nychealth / coronavirus-data Goto Github PK

Publishing the raw counts in a choropleth map is bad practice, since the data isn't normalized. The choropleths should reflect a rate.

Thank you all for providing this data. I'm wondering if we can obtain the zipcode-level data for age groups. The current zipcode data are all cases.

Thanks for creating this platform and regularly sharing the updated counts.

I would love to see these data presented by zip code in raw form so that it can linked to existing govt datasets on neighborhood level factors. Currently, I only see tests and positive frequencies by zcta. Are you planning to release hospitalizations, deaths and recovery by zcta as well?

Also, is there any plan to release these data by race/ethnic? For social epidemiologist who examine disparities, this would be crucial information to understand spread and burden in communities of color.

Stay safe everyone.

I've noticed the total number of positives in the zip file (132,001) does not match the total number of cases in the cases-hosp-death file (129,786). Any reason for that? I would think they would be the same.

Hello - I track the numbers daily and keep them for review. The differences between the cases-hosp-death file currently posted and the one from yesterday is quite substantial. For example - March 20 has been hovering around 3600-3900 cases since I started tracking on March 31. However, in today's file - they are now at 2594. In total - the number of positive cases for the day nets to be 3420 more since yesterday - which seems directionally correct. But the day by day numbers are way different. Any ideas?

Can the data be broken down to the borough level?

Interestingly, USAFacts is publishing confirmed and deaths down to the borough level, which even the NYTimes COVID19 repo nor the JHU COVID19 repo does not do.

https://usafacts.org/visualizations/coronavirus-covid-19-spread-map/

Where do they get the borough breakdowns if not NYCDOMH, which they list as one of their sources?

With the USAFacts data, I was able to pull together this quick boro-level viz using Google Colab
https://colab.research.google.com/drive/17-g0CIG57q13S4p26r6NpCFTLSYA1mBn

I see those in the governor’s daily briefings. Any chance of adding those here?

Thanks,
Owen Dall Sotomayor

Please consider group cases by census tract. Although zipcode could be more informative and understandable for the public, but it's really hard to use for researchers. Some cities like Milwaukee are doing that: https://county.milwaukee.gov/EN/COVID-19

Is the citywide death rate found in the age and sex files based on confirmed deaths only, or both confirmed and probable?

case-hosp-death.csv lists the first death (DEATH_COUNT: 1) on 3/11/20. News reports and statements from public officials suggest the first death occurred on 3/13/20. Which is correct? Thanks.

hello,

i've been tracking your daily PDFs. total case count on 3/30 was 38087, on 3/29 was 33474 etc.

this new site shows very different counts

I had found that for the most part the three sources all more or less agreed with each other until around a week ago, where NYC total cases reported by NY State and Johns Hopkins began to grow at a faster rate than cases reported by NYC's DOHMH. However, starting yesterday evening I noticed that NY State and Johns Hopkins suddenly were reporting far more deaths than NYC, and the gap grew by this evening (2,738 vs 3,485). I do not believe ~750 people were lost in a gap of a day or a few hours. Does anyone know anything about whether they are using different standards in confirming the deaths as COVID-19 related, or have any other explanation?

There is currently a substantial lack of metadata regarding what the numbers in this database actually represent. For example, what is the denominator for DEATH_RATE in by-age.csv? Is it confirmed positive cases in that age bracket, or is it total population?

Hey all - based on the line graphs here https://www1.nyc.gov/assets/doh/downloads/pdf/imm/covid-19-syndromic-surveillance-04082020-1.pdf

I was able to roughly compile total numbers for each age group and approximate totals - you can view those here https://github.com/briankoral713/NYC-COVID19-Data/blob/master/ER%20Visits-Admits%20NYC.pdf

Remember these are rough estimations based on the line graphs and don't reflect actual numbers. The trend is that overall visits are on a downtrend however the bulk of ER Visits and Admits reside with the elderly (65+) which has been steadily increasing.

The age group total population numbers are from https://www.baruch.cuny.edu/nycdata/population-geography/pop-demography.htm

As someone living in this city I very much appreciate having this data openly available. The landing page is also very well done. Thank you for doing this, it is very much appreciated.

Hello...

Thanks much for posting this data. It's an important resource.

Today's (early) update of case-hosp-death.csv has an updated count of cases, but neither hospitalizations nor deaths are updated -- they match yesterday's numbers. Similarly, the death/hosp numbers on the official NYCDOHMH data page are not updated, even though cases are.

Thanks for your attention!

--Charles Seife

I am concerned that the process of normalizing the data to represent 100,000 population may have been based on national coefficients and not NYC based coefficients. I am specifically concerned that the AGE-GROUP presentation here is showing that AGE does not really matter.

