Charts, below, are created automatically from data updated most weekday afternoons. OHA no longer releases a report on Saturday and Sunday so Friday’s hospitalization values remain constant until the following Monday. All data is from the Oregon Health Authority (OHA).
Charts are generated from 4 cloud-based spreadsheets and may take time to load.
Use the horizontal scroll bar to view right edges of charts, as needed.
If you stumble upon this page: I am not an authority – these charts should be treated as “FOR ENTERTAINMENT PURPOSES ONLY”. Please assume I am an idiot and that I have cherry picked all data and everything you see here is wrong or stupid.
- OHA periodically removes prior “new cases” or “deaths” after new information indicated the patient did not have Covid-19 or did not die (usually a data entry error).
- OHA periodically moves cases from one county to another, or to out of state, based on new information, decrementing earlier counts. Because the OHA does not say what day was decremented, I have to pick a random day to reduce by 1. This means my data may be off from a later OHA data release or chart.
- As of June 6th, OHA stopped releasing detailed data on Saturday and Sunday – this means no detailed data (hospitalization, ICU beds, etc) is available for Friday -Sunday.
- These charts are based primarily on the OHA “Daily Updates” and this yields different values than the OHA “Data Dashboard” data, presumably because the “Data Dashboard” contains updates to past data.
The following reflects test ratio and percentage testing positive since the very first test in March (which includes nearly two months of testing almost no one).
The test ratio indicates the total number tested for each confirmed case (e.g. 29:1 means out of 29 people tested, there was one positive test result).
Total tests done for each confirmed case
Percentage testing positive since first tests in March
The “daily body count” reflects the day OHA counted the death in their public information release. Each death may have occurred days to a week or more earlier. Therefore, use caution in interpreting a particular day’s death count.
Daily body count
Cumulative Deaths Statewide, Log Scale
Cumulative Deaths, Linear Curve
This data was recorded by hand from OHA “Daily Updates”, starting in late March. Note the “curve” is linear and not the much discussed “exponential” curve.
Seven Day Moving Average % Change in Daily Deaths
Estimated Case Fatality Rate for Oregon
Strictly unofficial and very, very, very approximate. Remember, this is for Entertainment Purposes Only. This chart shows that the fatality rate for Covid-19, in Oregon, has fallen sharply.
On July 9th, the Oregonian caught up with this trend, reporting that the mortality from Covid-19 in both Oregon hospitals and in the chain of Providence hospitals on the west coast, has plummeted. A trend I have been watching in the data, for months. But I’m not an expert so please ignore anything I notice. Sadly, the Oregonian quotes extensively from Dr. Bruce Goldberg, professor at OHSU, neglecting his role as the disgraced former head of OHA and his mismanagement of the nearly 1/2 billion dollar failure of Cover Oregon. Frankly, his comments should be ignored.
Ratio of 7-day Average of New Deaths / 7-day average of New Cases
This chart shows that over time, the mortality rate of Covid-19 is decreasing. This shows the approximate number of deaths occurring for each 100 new confirmed cases as of the date shown on the X-axis.
On June 24th, the Oregon Health Authority issued this statement in their weekly summary:
“Available evidence suggests average severity of illness among reported cases is lower than it was early in the outbreak: hospitalizations and deaths remain well below their peaks, even after reported cases have been surging for 4 weeks, and the percentage of emergency department visits attributable to COVID-19-like symptoms remains below 1%”
OHA suggests the mortality rate could be decreasing because:
- The disease is spreading among a younger population who get less sick.
- Far more testing and finding many asymptomatic or with such mild symptoms they never though to be tested. During the first two months, to be tested in Oregon the patient had to be either already hospitalized for pneumonia, or in a skilled nursing center – and have all symptoms, or have all CDC listed symptoms AND in have been in close contact with a confirmed case. This restricted testing to very few patients. Now, anyone can be tested with little or no symptoms including persons who may have been exposed (but are not certain). OHA suggests we are now testing and confirming cases in persons with no or little symptoms. For example, an outbreak at a fish processing plant led to offering tests to all workers. 95% of those tested were asymptomatic and would not have been identified earlier.
