ASA Connect

 View Only
Expand all | Collapse all

HHS directing hospitals to cease sending data to CDC

  • 1.  HHS directing hospitals to cease sending data to CDC

    Posted 07-14-2020 19:20
    In a recent letter, HHS has directed hospitals to send specific datasets directly to an HHS database rather than CDC. This was reported by The New York Times.

    Having multiple groups look at and use the same dataset is always better.  I do not see anything in the letter which suggests hospitals cannot report data elsewhere.  Indeed, once the collecting and formatting is done, sending a file to multiple places periodically is just a cron job.

    But I am concerned, because if there is any suggestion these data not be sent anywhere else, we may be seeing a politically motivated emergence of what was done with environmental and climate change related data at the beginning of this Executive, something which I, with a big team of others, tried to anticipate and deflect.

    Anyone heard any more?

     -- Jan

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------


  • 2.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-15-2020 12:50
    ​Hello statisticians.
    Yes, the HHS letter you cite contains this statement directing hospitals to NOT send data to the CDC.

    "As of July 15,, 2020, hospitals should no longer report the Covid-19 information in this document to the National Healthcare Safety Network site. Please select one of the above methods to use instead."

    Can we, the ASA respond, as an organization? 

    Hello statisticians.
    Yes, the HHS letter you cite contains this statement directing hospitals to NOT send data to the CDC.

    "As of July 15,, 2020, hospitals should no longer report the Covid-19 information in this document to the National Healthcare Safety Network site. Please select one of the above methods to use instead."

    Can we, the ASA respond, as an organization?



    ------------------------------
    Amalia Magaret
    Research Assistant Professor
    University of Washington
    ------------------------------



  • 3.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-15-2020 13:11
    Dear all,

    I think the letter is worrisome and I think we should encourage our leadership to respond as an organization.
    I personally would not trust reports coming from the White House administration. 

    Regards,

    Yolanda

    ------------------------------
    Yolanda Munoz Maldonado, PhD
    Owner and Statistician
    Statistical Savvy Consulting LLC
    4500 Williams Dr. Ste. 212-329
    Georgetown, TX 78633
    Phone: (512) 591-8727
    Statisticalsavvyconsulting.com
    yolanda.munoz@statisticalsavvyconsulting.com
    ------------------------------



  • 4.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 07:07
    Good Morning, All, 

    The ASA Board issued a statement that went online this morning: 

    https://www.amstat.org/asa/files/pdfs/POL-July10Covid19HospitalReportingGuidelines.pdf. The full text is pasted below. Thanks to the many of you who wrote on this through ASA Connect and directly to ASA leadership. 

    Best,
    Steve

    American Statistical Association Statement on New HHS COVID-19 Guidance for Hospital Reporting

     

    July 16, 2020

     

    The American Statistical Association is exceedingly concerned about the July 10 COVID-19 guidance for hospital reporting from the Department of Health and Human Services (HHS), specifically the removal of the Centers for Disease Control and Prevention (CDC) from the reporting chain. The CDC has thousands of scientists and administrators who are experienced experts in collecting and reporting these data. Disease surveillance has been its mission since the CDC was established in 1946 with strong transparency protocols in place, as well as secure and recognized mechanisms for providing the data to the public.

    Removing the CDC from the reporting chain denies the country of the CDC's unrivaled expertise and experience, introduces the risk of data reporting discontinuities that could obscure the spread and containment of COVID-19, and removes these data from public view. The new reporting guidelines also sidestep the CDC transparency protocols that are critical to ensuring the data are impartial.

    A vital role of the federal government is to provide timely, reliable, and objective data on important issues. For these data to be trusted, they should be collected, analyzed, and disseminated by recognized experts in a transparent fashion. Consistency in reporting, methods, and protocols are critical, so data can be compared over time. These data quality procedures and the processes in place to achieve them are especially urgent now, as the country deals with the COVID-19 pandemic.

    We urge HHS to reverse its decision and allow the CDC to continue to receive these data and serve its vital, time-honored government role of producing reliable and objective data for the American public and world-wide.

     



    ------------------------------
    Steve Pierson
    Director of Science Policy
    American Statistical Association
    ------------------------------



  • 5.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 11:12
    Thanks!  That's great.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 6.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 14:08
    ​Thank you Steve,
    Can you say how this letter is to be disseminated?  Can we take up a collection to get it printed in NYTimes?  How do you plan to get this picked up by the media? In addition, I confess I was hoping for something longer and stronger with references to potential damage incurred by misinformation. If there is a team coming together to address this further, I'd like to be a part.
    Amalia

    ------------------------------
    Amalia Magaret
    University of Washington
    Seattle
    ------------------------------



  • 7.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 10:45
    I agree with Amalia. I would like to see this statement published in a variety of media outlets. I am happy to donate and participate.


    ------------------------------
    Cathy Furlong
    Statistics Without Borders
    ------------------------------



  • 8.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 10:57
    ​I completely agree that we need this message to get out widely and picked up by major new networks.  I would also be happy to donate in support of this effort.  This directive by this administration is unacceptable and we need to get that message out broadly.

    ------------------------------
    Paul Bycott

    ------------------------------



  • 9.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 10:58
    ​IN addition to large media outlets such as the NYT and so on, we probably should disseminate similar messages to our local media outlets as well. So you could write to your local newspaper or TV station about this.

    ------------------------------
    Stan Altan
    ------------------------------



  • 10.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 10:49
    I will be glad to contribute for the diffusion of the letter.
    Can ASA collect the contributions?
    Valerio M. Genta, MD
    Pathologist

    ------------------------------
    Valerio Genta
    Virginia Beach General Hospital
    ------------------------------



  • 11.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 19:19
    ​Hello, everyone, I'd like respectfully to suggest that there is another side to this story. Yesterday (Thursday 07/16/2020), this very topic came up during Press Secretary McEnany's White House Press Briefing. In fact, the back & forth between her and the press members took up the final five minutes of the Briefing. You can watch and listen for yourselves on YouTube at: https://www.youtube.com/watch?v=fLOhcJlBBtw, starting at the 20:20 mark, and then you can decide for yourselves whether there is more to this story than we originally heard.

    Fair warning: the 24:14 mark is where Press Secretary McEnany begins to make a political point contrasting the current Administration's decision this week on COVID-19 testing with the previous Administration's 2009 decision to stop testing for H1N1 Flu. Even if one personally would not trust reports coming from this Administration, one may still wish to make oneself aware of what they are saying.