When I used Coefficients based on NYC demographics I came up with a different story. I have been especially concerned with the 65-74 age group (Young Retired) have had a disproportional high level of disease incident level from the onset.
Please see my further explanation & assessment here

The readme documentation points to CDC with regards to 100,000 standardization. My review of the CDC document talks about National Coefficients based on National Demographics for a National Survey.

Can the summary file including new cases, hospitalizations and deaths by date be enhanced to include a subset of the hospitalized column providing a count of cases that recovered and were released ?

Hi, thanks so much for making all these data available. How is "diagnosis date" defined (i.e., from test administered vs results received)? I note that there is a considerable difference between the figures in case-hosp-death.csv and the daily differences between the nyc.gov Daily Data Summary: Total Cases pdfs. Wondering if this might be explained by case-hosp-death.csv using the testing date and Daily Data Summary: Total Cases pdfs using the results date. Any insight appreciated. Thanks again.

Data quality question on tests-by-zcts.csv (cumulative count of New York City residents by ZIP code of residence):

What does it mean if MODZCTA is NA? Does this just mean you don't know their ZIP code but they are still a New York resident? Can we assume that all of these people live in New York if we are mapping these ZIPs to geography?

Would it be possible to release a cumulative time-series historical data for the following CSV:

https://github.com/nychealth/coronavirus-data/blob/master/tests-by-zcta.csv

DATE,ZIPCODE,CONFIRMED,HOSPITALIZED,DEATHS
2020-01-21,10001,1,0,0
2020-01-22,10001,2,1,0
2020-01-23,10001,3,1,1
...
2020-04-01,10001,265,59,33

Hi,

I’d like to see the tendency of positive, hospitalizations and death by borough, like a graphic showing day by day by borough, the number of people getting sick and dying. Thank you

It looks like the data is cumulative, is there a dataset available that shows the number of cases each week for the same attributes (zip code, race, etc)?

Does anyone have the raw data grabs from the PDF's published for data older than March 31? This is what I have from pdf's but wanted the historical breakdowns starting March 1st.

The zipcode-based counts have not been updated after April 4. Will the zipcode counts be updated?

First off, thanks so much for this! This data really provides insight into what is actually going on here in the epicenter, so thanks for all of your work in keeping this up to date.

I know hospitalization data is more or less estimated from ER visits/admits. What's interesting is that when you take a ratio of total deaths to total hospitalizations, converted to a percentage, the ratio always increases by a near perfect percentage point every day

...is this just coincidence?? Or is there perhaps a standard time frame from when a patient is admitted into the hospital to when they either succumb or recover?? (The inverse of this ratio - subtracted from 100% - would give you a "survival rate" of sorts, or the percentage of those hospitalized removed from total deaths.)

Or does it have to do with less people being hospitalized vs. succumbing day to day?? The 1% increase is just intriguing.

Thanks again!
-Brian

Likely don't want to be tracking that file

Hi,

Thanks so much for making this available.

I was wondering if the data for daily testing- hospitalizations- deaths is still available in testing.csv. I'm not seeing this in the updated repo and we were interested in looking at these numbers.

Best,
Ashley

If I look at the daily figures from nyc.gov

It reports 48462 total cases vs NYState's official NYC Count is 51809

Whats the source of the discrepancy? Ive been noticing it for a few days

I'm combining these data with the NCHS mortality surveillance survey, but NCHS has only reported total all-cause mortality for NYC through 3/28. Could you add all-cause mortality to these data?

Hi,

Readme says that there should be a file called testing.csv, which is supposed to include "counts of New York City residents with specimens collected for SARS-CoV-2 testing by day, the subsets who tested positive as confirmed COVID-19 cases, were ever hospitalized, and who died."

Will that be posted soon? Also, last week the mayor was promising a count of people admitted into the ICU as well has being hospitalized. Will that be in that file?

Thanks,

Is there any way to modify the tests-by-zcta file to include confirmed deaths by zip code? It would be helpful to know the confirmed death rate alongside the positive test rate. If not, where might I find this data (whether as a GitHub project or other form)? Thanks!

thank you for your service

summary.csv reports 4260
case-hosp-death.csv reports 4255

$ date ; date --utc ; curl -s https://raw.githubusercontent.com/nychealth/coronavirus-data/master/summary.csv | grep Deaths ; echo ; curl -s https://raw.githubusercontent.com/nychealth/coronavirus-data/master/case-hosp-death.csv | dos2unix | cut -d, -f4 | grep -v -e "DEATH_COUNT" -e "^\s*$" | paste -sd+ - ; echo ; curl -s https://raw.githubusercontent.com/nychealth/coronavirus-data/master/case-hosp-death.csv | dos2unix | cut -d, -f4 | grep -v -e "DEATH_COUNT" -e "^\s*$" | paste -sd+ - | bc
Thu Apr  9 08:44:49 EDT 2020
Thu Apr  9 12:44:49 UTC 2020
Deaths:,4260

1+1+2+6+9+7+21+25+44+36+45+80+89+110+167+179+230+247+266+319+337+361+362+336+379+327+235+34

4255

(where's my low-priority button)

Hi All,

First off, thanks so much to the maintainers of this github for all their effots!