- We are now finding asymptomatic people who are false positives. Due to Bayes Theorem, many – even most – asymptomatic positives may be false positives. A 95% accurate test sounds good. But when only 1 in 1,000 people are actually sick, testing people who have no symptoms means 5% of 1,000 or 50 people will be given a false positive test result – but only 1 is actually sick, meaning the accuracy is only 2%!
- We are now finding more asymptomatic people because, like the fish plant in Newport, when the outbreak occurred they offered testing to every worker – hundreds of people. And most of the positive tests turned up asymptomatic carriers that now count as confirmed cases – but will doubtfully show up in a hospital bed.
- Or, and this is not from the OHA, perhaps the disease is changing: “”Decreased-in-hospital mortality in patients with Covid-19 pneumonia” – ” In our institution, the proportion of patients requiring ICU decreased over time from 17% to 7%, without significant changes in patients’ age, suggesting a decreased severity of clinical presentation and progression“
- REMINDER: I have no expertise in any of this. Assume I am an idiot, these concepts are nonsense and this page is FOR ENTERTAINMENT PURPOSES ONLY.
Couple of related items – and Comments
42% of all deaths in the U.S. due to Covid-19 occurred in New York, New Jersey and Massachusetts, which together account for 10% of the U.S. population.
The media widely reported that Sweden’s relaxed approach to addressing Covid-19 was a disaster. But the data paints a different picture – follow the link for all charts of new case curves, hospitalization and ICU bed curves – all way down. As you can see, new hospitalized ICU cases are now minimal as the curve has flat lined:
Sweden was said to have done everything wrong, did not lock down the economy, kept schools open – and never had a face mask requirement.
You may remember that Georgia was the first to begin re-opening their economy and noted journalists wrote that Georgia was conducting “an experiment in human sacrifice”, which will result in killing thousands of people. The Georgia re-opening began at the center of this chart – look at what happened afterwards:
There is no consistent official explanation for these patterns. You can draw your own conclusions, what ever they may be. I have ideas but am keeping them to myself.
The media’s coverage of Covid-19 has been a fiasco of garbage reporting which influences your views on the disease. This is why I began tracking the data for my state myself – enabling me to spot patterns 1 to 2 months before the media noticed these patterns.
New Cases Trends
- Before May 4th: OHA reported *only* confirmed cases of Covid-19.
- Starting on May 4th: OHA reported *both* confirmed cases and “presumptive” cases (no test or even has a negative test result) as separate values.
- May 26th: OHA only reports a combined number. My charts use “confirmed” only before May 26th and the combined number afterwards.
- June 6: OHA discontinued providing detailed data on Saturday and Sunday. On all charts after June 6, the Friday data is copied straight across the weekend.
Daily new cases in Oregon (Line Chart)
May 28th: Protests began and continued every day and night up to the present data. Current thinking is that these events did not act as mass spreading events – but then that leads to questions as to why public health enthusiasts, in some areas, are still closing parks, trails, beaches and forests to public access. Those closures were apparently not necessary during “lock down” and even less necessary now.
7-day moving average of new cases in Oregon
Comparison of Test Results Over Time
As testing increases, we see more new cases discovered. The blue line charts the total number of tests done per week. The orange line charts the total number of positive result tests. The orange positive test result line is – so far – roughly tracking the total number of tests as recent weeks see the number testing positive in the 3-4% range. Naturally, as the number of tests increases, so too does the total number of confirmed cases.
The last week of June saw 2x as many test as the last week of May, and 3 1/2x as many as the last week of April
Trend in percent positive test results
Daily new cases in Oregon (Daily Update) (Bar Chart)
Cumulative cases in Oregon (Daily Update), Linear curve
This chart is based on the OHA “daily updates”. However, the daily updates have different daily values than OHA “Data Dashboard” data downloads. Data generally begin in late March because OHA did not release daily updates with data before then.