    ------------------------------
    Eric Siegel, MS
    Biostatistics Project Manager
    Department of Biostatistics
    Univ. Arkansas Medical Sciences
    ------------------------------



  • 12.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 19:47
    Thank you, Eric. The vid shows the question regarding the HHS  was taken by Emily Ngo of NY1 who asked the question in a backhanded way, repeating accusations that the CDC is deliberately manipulating data. 

    Whatever was done initially, modellers were deprived of their data feeds, because CDC took down their data access. Then, apparently yesterday, HHS reversed its position, possibly because of the howls of complaints that erupted. 

    While that's good, now the CDC and modellers need to contend with potentially HHS Protect saying one thing, and CDC data saying another. It's not uncommon for databases to disagree, but, as I noted, what's needed is to rationalize them and explain the differences.  A lot of concerns could be dissipated if the technical transition plan was briefed and made transparent. 

    All that said, there may be a need for some creative open source people to come up with a way for a designed survey sample of hospitals to submit data to another concentrator in the same formats, one which resides outside of government, and can provide a check.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 13.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 18:21
    Jan -  

    You said "...there may be a need for some creative open source people to come up with a way for a designed survey sample of hospitals to submit data to another concentrator in the same formats, one which resides outside of government, and can provide a check."  

    It's a very very sad day when we have now become a third world country where such Government data may no longer be considered the gold standard.   

    Was Franklin right to indicate we may not be able to keep our Republic? 



    ------------------------------
    James Knaub (Jim)
    Retired Lead Mathematical Statistician
    ------------------------------



  • 14.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 18:39
    Trust, but verify. It can't hurt to verify. The epidemiological modelers do it against one another all the time. 

    And certainly the geophysicists do. And certainly big data gathering projects, at least the good ones, put on the critics hat and try to poke holes in what they are seeing, asking lots of why questions. 

    It's how it was known the firearm homicide rate at BJS was different than that reported from NCHS.  In fact there was at least one paper written about that discrepancy. 

    I like Hanlon's Razor: "Never attribute to malice that which is adequately explained by stupidity." However, as Heinlein added: "But don't rule out malice."  I like the full statement better.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 15.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 18:59
    The point is that the Government official statistics that were previously transparent are now being clouded.   At the Energy Information Administration, so many people use the data and various people check various parts, that once in 1999 when there was a programming mistake regarding just one instance about where I said small area statistics should borrow strength and where it shouldn't, we heard about it immediately after the first month's reporting that way!  

    You could at least trust that no one was fiddling with the numbers.  The Pentagon might, but the core Federal Government statistical agencies could be trusted.

    ------------------------------
    James Knaub (Jim)
    Retired Lead Mathematical Statistician
    ------------------------------



  • 16.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 20:12

    Stupidity and malice are not mutually exclusive. The present administration has provided ample examples of the two together. 

    David

    (Thank for the reference to Hanlon's Razor! I failed only yesterday to remember the name when discussing Georgia's poor COVID-19 maps that re-scale at each date, thus concealing the recent marker increase. I will edit my remarks on those channels. I too prefer Heinlein's corollary.

    DJC) 



    ------------------------------
    David J Corliss, PhD
    Director, Peace-Work www.peace-work.org
    davidjcorliss@peace-work.org
    ------------------------------



  • 17.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 21:00
    David, can you include a link to the offending maps?  I couldn't find them when I googled Georgia COVID-19 maps.  Have they since been fixed?

    ------------------------------
    Raoul Burchette
    Biostatistician
    ------------------------------



  • 18.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 11:11
    Georgia reports its COVID-19 cases here:

    https://dph.georgia.gov/covid-19-daily-status-report

    The map comparisons come from screen shots on separate days. I see that the scales on today's map are slightly different from other screen shots I've seen, so no, nothing has been fixed.

    A variety of people have commented on the changing scales of the map on twitter.

    Barb

    ------------------------------
    Barbara Graham
    Biostatistician
    Colorado State University
    ------------------------------



  • 19.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 17:54
    Thank you, Barbara, I guess I did see this chart, but I wasn't clear what the issue was, since I only saw it on one day.

    Has anyone inquired what the logic was to the Georgia Department of Public Health?  I think the map serves its stated purpose to show where the most cases were.  Whether there may be better ways to do so, is open for suggestion.

    Speaking of which, does anyone have any suggestions on how this should have been done (preserving the stated purpose of the graph), seeing that the upper bound is unknown and continually changing and there are five colors plus white to choose from to show the ranges?  In fact, there are basically four colors available to indicate ranges with the lightest color representing the levels below the first threshold, and white being reserved for zero cases. 

    I did happen to notice that the boundaries are approximately linear on a log scale.

    ------------------------------
    Raoul Burchette
    Biostatistician
    ------------------------------



  • 20.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 13:04
    I totally agree, the ASA should, in the strongest terms possible, urge the administration to rescind this directive. The CDC has been doing this function and changing in midstream is just asking for problems under normal conditions, but in the midddle of a pandemic the logic in extremely suspicious.

    ------------------------------
    Michael Mout
    MIKS
    ------------------------------



  • 21.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 10:59

    Michael Mout wrote "...in the middle of a pandemic the logic in extremely suspicious."

    Well said, yet let me add what I find extremely suspicious is that this is happening in the middle of an election campaign. The concern is that safety is taking a back seat to politicking. We have already amply seen how political priorities driving the response instead of science has led to increased sickness and death. For example:

    COVID-19, Deaths per Day, TX FL CA MI


    ------------------------------
    David J Corliss, PhD
    Director, Peace-Work www.peace-work.org
    davidjcorliss@peace-work.org
    ------------------------------



  • 22.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 11:31
    Is it odd that the graphs in the email from David Corliss look like the president's signature?? Coincidence? I think not.

    ------------------------------
    Michael Mout
    MIKS
    ------------------------------



  • 23.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 16:05
    ​Here is a copy of a letter I just wrote to my local newspaper's public opinion column. I would like to urge all of us to send in something similar to their local papers, and let's develop a groundswell of opposition from the bottom up. 

    We've all heard the saying that love makes the world go around, but in this modern era, it's really data that makes the world go around. Both public and private sectors depend on accurate and reliable information to make business decisions and develop policy. So it came as a little bit of a jolt to read that the Trump administration has ordered hospitals to report their Covid-19 related numbers to the Health and Human Services Administration (HHS), rather than the Center for Disease Control (CDC) reported on July 16 on page A3. The CDC has been the central repository and clearing house for morbidity and mortality data for decades, serving many administrations in an efficient and nonpartisan way. So why would the Trump administration change this practice, in the face of rapidly rising Covid-19 cases throughout the country. Given the context of this decision, it's clear that the administration is making a power grab to seize the ability to manipulate the numbers and deny the public access to accurate and timely information. This is a gross mistake that should not go unchallenged. The respect for the principle of data integrity free of political interference must be protected. The Trump administration should rescind this decision immediately and apologize for this transparent attempt to violate the public trust.

