Sure. There are uncontrolled variables throughout. No one is talking about trying to draw conclusions from any specific day or even a few days—that’s why the trend line is the 7 day average—it smooths over the inevitable inconsistencies from day to day.
No one is trying to use this data to pinpoint an exact number of cases—but I think any reasonable, unbiased actor would look at the consistent rise in the 7 day rolling average and reasonably conclude with a high degree of confidence that SOMETHING started to change at the end of June in VA. Since that change can’t be attributed to something like tighter testing requirements (they’ve been relaxed) or fewer tests being conducted overall (rationing tests to only the most needy)—I don’t see how anyone could possibly not interpret the data as an indication of a likely overall increase in cases in VA.
Trying to concoct some elaborate alternative justification for the very significant spike in the 7 day rolling average without any actual data to back it up just strains credulity frankly.
I'm NOT going to argue there is NOT an increase in total cases. But, I'll play along and chime in with non-elaborate alternate justifications.
1. Increase in false positives from before
2. Decrease in false negatives (as Nicole pointed out, there is more likely an increase in false negatives)
3. Decrease in non-Covid illnesses brought on by the warmer weather
4. Self selection bias (while standards for testing are more lax, self selection to go get tested is becoming more self selecting towards positive)
Hopefully, those in government have an idea of the false test results, and can build that into their standard error. I don't think any jurisdiction is publishing their standard error, unfortunately. So, what percentage change is a significant change? For example, if the percent change is less than1%, that could entirely be attributable to false test results, and not any change in actual number of cases. (Though, I will say that if the 10% change in positive test results is within the standard of error, then these tests would be a complete waste of time and money).
Now, the final two possibilities could have definite effects.
Back in March, if someone felt sick, there was a much greater probability it was some other bug (compared to now). Now, with non-covid19 illnesses decreased with warmer weather (which happens every year), the
percentage of sick people who have covid will increase. Therefore, though the total number of covid cases could be steady, testing people with symptoms would result in an increase in the percentage positive covid.
You can also make a case that relaxing the standards to get tested would result in an increase in the percentage of positive tests as a result of self-selection bias. Before, only those with symptoms would be tested. Therefore, we were missing almost all those who had covid but no or mild symptoms. Now, with the lax standards, we are able to catch those without symptoms. As hospital admissions are pretty steady from before this spike, it's clear that the increase in positives is coming from those with no or mild symptoms. These are people who probably wouldn't have been tested with more stringent testing protocols.
Therefore, if we look at the people who would not have been tested before, but are testing now... is there a self selection towards positive? Very possibly people won't get tested unless they have reason to be tested (they know someone who is positive, or have mild symptoms consistent with covid). So, the relative question here is this: if the overall Covid rate were the same, would the the self selection bias result in a positive test rate that is greater than or less than the positive test rate from those who were tested earlier on.
Just to reiterate, I am not arguing that these explanations are able to account for the 10% increase. My point is that the data is far from self explanatory, and we really don't know what is going on. Which is why everyone saying they are making decisions based on science and data is really only half-truths. They are making decisions based on risk assessment probabilities (at least I hope they are), which is a combination of statistics and relative, subjective, value judgments of negative outcomes. And, I would hope they are taking the consequences of the lock downs into account in these risk assessment models (unemployment, poverty, behavior health issues, domestic violence, neglect of other medical issues, etc.). Maybe there is an increase in total numbers, but maybe that increase is half what the testing data would suggest.
And, maybe the number of cases is skyrocketing more than the data suggests. Respiratory infections are (in general) more likely to cause severe complications in cold weather than warm (which is why having a spike now could be a blessing in disguise - if there's going to be a spoke, it's better to have it now than in the winter). If this holds true for Covid (and it very likely does), then a steady rate of hospitalizations would mean a higher infection rate in the population.
To bring this back to Bush Gardens
Nothing particularly newsworthy in
the latest local story focused on the closure's impact on tourism in the area, although I did find this part interesting:
Initially it had seemed like Lembke was going to be active in lobbying the state, but it feels like the park realizes that such efforts would read very differently in a situation where the state is back up to averaging 1000 cases a day, compared to the 600 or so when the Phase 3 restrictions were first revealed.
Sometimes, appearance is everything. Just the fact that the numbers look worse, means they can't advocate for reopening. It would just look bad, even if you could guarantee that the park opening at 5000 capacity wouldn't cause any increased cases.
Assuming there is an increase in number of cases, the question is where are they coming from? Travel to other states? Beaches? Bars? Bowling alleys? Hopefully, they can conduct interviews with those who test positive, and do a statistical analysis to zero in on the activities that correlate with infection. And, hopefully they have been doing this all along. Then, the results could be compared to other states, to further zero in on the problematic activities, and hence the shut downs could really become specific to the problematic areas.
Maybe we'll find that theme parks can open, but beaches shouldn't be. Maybe open movie theaters, but close bowling alleys.