Note that most of the “curve” is linear and not “exponential” – it’s two straight sections with an inflection point.
Total Cases (Linear) from OHA Data Dashboard Data
Most of the curve timeline is linear, not exponential.
Daily Increase as a Percent Change
Cumulative cases based on OHA Data Dashboard data, linear curve
OHA releases a total of new cases for the prior day in a “Daily Update” release.
Comparing the “Daily Update” values to the “Data Dashboard” values shows discrepancies. This chart shows the “new cases” as reported by the Daily Update and separately by the Data Dashboard. This chart is updated only once per week or two.
OHA Chart of Hospitalizations
OHA chart shows peak day as March 28th with 28 admissions which is DIFFERENT than their Data Dashboard data which shows peak day of April 1 with 28 admissions.
Flattening the Curve
The original “flattening the curve” idea was expressed by a public health doctor with his illustration, below. The idea was that strict measures would shift the early peak into a longer term event – to avoid overloading hospitals. The intent was to shift those hospitalizations into the future, not eliminate them. We “flattened” the red curve and shifted those hospitalizations to the “blue curve”.
A flatter curve, on the other hand, assumes the same number of people ultimately get infected, but over a longer period of time. A slower infection rate means a less stressed health care system, fewer hospital visits on any given day and fewer sick people being turned away.https://www.livescience.com/coronavirus-flatten-the-curve.html
Unofficial “Hospital Intensity” Metric
For unknown reasons, the number of persons hospitalized or dying from Covid-19 drops relative to the new number of diagnosed cases. No one knows why this is occurring but as noted, above, the OHA also sees this now.
This chart graphs a ratio of the 7-day sum of current hospitalizations to the 7-day sum of new cases. As time goes on, the number of daily
new cases goes up, but the number of hospitalized patients relative to newly diagnosed cases, has been going down. This suggest we are now diagnosing cases that are much milder than before. Or some other factor. For now, “officials” only speculate as to the cause.
This chart is similar to the 7-day average of new deaths / 7-day average of new cases. Rather than showing mortality, it shows that hospitalizations per new case are also dropping.
In early April, the OHA Daily Update reported that up to about 25% of confirmed Covid-19 patients were hospitalized at some point. As of July 8, 2019, OHA reports that the cumulative hospitalization percentage is now about 10%. Since this is a cumulative figure, for this value to fall, presumably the currently hospitalization rate is less than 10%.
OHA does not provide information on how many new cases are newly hospitalized each day. All we have is the total number of people occupying a hospital bed each day – we do not know how many were admitted or released on any day. Consequently, I cannot calculate a # of newly hospitalized per 100 new cases like I did, above, for deaths. All we can do is look at a broad ratio to see that the trend is fewer hospitalized patients per new cases being found.
Hospitalization and Hospital Resources In Use
I believe the best metrics are the number of hospital beds occupied, the ICU confirmed cases, the number of ventilators in use by confirmed Covid-19 patients and the cumulative number of deaths attributed to Covid-19.
These values will remain what they are regardless of the total number of people in the population being tested and are therefore not biased by increased testing or a more expansive definition of “presumptive” cases.
- Beginning in June, the OHA discontinued issuing hospital data on Saturday and Sunday. The effect of this is that the values for Friday remain constant Friday-Saturday-Sunday – or stated another way, we now only have data for 4 days of the week. One would think – as the number of cases exploded in June and July that this data would be as important as it was the first three months – but the data is no longer publicly available.
- On the 4th of July weekend, OHA gave no report on Friday, Saturday or Sunday. This means the Thursday hospital data values remain constant for Thursday-Friday-Saturday-Sunday. And we have data for only 3 days of the week – again, during a time when public health and the media tell us things are getting worse. Does this make sense to cut off data now?