    ------------------------------
    Stan Altan
    ------------------------------



  • 24.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 17:27
    Stan's idea to write a letter to the editor sounds good.  However, my local paper is the Washington Post,  and over the years I have gotten maybe three letters published there, and one in the New York Times.  So I modified Stan's idea and just decided to make sure my US House representative knows about both the ASA Board releases on the recent Census appountments, and the current CDC data situation.  In two separate submissions, I first gave the relevant URL in each case, and then, especially in case they cannot open links for security reasons, I copied the most relevant paragraph or so into the message.  

    If anyone wants to do this in lieu of, or in addition to a letter to the editor, here again are the two URLs noted above: 

    https://www.amstat.org/asa/files/pdfs/POL-June23CensusBureauAppointments.pdf

    and

    https://www.amstat.org/asa/files/pdfs/POL-July10Covid19HospitalReportingGuidelines.pdf

    Also, please note that if you do write a letter to the editor of a newspaper, your chances of being published are greatly enhanced if you are commenting on an article or editorial published in that paper in the last day or so.

    Cheers.


    ------------------------------
    James Knaub (Jim)
    Retired Lead Mathematical Statistician
    ------------------------------



  • 25.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 17:05
    Dear Colleagues,

    I fully agree with the statements of Ms Maldonado, Mr. Bycott, Mr. Alton, and others above, that this request "to have the US hospitals to send data NOT to the CDC" appears very clearly to be data suppression attempt.  I support all civil efforts by ASA to be opposed this action. 

    I would expect this type of government action from some authoritarian third or second world government, but not an American one!

    Best regards and good health to you all,

    David L Kimble





    ------------------------------
    David Kimble
    Statistician / Data Scientist
    Consultant - Self Employed
    ------------------------------



  • 26.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 10:45
    Hi All,

    I agree that I would like to see the ASA leadership respond against this proposal by the White House.  I personally would not trust anything from this administration at that White House.  I have also reached out to the American Society of Clinical Oncology, which I am a member of as well, to send a letter condemning this request.


    ------------------------------
    Paul Bycott
    Pfizer Inc.
    ------------------------------



  • 27.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-27-2020 19:16
    I am still puzzling.  I am finding these bits of information (cited below) at HHS and CDC websites about the new data repository and wondering what it all means.
    If anyone has more insight into whether the CDC is encouraging this transfer of data handling policy, I'm interested to understand better.

    https://www.cdc.gov/media/releases/2020/s0716-covid-19-data.html
    Seems CDC director Redfield helped bring about the change in reporting?  These are his words below:
    "In April, HHS leaders, with input from CDC, created a new system, called HHS Protect, that allows us to combine data through systems like NHSN, as well as other public and private sources. The data reported from hospitals that went into HHS Protect either came through the NHSN, directly to HHS Protect from the states, or through a system called TeleTracking. What we have now asked is that, going forward, states provide data from hospitals directly through the TeleTracking system or directly to the HHS Protect system."

    https://www.hhs.gov/about/news/2020/07/20/hhs-protect-frequently-asked-questions.html
    HHS is claiming that CDC has direct access to data:
    "Yes, hundreds of CDC staff have access to the system. CDC has been delegated the ability to directly authorize CDC personnel for access to HHS Protect. Appropriate CDC personnel have sole authority to grant access to CDC personnel for HHS Protect."

    https://www.propublica.org/article/out-of-view-after-public-outcry-cdc-adds-hospital-data-back-to-its-website-for-now
    https://www.politico.com/news/2020/07/16/who-took-down-the-cdcs-coronavirus-data-the-agency-itself-367018
    So then why did the CDC take the data down from its website in the days following the change in policy?
    From former head of CDC Tom Frieden:
    "We're in the middle of the worst pandemic in 100 years. More than 137,000 Americans have died. Covid is exploding in Arizona, Texas, South Carolina, Florida, and other states. And the administration has chosen to sideline the CDC. Where did this data go?"

    And if the government is managing the data well, why is an independent public health organization led by connected to Tom Frieden issuing recommendations to states about what and how frequently they should report? (Perhaps this is a bit off the point.)
    https://preventepidemics.org/covid19/resources/indicators/


    ------------------------------
    Amalia Magaret
    University of Washington
    Seattle
    ------------------------------



  • 28.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-27-2020 19:52
    This, along with the recommendation to open schools when many children are over 10 and the younger ones who may be less contagious are going to make up for that by coughing on teachers, etc, would indicate that the CDC has been subverted/hijacked from the scientists.  

    Changing the data system at this point, and what they have done, is at least sloppy practice.  That would be the charitable interpretation, Amelia.

    ------------------------------
    James Knaub (Jim)
    Retired Lead Mathematical Statistician
    ------------------------------



  • 29.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-27-2020 19:58
    In addition to everything Amalia said, there is a troubling emphasis upon security of the data in the general descriptions given HHS Protect by, for instance, HHS Chief Information Officer Jose Arrieta. Specifically, he stated:

    You heard me mention security, which I want to emphasize we take very seriously.

    Access is only granted to authorized federal/military employees and contractors, who are granted access as necessary by mission need. We authenticate and authorize every user to ensure only mission essential activity is occurring within HHS Protect. All data in HHS Protect is de-identified, meaning that there is no personally identifiable information attached. HHS has made the security and protection of the data involved a top priority. Least-privilege and National Institute of Standards and Technology (NIST) cybersecurity frameworks have been applied to support confidentiality, integrity, and availability. These are actually higher standards than are applied to protecting healthcare data in many other parts of the American healthcare system. Controls and platforms are tested for vulnerabilities, which are mitigated quickly, and mechanisms are in place to prevent exfiltration of data.

    CDC has complete control of who access their data and what users at CDC log into HHS Protect. Currently, we have 1,200 users and approximately 950 state CDC partners and CDC partners.

    You also heard me mention transparency: We recognize that experts creating and using these data sets want to know their sources and the lineage of the data sets and how they're being used. We do that through HHS Protect, and we also give the actual owners of data sets-CDC for NHSN, for instance-control over who within HHS Protect has access to those data sets.

    In the case of a scientific enterprise, "preventing exfiltration of data" is the opposite of transparency. Data is validated by continuous checking and criticism. Transparency isn't merely knowing "their sources and the lineage of the data sets". Moreover, giving the originators of the data sets, say, the state of Florida, "control over who within HHS Protect has access to those data sets" is likely to create a sampling nightmare. Consider, for example, if a state doesn't want to release a dataset for a particular week, and other states pick other weeks.

