I am likely overestimating the numbers.
*** Dashboard cases are only counted if I can find the dashboard. In some cases, it is difficult to distinguish positive tests from cases (one case may have multiple positive tests; that occurs mostly at VA Tech). I include active cases if reported; otherwise, I use total cases.
**** % population uses the total reported number of students rather than just those on campus; it may be off when the percent positive is above >40%.
***** Old data, not updated for today.
****** Active cases, not total cases
Large Scale Community Spread:
VA TECH: VA Tech is finally getting a handle on COVID with only a few new cases a day. VA Tech is more than a college...it is a football team. And the football team has been decimated by COVID-19. The real test is when VA Tech can field a full team. On Sat, 10/17, they were down to 13 inactive players . I may promote it to yellow.
JMU: JMU returned to in-person classes 10/5; here is their plan. So far so good with the return. The case count at JMU continues to rise, but slowly. JMU has been fairly transparent with the situation, but could not get ahead of it initially. As it stands now, caseloads continue to smolder. They are neither increasing nor decreasing. My one significant concern is reporting: there are a lot more cases in Harrisonburg than there are new cases at JMU even though they are in the same city. This may be due to students failing to report, or it could be due to cases in the non-student population.
CNU -- I am not sure what is happening at CNU because they are in a populated area and their dashboard does not supply the data required to understand variation over time. They are showing a significant increase in caseloads and I will be monitoring them more closely now.
ODU -- ODU is now showing a significant increase in the percent positive which suggests there may not be a complete count; however, the numbers remain relatively low for the size of the university. It is difficult to track ODU because it is in a populated area and they only update the dashboard weekly.
UVA: With the students’ return, there has been a marked increase in cases, week after week. It appears with the much stricter rules UVA imposed two weeks ago that the caseload is abating. As a Hokie, nothing saddens me more than having to move UVA down on the watch list before I do VA Tech. Oh well. They still messed up COVID. There is that. 😊
RADFORD: Radford was moved back up primarily because of the % positive. Radford updates the dashboard weekly which is insufficient in my opinion. Fortunately, they report to VDH regularly (as mandated by law). While it is not clear how many students at Radford contracted COVID-19, it would not surprise me to see enough that the community, when isolated effectively, has herd immunity (meaning over 60-80% contracted COVID-19). The only way to know would be antibody studies. Hopefully, VDH will look into that. Currently, it seems that Radford has a manageable number of cases. While only 57% of the cases in the last two months are attributed to students, there were so few cases prior to the students’ return, this means either the dashboard is undercounting, or there were a lot of non-students infected by students.
William & Mary -- There is at least a small-scale outbreak in the athletic department. W&M tested all athletes and stood down on athletics after having had a total of 20 positives; However, W&M (and Williamsburg) have gone a full week without a positive case (and completed nearly 1,000 tests). Most of the active cases are among athletes. It is worth noting that W&M is quick to update the dashboard; they update prior to reporting to the VDH. W&M tested all students prior to arrival in town--with the current outbreak (albeit small), the college probably is a bigger threat to the town than the other way around which is a reversal of the last several weeks.
When I started talking about communities, the focus was on the safety for incoming students. Unfortunately, that concept has changed. Now we are seeing the colleges impacting the surrounding communities. If we look at the age distribution of cases in the communities of JMU, Radford and VA Tech (New River, and Central Shenandoah health districts), we see that, starting about four weeks ago, the number of cases for non-college age citizens has been increasing -- about two weeks behind the college-age curve. This suggests the disease is infecting the general population.
So far it looks like about an extra 372 middle-aged and senior citizens have been infected; there have been 12 cases to 24 cases per day in those groups. Furthermore, in the five months from the beginning of the epidemic to early September, there was an average of 10 deaths per month; in the last seven weeks there have been 35 deaths. This is concerning as the return of students has appeared to result in 21 extra deaths in the Commonwealth today,in just the New River and Central Shenandoah health districts. None of the additional deaths were of student age. This trend has been observed elsewhere in the country where the college students infect the more vulnerable populations.
Since Donald Trump has been infected, or more accurately, with the infection of Trump, we have learned:
1. People can test negative and still spread the disease.
2. Even if everyone who interacts with you is tested immediately prior to the interaction, you can still get sick (see 1).
3. If you have an event with hundreds of people, there is potential for it to be a super-spreader event, even if you test everyone at the event. If you do this frequently enough, you will get unlucky.
4. A non-spaced, but outside event is still dangerous.
This has two impacts for us: first, some of the attendees and support people from the White House live in our community. I have no information about what impact this may have had. Second, we now know that any crowded situation is a potential spreading event, even if outside. his includes sporting events, concerts, and worship. Please think about this before planning activities.
There are safety concerns with athletics, but those can be mitigated. I am more concerned about the fans in the stands.
I have reason to believe about 100 or more of the last week's VA cases are a direct result of the combination of the Newport News Trump rally and the White House outbreak.
1) You can repost/ share this information in its entirety by forwarding the entire link, or, 2) If you want to share partial content, you must receive my permission. This is proprietary information and I need to make sure you understand what I am saying. If anyone sees that this work being used without attribution, please let me know as soon as possible. I am willing to have an informed discussion/debate on my approach, but I want to make sure the proper context is captured.