- The number hospitalized for Covid 19 confirmed AND presumptive is always about twice as many as confirmed cases. We assume a presumptive patient in the hospital is going to be tested, right? But since the confirmed case counts are always half the confirmed+presumptive total, this implies most of the presumptive cases were not actually Covid-19? Something makes no sense here.
Cumulative Hospitalized Cases from OHA Data Dashboard Data
Note that this is a linear chart and the curve is not exponential.
Hospital Confirmed and Unconfirmed Cases
Hospital Confirmed Cases
The next 3 charts are very important – number of confirmed cases occupying a hospital bed, number of confirmed cases in ICU beds, and the number of confirmed cases on ventilators. These numbers are independent and not biased by a large increase in tests being given.
ICU Confirmed Cases
Ventilators in Use
OHA Chart of Peak Day of Covid-like Symptom Presentation to ERs
Peak Day is March 13th, with a slightly lower second peak on March 21. At the time, models were predicting substantially higher peaks in April to May time frames but those peaks never happened.
Cases by County Trends
Top Counties Only
New Cases in Counties Having Large Numbers of Cases
Percent of Daily New Cases in the Top Counties
Number of new cases in the top counties
I live here so I track this one separately.
Daily new cases in Deschutes County
Deschutes County 7-day moving average of New Cases
Deschutes County Cumulative Cases
Deschutes County Estimated Active Cases
In this chart, I defined a case as “Active” for 20 days after it was first reported. According to the OHA, the median time to recovery is 20 days. The chart shows s a sum of the total outstanding cases for the prior 20 days. This is completely unofficial, just a random concept FOR ENTERTAINMENT PURPOSES ONLY.
Current counts for Deschutes County (and other counties) may be found on this OHA page. My chart estimate always runs quite a bit higher than the actual numbers shown by the County and OHA. Because its just a very crude estimate.
These values may be meaningful or not meaningful.
Percent of Oregon population that has had confirmed Covid-19
Estimated Percent of Oregon population that has had Covid-19 including IDM estimated unconfirmed cases (5x)
Percent of Oregon population that has not had Covid-19
Percent of Oregon population that has died of Covid-19
Relative Risk of of Covid-19 diagnosis versus being injured in a traffic accident today
If the number is greater than 1.0 (e.g. 2.5), this means you have a 2.5x higher likelihood of being diagnosed today with Covid-19 than you have of being in an injury vehicle accident today. I created this item to help understand the relative risk of contracting Covid-19 by comparing it to something else we can identify with.
OREGON COVID-19 TIMELINE
- March 8: Institute for Disease Modeling (in a late May report) says Oregon’s curve began to bend on March 8 owing to voluntary measures (IDM)
- March 15: Peak day of Covid-like illness presenting at ERs in Oregon (OHA chart)
- March 23: Governor announces state-wide “lockdown” program. Charts show this lockdown had no impact on the already decreasing trend line.
- March 23: Earlier OHA charts and CLI data implied “peak” hospitalization was around March 23d. Their “data dashboard” data, through June 15th, puts the peak days at 26 on 3/20, 27 on 3/23 and 28 on 4/1 (OHA)
- April 5: Peak day of hospitalizations in Deschutes County (St Charles Hospital, Bend)
- April 5, 9 and May 20 [*]: Peak days of “new cases” in Deschutes County
- April 9: Peak day of state-wide ICU bed usage by confirmed patients (OHA)
- April 6: Peak day of state-wide ventilator usage by confirmed patients (OHA)
- April 27: Peak day for state-wide deaths (OHA)
- Jun 24th: Governor mandates public mask wearing in Multnomah, Washington, Clackamas, Hood River, Marion, Polk and Lincoln counties. My hypothesis is this will have little impact on the new cases curve (social distancing, work place restrictions and hand washing are likely to be significant factors – and face masks add little to that in most situations). The ineffectiveness is, I believe, due to real world considerations – people wearing unfiltered N95 masks, people not covering their nose, people removing their mask to talk to other people, people storing their mask on their neck and rarely washing them, so they accumulate “germs” over time. Basically, in the real world where they are misused by real people, their effectiveness when added to social distancing and sanitization may not be high. In fact, several regions that mandated face mask wearing in mid-April to mid-May are as of July, undoing their re-openings, and Los Angeles is considering new lock down measures. Yet they’ve had a mandatory face mask requirement since April 18th.