    Moreover, the key problem I have from what I know now, is how, using such a system, whoever makes it, will the error checking process which apparently is used in the CDC's original NHSN system be done:

    Where can I see what data will be sent to CMS from my NHSN facility?

    • Facilities can see the NHSN data that will be submitted to CMS using the special NHSN analysis output options for their specific facility type. To find the reports applicable to your facility type, log into your NHSN facility and go to Analysis > Output Options > CMS Reports > CDC Defined Output. Detailed guides for how to run and interpret the CMS reports can be found on the NHSN CMS Requirements webpage.

    What if the data in CMS' Hospital Compare don't match the data that I had in NHSN?

    • Please keep in mind that any data entered or modified within NHSN after the quarter freeze date will not be reflected in the data posted on Hospital Compare. Also, the frozen data from each quarter are aggregated to produce the measures on Hospital Compare. If you still have questions regarding the HAI data from your facility that is displayed on Hospital Compare, please contact the NHSN Helpdesk (NHSN@cdc.gov).

    CDC data validation procedures are given here.

    Where are the comparable ones for HHS Protect?  And where is the public dashboard for HHS Protect? It's supposed to be here.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 30.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 16:02
    This is another unsurprising attempt by a data and science averse administration to seize the ability to manipulate data. This crosses a red line that the ASA board has rightfully oppsoed:  https://www.amstat.org/asa/files/pdfs/POL-July10Covid19HospitalReportingGuidelines.pdf .
    In a weird kind of way, we have to be grateful to Trump and his administration for bringing about a national conversation on the critical need to maintain data integrity and respect for science free of political interference. I believe this principle is something all of us as statisticians can appreciate and support. Thank you ASA board for this important statement.

    ------------------------------
    Stan Altan
    ------------------------------



  • 31.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-16-2020 20:52
    Dear All, 

    In addition to the ASA board statement shared earlier today, the ASA has signed onto a multi-society letter on this issue. If you belong to other professional associations, you could encourage them to also sign on: 

    I wanted to share this sign-on letter opportunity with you regarding the recent decision to bypass CDC in the collection of of COVID-19 patient data - deadline is tomorrow, Friday, July 17 at 3pm ET.
     
    Protect COVID-19 Patient Data Collection:  The Infectious Diseases Society of America, HIV Medicine Association, American Society for Microbiology and American Public Health Association are circulating this letter urging the administration to reverse its decision to bypass the Centers for Disease Control and Prevention (CDC) in the collection of COVID-19 patient data. The letter urges the administration not to bypass the CDC to maintain the integrity of COVID-19 data, to keep public health data public, to invest in data reporting at the CDC and highlights the importance of the data to inform state and jurisdictional responses.  Please complete this form to endorse the letter by Friday, July 17 at 3 pm ET.
     
    Best,
    Steve

    ------------------------------
    Steve Pierson
    Director of Science Policy
    American Statistical Association
    ------------------------------



  • 32.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 16:40
    I read the HHS statement that Jan Galkowski included in his message, and it appears the stated purpose of the procedure was to reduce duplication of reporting efforts.  Here are the first two paragraphs of that document:

    "On March 29, 2020, Vice President Pence sent a letter to hospital administrators across the
    country requesting daily data reports on testing, capacity and utilization, and patient flows to
    facilitate the public health response to the 2019 Novel Coronavirus (COVID-19). Many separate
    governmental entities are requesting similar information, resulting in stakeholder requests to
    reduce duplication and minimize reporting burden. This document details the Federal
    Government's data needs, explains the division of reporting responsibility between hospitals
    and states, and provides clear, flexible options for the timely delivery of this critical
    information. The objective is to allow states and hospitals either to leverage existing data
    reporting capabilities or, where those capabilities are insufficient, to provide guidance in how
    to build upon existing capabilities. These FAQs will be posted to the various HHS and HHS
    division websites, and will be as necessary.

    "It is critical to the COVID-19 response that all of the information listed below is provided on a
    daily basis to the Federal Government to facilitate planning, monitoring, and resource
    allocation during the COVID-19 Public Health Emergency (PHE). This data will be used to inform
    decisions at the federal level, such as allocation of supplies, treatments, and other resources.
    We will no longer be sending out one-time requests for data to aid in the distribution of Remdesivir or
    any other treatments or supplies. This daily reporting is the only mechanism used for the distribution
    calculations, and the daily is needed daily to ensure accurate calculations."

    While it appears that much is being read into this document, its purpose is to streamline the reporting.  I would recommend that people read the documents themselves before rushing to judgment.  CDC is itself under the umbrella of HHS, so I don't see how this will reduce the data available to the CDC.  Am I missing something?


    ------------------------------
    Raoul Burchette
    Biostatistician
    ------------------------------



  • 33.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 17:37
    Dr Burchette,

    I'll be interested in hearing what others think, but I have three comments.

    First, this action needs to be interpreted in context with all the other actions by this administration, most pertinently the criticism of CDC by HHS and others as noted by, for example, Dr Josh Sharfstein of the Johns Hopkins School of Public Health

    Second, while reducing the number of reporting channels may seem smart from an effort and business perspective, with public health data and its myriad sources, namely, multiple states, multiple jurisdictions within states, and multiples of hospitals, there are hundreds of opportunities for inconsistencies and errors to creep in.  In the end, these only get ironed out if several groups and many people review and learn of these. By imposing this burden entirely upon HHS and, in particular, by imposing it upon a third party private contractor, that scruitany will be removed and these inconsistencies no longer reflected in the interest of efficiency. 

    Third, this is not the time to make such a major change. By imposing a format change mid-pandemic, a format which has not been reviewed for reasonableness with hospitals and others, whatever the long term efficiencies and betterment of such a system, this will burden already overloaded hospital administration, slow the reporting of COVID-19 cases and data, and introduce inconsistencies in itself. As a quantitative engineer and data scientist with 40 years experience, rolling out such a major change without a transition period in which old and new systems operate concurrently so these problems can be identified and debugged is foolish and risky. And it is incredibly unwise to do it in the middle of an emergency.

    Because this third element is so well known in the information processing industry, that it is being proposed either means that the people choosing to do it now have no experience with operating such big systems and the consequences of doing these poorly, or there is an ulterior motive to this change.  Readers imaginations can supply what some of those motives might be. This would, in any other circumstance, seem paranoid, but the context I mentioned in the first comment gives much cause for pause. 