My question is if the data tables be updated to include:

• historical daily numbers
• demographic info?

As of now, the "summary" csv only contains totals, but there must be a data-set that includes the historical dailies somewhere, as this underlies this view. Would it be possible to get access to that data-set? I am trying to calculate the doubling rate as data evolves, which is not really possible without having access to this historical data.

In short, I'm looking to find a data-set that has all the info available in the pdf historics, e.g. for Total Cases, Deaths and Hospitalizations. As above, these datasets must exist because there are dynamic graphs that display these numbers available here.

It doesn't really look like it would be possible to rebuild this set with the summary data points or calculated data, such as case rate per hundred thousand.

Ideally, a table would look (something) like this, with one for each borough - the age demographics could be built into additional columns, but regardless the point is that this data would be very helpful to run deeper analysis.

Does anyone have something like this?

Thank you again, everyone - be safe, be well and please wash your hands!

Thank you for managing this repository.

1. How will DOHMH begin to count patients (hospitalizations, deaths) transferred outside the city system to relieve pressure to the system? https://www.syracuse.com/coronavirus/2020/04/new-york-city-hospitals-begin-transferring-patients-to-upstate-ny-report.html ?

2. Will data start reflecting cases taken up in USS Comfort and Javitz Center and on what date?

3. Will NYC hospital discharge data start being published ?

why does the current and previous versions of tests-by-zcta.csv have NA values in the first row? Is this an unincorporated area of NY?

How many Zip code (zcta) areas are in the NY city? Corona virus data are posted for 177 zcta areas here. Does any one have the shape files for these 177 zip codes or knows where to get these? I have two different shape files: one with 200 and the other with 214 zipcode areas but those don't have some of the 177 zcta areas for which we have the corona virus data.

Is it possible to report the current hospitalizations each day as opposed to the "ever hospitalized?" The current hospitalizations are what all other NY jurisdictions are reporting, apparently from HERD.

Your binning is off in the by_age table such that I'm not sure whether 75yos are being counted twice, or if there is simply a typo in one of the bin labels:
65-75 years
75 and older years

Thanks for your work to help the people in need! Your site has been added to the Open-Source-COVID-19 page, which collects open source projects related to COVID-19, including maps, data, news, api, analysis, medical and supply information, etc. Please share to anyone who might need the information in the list, or will possibly contribute to some of those projects. You are also welcome to recommend more projects.

http://open-source-covid-19.weileizeng.com/

Cheers!

For coronavirus-data/case-hosp-death.csv the first column header has an issue:

Expected:
DATE_OF_INTEREST

Currently says:
Retrieving data. Wait a few seconds and try to cut or copy again.

Thanks for this great resource. It would be very helpful if you added deaths to the borough-level dataset.

The "probable deaths" that you started publishing on the 14th are currently increasing NYC's total death toll by about 50%. Could we also say that there are probable hospitalizations that would similarly increase the total hospitalization count by around 50%? Or are those already factored into the published figure?

Please publish deaths and hospitalizations by borough, as it appears here: https://www1.nyc.gov/assets/doh/downloads/pdf/imm/covid-19-daily-data-summary-deaths.pdf

First off, thank you for providing this data. I started looking at the data just a few days ago, and I noticed that the test-by-zcta.csv file has been the same for the past 3 days (with a total number of cases standing at 94,499 for the city. Is this data being updated?

Does the dataset have a DOI for citation purposes?

Hi,

I was wondering if there is a repository with revised daily data for each zip code. Since there is a delay in reporting, the past files have stales numbers.

I am trying to plot the trend of new cases for each individual zip codes.

Thanks

It would be useful if each update was just a new row or rows appended to each table so users of the data can track changes over time.

Also, using ISO8601 timestamps for the "as of" value would also be helpful for machines reading the dataset.

I'm curious, is there a data standard you are complying with that wants the data in the shape it's currently published in?

For summary.csv this would look like:

as_of,cases,total_hospitalized,deaths
2019-03-25T17:28,20011,3922,280
2019-03-26T17:30,23112,4712,364
2019-03-27T17:30,26697,5039,450
2019-03-30T17:30,38087,7741,914
2019-03-31T17:30,41771,8549,1096

Hello,

I am an engineer at Google interested in using your project. If you are comfortable with permitting this, could you please add an open source license (https://opensource.org/licenses) to your project directory?

First I want to thank you for this great effort of providing this data. It is very helpful and critical.

Them I just curious, are there any data about the recovered cases in NYC? Since the outbreak has been about one month long, are there any recoveries cases?

Thanks.