- June 27th: Clatsop county is added to mask wearing mandate on June 27th.
- June 29th: Oregon Governor mandates covering your face with a face mask at all indoor locations effective July 1. This will be an interesting test case to see if it results in any changes in trends.
- June 29th: OHA changes new case reporting: “Note: Starting today and moving forward, epidemiologists are using a new method for reporting daily cases. The new method assigns a date to each case when the case is first known to the state or to local health department as confirmed or presumptive. This is a better representation of the number of cases reported on any given day. Previously, the method was to subtract today’s case counts from the previous day’s count. Today only, the daily numbers from the weekend press releases will not add-up. Weekend numbers were calculated using the previous method. Moving forward, every day will use the date each case is first known to the state or to local health departments.”
- July 2: Oregon announces +375 new cases, an all time high. This is the day I added the two new charts showing total tests versus positive test results, and the chart showing the trend in positive test results. So far, the new case increases closely track the increase in tests being given.
Deaths per day in the United States
The link for this chart shows the current update (this screen shot taken on June 22), plus charts for selected U.S. states. Overall, deaths have been declining in the U.S. since mid April.
Here is a newer chart from the NY Times:
International Case Fatality Ratios
CFR’s are perhaps of interest in comparing how the disease has affected countries in different ways.
Economic Data Section
Apple Mobility Data
Click here to see Apple’s chart.
These charts show the trillions of dollars printed by the Federal Reserve. Longer term, this leads to inflation – first in asset prices (hence stock market going up) and then later in consumer price inflation, although there are multiple factors at work in determining future inflation.
TSA Passenger Data
TSA continues to count data on holidays and weekends but only releases the full set of data on Monday – Friday. Consequently, this is only updated Monday to Friday.
Passenger Count as percent of prior year
Passenger count as percent of same day the prior week
LINK TO OHA CHARTS
GO HERE for OHA official charts (new features from OHA).
OHA INFLUENZA PAGES
Here is a link to the Oregon government ILI pages. Chart of ED Visits for influenae-like illness (ILU) provided by OHA. “Covid-like” is a subset of ILI and is determined by clinical observations of symptoms. Neither it nor influenza diagnosis is based on a test per the CDC.
This is a snap shot from April, after which OHA discontinued making updates.
Highlights from the report at the first week of July – noting that hospitalizations have fallen to a low level and the combine mortality from pneumonia, influenza and Covid-19 has decreased to borderline epidemic levels.
Based on death certificate data, the percentage of deaths attributed to pneumonia, influenza or COVID-19 (PIC) decreased from 9.0% during week 25 to 5.9% during week 26, representing the tenth consecutive week during which a declining percentage of deaths due to PIC has been recorded. The percentage is currently at the epidemic threshold but will likely change as additional death certificates for deaths during recent weeks are processed.https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidview/index.html
Here is a link to CDC ILINet State Activity Report
CDC produced chart of ILI for OREGON
- Link: Access to CLI presentation at ERs, % tests that are positive, New Cases, Cases traced to sources, hospitalization data by state and county, active monitoring capacity, and county indicators.
- Link: Demographic data, disease severity data
- Tests and outcomes by county
- Covid-19 press releases
Charts here may look slightly different than charts provided by others. This occurs because OHA sometimes updades past data. Some times I catch the changes, sometimes I do not and often, OHA just says they decremented a count by one, but does not say on what day that occurred – so I have to select a random day.