    As I noted at the outset, a number of us worried about the state of climate and environmental data on the U.S. government's public facing Web sites after the November 2016 election.  Anticipating problems, and worrying about the possibly deliberate corruption or falsification of such data, a large group of people undertook the systematic copying and archiving of all environmental datasets we could reach.  In the end we were successful, even if the archives are no longer being updated. The Commonwealth of Massachusetts was one of the entities which took the threat seriously enough to do a replica themselves. And, eventually, all those datasets were removed in the sense that they were deemed part of an out-of-date version of the Web sites, and all quantitatively-backed mentions of climate change at the EPA sites and interior sites were stricken.  Eventually the links to the out-of-date Web sites disappeared. 

    With the pattern of deceit that's been documented in so many other areas, and the motives to understate the seriousness of the pandemic and its effects upon the people of the United States being so keen, controlling information by making it flow through a single choke point rather than a leaky public and transparent CDC seems a substantial basis for concern, one shared by many, including the Republican Governor of the Commonwealth of Massachusetts, Charlie Baker.




    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 34.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-17-2020 20:36
    ​Jan,

    I think you summarized it perfectly and I completely agree.  I will take it one step further and to be blunt that I have no faith in our administration to do the right thing and to report accurate and timely results.  This administration has repeatedly, from my perspective, shown poor judgement.  Also, Trump has banned the CDC from testifying at hearings next week at the educational board regarding whether students should go back to the classrooms or not given where we are right now with the virus and the risk also to staff.  My only response to this is "Seriously!!!".  There was an expression I heard earlier in the pandemic regarding this administration and it is ringing more true then ever:  "Dollars over deaths!"

    ------------------------------
    Paul Bycott

    ------------------------------



  • 35.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 14:05
    I cannot say I find support for these three claims against this administration.  It is a reasonable thing to do to streamline the data flow.  People can still get access to the data at HHS, they just need to change the search location.  The hospitals have plenty to do these days as far as reporting and caring for patients.  If people were inappropriately manipulating the data before it was released for general consumption, this is a problem, but I don't know what the evidence was for that suggestion (mentioned by someone in the audience during the news briefing).  I don't know if it can be proved that there was any inappropriate manipulation of the data.  I don't have any problem with taking the action at face value, and I feel it is folly to rush to judgment.  I would feel there was a problem if there was a difference between what was said and what was written.  I don't feel safe trying to guess between the lines.  It is hard to be objective while trying to read (guess) between the lines.  I do not appreciate my organization being moved to publicly object to something based on speculation and assumptions about its "motives".  There are plenty of things to go on record for that are plain and worthwhile.  This is not one of them.



    ------------------------------
    Raoul Burchette
    Biostatistician
    ------------------------------



  • 36.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 14:52
    Dr. Burchette,

    Note that the US Forest Service collects and reports on timber production and forest reserves.  Not the White House.  Elsewhere in USDA reporting on acres of crops and meat production takes place.  Not the White House.  The Census Bureau does the census.  Not the White House.  (And on and on...)

    And you support moving the reporting and data collection from CDC to the White House because the White House can do it better?

    "T
    here are none so blind as those who will not see."

    Jim


    ------------------------------
    Jim Baldwin
    Retired
    ------------------------------



  • 37.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 16:09

    I am sorry, I didn't read anything about reporting to the White House in the directive.  The data was supposed to be sent to HHS.  What makes you think the White House is going to look at the data?  If the White House did, I would say, good for them, but the data is going to HHS, not the White House.  I am not following your reasoning.



    ------------------------------
    Raoul Burchette
    Biostatistician
    ------------------------------



  • 38.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 17:04
    Using the term "the Trump-appointed administration" might have been more justified than the term "the White House" but reading between the lines that you state you are reluctant to do, does not result in any practical difference.

    Jim

    ------------------------------
    Jim Baldwin
    Retired
    ------------------------------



  • 39.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 17:55
    Good job Jan.

    ------------------------------
    James Knaub (Jim)
    Retired Lead Mathematical Statistician
    ------------------------------



  • 40.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 20:24
    Dear Colleagues,

    The points enumerated by Jan Galkowski seems reasonable and supportable by me.

    I see no value in the the "hospital time saving on reporting" argument when the reporting is still going to happen but not to the CDC.

    The conclusion of "why any government, in the middle of an out-of-control pandemic with less than 4 months to a general reelection, would change the statistical reporting procedures of numbers that reflex poorly on their handling of that pandemic to be numbers in their immediate control?"  looks more than suspicious is justifiable.  I think most reasonable people would agree with this conclusion!   

    As a professional statistician, I support -as I am sure you do- the concept that US national statistics should to be "above" the suspicion of being modified by any government action that "might" politically benefit from these modifications. This reporting change would looks suspect, regardless of the country or the government is "requiring" it.

    I think you all will agree with me: "We need objective vital statistics!" to make informed national decisions especially during this pandemic.

    Best regards,    

    David Kimble



    ------------------------------
    David Kimble
    Statistician / Data Scientist
    Consultant - Self Employed
    ------------------------------



  • 41.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 09:42
    Let's say we accept that as an argument for hospitals to begin sending data directly to HHS. Fine. Totally reasonable. What could possibly be the rationale for hospitals to STOP sending data directly to CDC? As Jan said to start this thread, it's pretty simple to just add an additional job to a script.

    ------------------------------
    Robert Pearson
    Asst. Professor
    Grand Valley State University
    ------------------------------



  • 42.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-20-2020 15:23
    Are the data now available on the HHS website, and if so, where can the data be found?

    ------------------------------
    Mark Pierzchala
    Owner
    MMP Survey Services, LLC
    ------------------------------



  • 43.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-21-2020 20:59
    Edited by Jan Galkowski 07-23-2020 20:12
    (slight update to add OWiD and COVID Tracking Project  to the list of alternative compilers)

    Just to follow-up a bit. Thanks for everyone's enthusiasm on this issue.

    I just noticed that the Web site and organization usafacts.org says the following about sourcing COVID-19 case data:

    This interactive feature aggregates data from the Centers for Disease Control and Prevention (CDC), state- and local-level public health agencies. County-level data is confirmed by referencing state and local agencies directly. Cases, deaths, and per capita adjustments reflect cumulative totals since January 22, 2020.

    For most states, USAFacts directly collects the daily county-level cumulative totals of positive cases and deaths from a table, dashboard, or PDF on the state public health website. This data is compiled either through scraping or manual entry.

    This can be found on their methodology page

    Not to disparage other compilers of data or presenters, such as Johns HopkinsNew York Times, Florida COVID Action, GoogleFinancial Times, Avi Schiffmann, OurWorldInData.orgThe COVID Tracking Project, and others, which all contribute their own useful views, but the usafacts.org visualization is quite nice.

    Interestingly enough, the CDC itself now recommends them as a data source


    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 44.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-22-2020 10:40
    I'm usually a day or so behind in reading the Connect Digest, so others may already have pointed this out.  Also, retired, I'm not necessarily the most up to date and I don't usually comment in Connect.  But I would point out that CDC is ordinarily the recipient of all data on notifiable diseases and, currently, COVID appears on this list (https://wwwn.cdc.gov/nndss/conditions/notifiable/2020/).  This, among other things, ensures a measure of consistency in the data handling and also means not only that hospitals need only report to one place but the public can consult data in one place, not multiple places.  


    ------------------------------
    Elizabeth Margosches
    ------------------------------



  • 45.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-28-2020 13:54
    Why not have two parallel systems, which includes CDC, until the new system has proven its muster. Anyone heard of SNAFUs? American GIs came up with the acronym, It would be prudent and consider that and many other experiences -- Anyone heard of the 737 Max Jet, the Apollo 1 fire, Challenger, Titanic, even the start of the ACA until it righted itself etc. etc.?,

    ------------------------------
    Mark Czarnolewski
    ------------------------------



  • 46.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-28-2020 14:13
    Mark,

    That is an excellent suggestion.  And, in fact, it is standard practice for any large data or accounting system, whether in the sciences or in business. I should hope HHS CIO Arrieta knows that. 

     - Jan

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 47.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 07-28-2020 16:42
    Jan,
    Thanks. But, unfortunately, skepticism and wariness are necessary, given we are in survival mode. So, "know" and  "do" are two separate things; and the corollaries. "how much know" and "how much (can) do (despite knowing better)" are in order with the current regime. One need only look at the testing rollout debacle, the PPE, testing  and contact tracing limitations.  How much pressure was there to have that "third test" in the test kit, despite people "knowing" it was unnecessary; how much "group think" took hold while people rushed despite their skill set and better judgment?
    "Trust, but verify", again, unfortunately is in order.
    Mark

    ------------------------------
    Mark Czarnolewski
    ------------------------------



  • 48.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-05-2020 16:32
    Dear ASA Members,

    Thanks to the many of you who spoke up on the new HHS Covid10 hospital reporting guidelines and shared your thoughts in this forum. I also appreciate that at least a few of you had letters to the editor published. 

    We are now at three weeks since the switch. I'm aware of a few news articles on the impacts to date: Bloomberg, NPR, and COVID Tracking Project blog posts (led by The Atlantic):

    I'd like to hear more first-hand reports by statisticians for the impacts on data quality from this switch. Please share them with me: pierson@amstat.org.  

    Thank you,
    Steve



    ------------------------------
    Steve Pierson
    Director of Science Policy
    American Statistical Association
    ------------------------------



  • 49.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-12-2020 15:59
    Update from The New York Times : "34 Health Experts Warn New Data Rules Left Hospitals 'Scrambling' and Could Mar Data Integrity".

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 50.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-12-2020 21:38
    Hi Jan,
    One initial approach is to focus on states that started to surge from the end of May till about now. This would include states, such as Florida, Texas, Georgia, Arizona, California and Louisiana and compare their trends to earlier states that experienced surges, such as New York, New Jersey, Connecticut, Mass, Maryland and Michigan. You can correct me, but I recall that the  earlier states had trends that were accelerating a great deal, then accelerating somewhat less, then plateauing for a while and finally started decelerating. (Michigan had gone down, but I think accelerated again when the auto plants opened up.).  Are the more recent states with surges following the same pattern, or are they plateauing for a shorter time period and decelerating sooner? Are all of these recent surge states following similar patterns (e.g., does California have the same pattern as the other recent surge states mentioned?) My hypothesis is that, at least most of the latter states, started reporting their data differently. I and others noted inconsistencies in Florida's reporting of their data and others found questionable trends in other states. A second hypothesis deals with the criteria for reporting Covid deaths. An analogy is drunk driver deaths. If a drunk driver crashes into another car and the victim in the other car experiences a fatal cardiac arrest will the drunk driver be charged with vehicular homicide or just charged with damaging the other car and not charged with the victim's death? Guess what the judge will decide?. Well, Alabama stopped counting a patient experiencing cardiac arrest after being hospitalized with Covid as a Covid death. Are the latter surge states changing their definitions of a Covid death mid-stream in the surges, how do the trends coincide after the change of definitions? If the new DHHS reporting system is not as current  as the CDC system, all things being equal, the trends could still be the same, but over much larger aggregated time periods. If the definitions have changed, then the trends would not be the same even over aggregated time periods. Obviously, given predictors of face mask policy, population density, ethnicity, concentration of underlying conditions, etc. can, perhaps,  further clarify the impact on trends from definition changes, reporting changes and system gathering changes..

    ------------------------------
    Mark Czarnolewski
    ------------------------------



  • 51.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-13-2020 10:03
    Sounds like a good project for a grad student or a stats guy looking for projects to look at trends over time related to various events, such as data reporting changes, state policies, local events...


    ------------------------------
    Michael Mout
    MIKS
    ------------------------------



  • 52.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-13-2020 11:05
    Perhaps, others can look at the trends, too. The papers, e.g., NY Times, Wash Post,, have had research groups analyze data so that their staff can write the results to a lay audience. Perhaps, papers in the hots spots, like Miami-Dade and Houston can do the same. Again, the public wants to know and is dependent on  accurate and timely Covid information, That's the issue. I would extend the same point of accurate and timely to testing. Admiral Giroir of the U.S. Public Health Service Corps recently testified that he briefs the public on the testing data. But does he release the data for more extensive research? For example, stating the number of tests administered, does not consider the number of times a person is tested. People typically go through multiple testing because that's the protocol; One protocol is:  if you're positive, get two follow up negative tests to be considered cleared of the virus.. So, how many different people have been tested? Anyone know? Is turnaround the same in different parts of the country, different parts of a state? Anyone know?

    ------------------------------
    Mark Czarnolewski
    ------------------------------



  • 53.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-13-2020 13:36
    DataLumos, ICPSR's archive for safekeeping and disseminating US government and other social science data, now lists these two datasets:
     


    ------------------------------
    Margaret Levenstein
    ------------------------------



  • 54.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-13-2020 13:52
    Thank you.

    ------------------------------
    Mark Czarnolewski
    ------------------------------



  • 55.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-14-2020 18:11
    "The manager of the Trump administration's new virus database refuses Senate questioning, citing a nondisclosure agreement"


    From The New York Times.

    Quoting from a letter by the attorneys representing TeleTracking

    TeleTracking recognizes your concerns and seeks to fully cooperate in response to your questions regarding the Contract. However, we are obligated to advise you that the Contract is subject to a broad Non-Disclosure Agreement (NDA) that restricts TeleTracking's release of non- public information,including, but not limited to, confidential privileged and procurement- sensitive information.Notwithstanding these contractual restrictions, we are pleased to cooperate and provide other information ,as permitted, in response to your letter.

    That letter also includes the standard language HHS is using in response to criticism, namely, 

    On the back end, whether collected by the CDC's system, the third party vendor, or the states, the data ends up aggregated in the HHS Protect platform , where the CDC team and other federal response teams still have access to this information for their use in the response. Additionally , state and local public health departments also have access to this information in HHS Protect which allows them to access and use the same information that the federal response teams are using.
    No one is taking access or data away from CDC.
    .
    .
    .
           This has no effect on CDC's ability to use this data and continue churning out the daily data, the MMWRs, and the guidance we publish . In fact, the new
           infrastructure we have now actually provides our CDC team with easier access to a much broader variety of data sets than they would have without it.

    Apparently, then, the methods and procedures being used to clean and rationalize fusings of data from different sources are protected by this NDA and so are not subject to any kind of independent review, peer or otherwise. 

    I don't know, and no one here has responded in detail to Dr Steve Pierson's comment, but it looks to me that the only way to assess whether or not this kind of cleaning is being done properly is to have one or more independent fusion centers collecting the same data and to make comparisons. 

    The COVID Tracking Project has assessed the state of information here.  They say

    These problems mean that our hospitalization data-a crucial metric of the COVID-19 pandemic-is, for now, unreliable, and likely an undercount. We do not think that either the state-level hospitalization data or the new federal data is reliable in isolation. (As we describe below, the new federal hospitalization figures are substantially higher than the same data as reported by most states.)

    That said, these problems only affect hospitalization data. Our case, testing, and death data from states continues to show the effects of testing and reporting backlogs and day-of-the-week reporting differences, but are as reliable as they were before the new directive. And we find no evidence to support a popular online conspiracy theory that the switchover from the CDC system to the Health and Human Services system explains a national plateau in new coronavirus cases. The theory is unsupportable: As we explain below, hospitals do not report case count or testing data to the federal government, and those trends in the data are reflected across too many different independent sources to be subject to centralized tampering. 


    Emphasis is in original.



    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 56.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-15-2020 21:34
    So,the DHHS conversion of  the hospital data is of concern, but not the death data. First, the hospital data still address specific data requests, among which are time dependent data. Second, the death data may not be affected by the DHHS conversion perhaps, the word "conversion" is being used in more ways than one. The death data are also of concern. If for example, states are changing the criteria for attributing deaths as being due to Covid-19 in "mid-surge", which may fortuitously have the data surges slow down, with resulting decrease in expected surge numbers as compared to surges earlier in the year for similar types of geographic areas and demographic groups..

    ------------------------------
    Mark Czarnolewski
    ------------------------------



  • 57.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-17-2020 12:43
    One of the most reassuring things I've found of late is the COVID-19 Forecast Hub which assimilates forecasts from up to 49 different COVID-19 modeling and forecasting groups, using them to produce an ensemble forecast and, at the same time, showing through their dashboard how different forecasts compare with one another. 

    What's heartening about this is not only the principal purpose of the project, which is noteworthy, having wished many other fields had such an ongoing open competition among scientific models and standard figures of merit for them, but that it enlists the combined brilliance and approaches of so many different teams, each have different methods, perspectives, and, necessarily, different sets of data feeding their models.  Some have a single U.S. state as their scope, some the entire United States population, some the world. 

    While these groups use U.S. data sources, those are not their only inputs, as they rely upon the standard New York TimesAtlantic MagazineJohns Hopkins, and USAFacts.org sets, but many others, some grounding themselves in reports from individual counties across the United States, for instance.  

    I have cloned and regularly update the Githubs of many of these groups, so I see the work they do.  When Github changes are rolled in, justifications for changes are included, so I have watched many of them wrestle with data irregularities in counties, states, and countries, repairing them. 

    My primary point is that no matter what the HHS and CDC do, there is now this independent replica of status of the pandemic out there, which can be used to checkpoint anything resulted from HHS or the administration.  I don't think these can be biased. They compare and cross-check against one another, and publicly assess their short term and, where they have one, their long term forecast accuracy. 

    Indeed, "looking over the shoulders" of these teams at their Githubs is why I wondered about the data cleansing efforts which CDC invariably had to do being maintained by TeleTracking and Palantir for HHS. 

    But it no longer matters. For biostatistical research and other purposes, these decentralized projects exist and are cloned and safeguarded many times over, so our collective scientific memory of these events are assured. Even underreporting at the county level can be exposed. To the degree to which these forecasting models are sensitive to their diets, if something untoward did happen, the models would react, and if the change was abrupt, their operators, in my opinion, would investigate. It's unfortunate that the federal government cannot keep up. But I think we're safe, thanks to the enormous collective and continuing efforts of all these teams. 

    I think they all someday should receive some kind of award from the ASA.

    The only area which remains pretty weak is knowledge of "recovered cases". The sobering reasons is we do not have a medical definition (yet) of what it means to be "recovered" from COVID-19. 

    Thanks.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 58.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-17-2020 14:12
    Thank you.






  • 59.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-17-2020 14:29
    Thank you--that is indeed reassuring. Is there a way to bring the COVID-19 forecast hub to the attention of the "powers-that-be" and do you think it would be  worthwhile?

    I wonder about the reaction of the general public if the HHS or administration were caught making false claims, and felt the need to double down on their statements and discredit the reports on the COVID-19 forecast hub. There are plenty of people in this country who would trust the word of an administration they support over the conclusions of independent scientists.

    ------------------------------
    Sudip Bose
    Department of Statistics
    George Washington University
    ------------------------------



  • 60.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-17-2020 19:28
    Hi, Sudip, and thanks. 

    I see little benefit of bringing extra attention.  I'm curious, though, what you thought might be the outcome?  Surely, if the decentralized set of data are functionally as good or better than what comes from HHS Protect, people can just use those. Someone should compare the composite data with what HHS Protect is producing. Not sure if any of the individual projects are doing that. They might, especially operations like The New York Times or The Atlantic

    But what do other people think?

     - Jan

    ------------------------------
    Jan Galkowski
    Westwood Statistical Studios
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 61.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-17-2020 21:26
    You're welcome. A primary concern is that the people who take the word of the administration over independent scientific studies would refuse to take precautions such as wearing masks, and practicing social distancing. Also, it would encourage anti-vaxxers. There seem to be large numbers of people who think that health warnings about Covid-19 are overblown. False claims suggesting that it's been contained in the US could lead to resistance against basic safety practices.

    ------------------------------
    Sudip Bose
    Department of Statistics
    George Washington University
    ------------------------------



  • 62.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-19-2020 18:25
    Important letter related to this subject from U.S. House Committee on Energy and Commerce to Comptroller General, U.S. GAO.

    Note in particular the questions asked on its pages 3 through 5.

    Related article in The New York Times.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 63.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-20-2020 14:42
    From the Wall Street Journal:

    Troubled Covid-19 Data System Returning to CDC

    The U.S. Department of Health and Human Services is reversing course on a change to the way hospitals report critical information on the pandemic to the government, returning the responsibility for data collection to the Centers for Disease Control and Prevention.

    The HHS takeover of the process was plagued by delays and inconsistencies.

    https://www.wsj.com/articles/troubled-covid-19-data-system-returning-to-cdc-11597945770?mod=djemalertNEWS

    ------------------------------
    David Mangen
    ------------------------------



  • 64.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-20-2020 15:54
    Like many on this forum said. Don't make major changes to a system in the middle of an emergency.

    ------------------------------
    Michael Mout
    MIKS
    ------------------------------



  • 65.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-21-2020 15:58
    HHS denies report COVID-19 hospital data going back to CDC
    The Department of Health and Human Services says a report by the Wall Street Journal about the future reporting of hospital COVID-19 is wrong.

     
    For what it's worth, here's the NPR report along the lines of the WSJ piece: https://www.npr.org/sections/health-shots/2020/08/20/904450628/white-house-stokes-hopes-that-key-hospital-data-tracking-will-soon-return-to-cdc



    ------------------------------
    Steve Pierson
    Director of Science Policy
    American Statistical Association
    ------------------------------



  • 66.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-24-2020 15:03
      |   view attached
    Dear All, 

    Today, we transmitted the attached letter from ASA President Elect Rob Santos to HHS Secretary Azar urging "NHSN be restored as a direct recipient of the COVID-19 hospital reporting data concurrently with TeleTracking" but not in a way that newly burdens the hospitals reporting the data. The letter is also available at https://www.amstat.org/asa/files/pdfs/POL-HHS_Covid19HospitalReporting.pdf

    Best Wishes,
    Steve

    ------------------------------
    Steve Pierson
    Director of Science Policy
    American Statistical Association
    ------------------------------

    Attachment(s)



  • 67.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 08-24-2020 18:01
    Thank you Steve, and Professor Santos.

    There was also an article in the New York Times yesterday, updated today, which tied federal reimbursements to hospitals for COVID-19 to their cooperation with TeleTracking in collecting data, making cooperation a prerequisite for payment, but not promising any specific amount. These were the funds voted for relief by Congress.

    The email was from Secretary Azar. 

    Not sure there'll be much cooperation there.

    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 68.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 09-04-2020 08:33

    Nature just published an article on the slowdown in COVID-19 data dissemination & sideline of the CDC. While countries like South Korea, Singapore, and New Zealand update publicly sourced outbreak data in real time--fast enough to be of use in controlling the outbreak--"... By contrast, the United States offers vanishingly few details on how the disease is spreading, even as people increasingly socialize and travel, and authorities reopen schools and businesses.
    "...political meddling, privacy concerns and years of neglect of public-health surveillance systems are among the reasons for the dearth of information in the United States.

    https://www.nature.com/articles/d41586-020-02478-z?utm_source=Nature+Briefing&utm_campaign=2dc8b44334-briefing-dy-20200901&utm_medium=email&utm_term=0_c9dfd39373-2dc8b44334-44664489



    ------------------------------
    Dennis Sweitzer
    Principal Biostatistician
    ------------------------------



  • 69.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 12-10-2020 19:38
    On the continuing saga ...  CDC official says she was ordered to destroy an email showing a Trump appointee interfering with publication MMWR.


    ------------------------------
    Jan Galkowski
    Westwood Statistical Studios
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 70.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 04-30-2021 10:32
    Edited by Jan Galkowski 04-30-2021 10:32
    The COVID Tracking Project, an effort partly supported by The Atlantic, a venerable news and opinion magazine previously known as The Atlantic Monthly, had, as several other groups, been attempted to track COVID-19 cases, deaths, and hospitalizations from many sources.  Because many groups are more suited to doing this ongoing, such as the famous John Hopkins engineering group and, now, a refurbished federal data gathering apparatus, the COVID Tracking Project decided to shut down in March 2021.  

    In that process, they are making public notes they had gathered about anomalies in states and other reporting, and making that available for scholarship.  I decided to include these links to those sources as a reply to this thread because the general theme here is data gathering, availability, and reporting in a crisis.  I have not previously linked it here, but we now know, due to emails found by the current administration, that higher HHS and CDC officials were censoring and modifying findings of CDC data and cases. 

    For an organization committed to the use and curating of datasets, the need to have key data sources protected, and protected especially in a politically charged crisis could no better be highlighted than the experience we have all had.  The results from the COVID Tracking Project suggest that integrity of reporting must be guaranteed back into its sources, in many cases, these being state health departments which are subject to political influence and manipulation. 

    The COVID Tracking Project states data log.

    How the COVID Tracking Project entered testing and outcomes data every day for a year.

    How federal COVID-19 testing data is getting better

    How inconsistent reporting practices hampered the COVID Tracking Project's ability to analyze COVID-19 data: Three common problems.

    COVID Tracking Project's annotations on state COVID-19 current hospitalizations data.











    ------------------------------
    Jan Galkowski
    Westwood, MA 02090

    bayesianlogic.1@gmail.com
    ------------------------------



  • 71.  RE: HHS directing hospitals to cease sending data to CDC

    Posted 04-30-2021 16:52
    Thank you for posting this Jan. This is very valuable information. I haven't had a chance to review all of the links you referenced, but on the face of it,  the evidence is compelling that there were cases where data integrity was compromised to promote a political agenda. I guess my question is, what is the appropriate role of the ASA in leading an effort to promote laws and practices that ensure that this kind of political interference of the free flow of public information does not happen again. One might argue that this was a one off by an ethically untethered administration and their supporters, but there is no guarantee that this won't happen again in the future. Some attention to what we can do institutionally to protect data integrity and transparency in the public interest is worthy of our time.

    ------------------------------
    Stan Altan
    ------------------------------