4 ms·

Hello. I work for public health. My job is to support the computer systems that allow epidemiologists to track 'superbugs' (and regular ones). Deathcerts is

Hello. I work for public health. My job is to support the computer systems that allow epidemiologists to track 'superbugs' (and regular ones). Deathcerts is an interesting dataset. It's not the only one. Even if death-by-MSRA isn't tracked, I assure you that MSRA itself is tracked. There's a specific class of lab test called "antibiotic susceptibilities". The results report includes a bunch of numbers that represent just HOW resistant a particular bug is to a standard battery of antibiotics, broken down per antibiotic. That data is transmitted from the labs that performed the test to the state health department where it is analyzed. Fascinating article, but there's NO WAY that Reuters had that much contact with state public health folks without being aware that they track lab results. Misleading! Edit: elaborated about the susceptibilities results.

daveloyall 10y ago
Update... I asked around the office and learned more about this. TL;DR: I'm less right than I thought. Specifically regarding tracking cause of death: Not every disease or condition is "reportable". Each state maintains a list of reportable diseases. When a hospital/lab (hereafter: facility) encounters a disease on the list, then they are required to report it. State health departments will open investigations in various circumstances. You've seen events like e. coli outbreaks in the news--those always precipitate an investigation. There are also investigations associated with individual infections, i.e. every case of tuberculosis. If a patient involved in an investigation dies, public health will find out. But, if a patient dies from (because of) an infection that wasn't on the reportable list, then public health won't know the cause of death unless it is on the death certificate. Just like the article says! Regarding tracking the susceptibility of bacteria to antibiotics (superbugs): Any sample that is tested for antibiotic susceptibility is reportable. The results of that testing generally include the specimen source, but don't include why the test was ordered (i.e. because some person has sepsis and might die) and certainly there's no (deliberate) relation between this data and cause of death.
pg_is_a_butt 10y ago
Hello. I read the lies published by Y Combinator. Misleading!
thr0waway1239 10y ago
From the subhead: "government agencies remain unwilling or unable to impose reporting requirements on a healthcare industry that often hides the problem" I ask as someone who is quite a novice when it comes to knowledge of what is the expected reporting requirement: That subhead is a blanket assertion, but when applied to the death-by-MRSA issue, do the public have access to accurate, up-to-date information? If not, isn't that assertion partially correct?
dmix 10y ago
Strange, so that basically refutes the a pretty core point of the article (that it isn't being tracked). The reason they said doctors don't put it on death records is because they dont always want to wait the 2-3 days it takes to get results back from the labs. So it makes way more sense that the labs record the bugs. Instead of doctors in the field. Those are the people who are trained to detect the bugs... so they should be responsible for tracking them. I presume they could get indications from the doctors on whether the patients died as well. Couldn't the doctors update death records after-the-fact? If they found out afterwards what the actual cause of death was, they should really being recording that.
toss1 10y ago
OK, so somebody's tracking it in secret, doing the 'antibiotic susceptibilities' tests and logging the results, and the state health dept receives the data, but the info is being kept from the patients' relatives and from the public in general? This creates two obvious results. 1) suppression of legitimate claims from relatives, and 2) suppression of valid concern and pressure for govt action such as increased research funding. How is this good?
- dv_dt 10y ago
  In particular, a potential patient might want to know MRSA case rates when selecting hospitals or other medical treatment facilities.
  - apathy 10y ago
    Or (perhaps more likely) Medicare/Medicaid and/or private insurers might start penalizing the ones that are screwing it up (by renegotiating contracts and reimbursements). I know of one center where executive laziness led to a $2 million fine (completely unjustified) for a line infection in an immunosuppressed patient. Which was cleared up in hours, probably didn't come from the line to begin with, and got said executive fired (because if there's one thing hospital "leadership" understands, it's money). Actually I have all sorts of examples of the above, but the point is that oftentimes the data is "corrupted" on the way to the collection point(s). Anyways, the executive (VP of some bullshit) got his walking papers and funnily enough, no more reportable line infections that didn't come from lines. So at least sometimes there are happy endings, sort of...
    - thr0waway1239 10y ago
      Sorry, I am a little confused. Why was it a happy ending? Because the VP was just a titular post? Also, looking at your comment, it reads like the problem went away not because the hospital improved, but because the data stopped being reported. This does not look like a happy ending either. And if the line infection was cleared up in hours, how did it also "probably didn't come from the line to begin with"?
kendallpark 10y ago
Daveloyall is correct. I'll add that pretty much every local hospital system has their own database of antibiotic susceptibilities that is updated regularly so that doctors can prescribe the best antibiotic for the microbe populations specific to their region. Pretty nifty.
- nkrumm 10y ago
  If you are curious as to what this might look like, here is the University of Washington 'antibiogram' for 2015: http://depts.washington.edu/labweb/Divisions/Micro/antibiograms/2015_antibiogram.pdf http://depts.washington.edu/labweb/Divisions/Micro/antibiogr...
refurb 10y ago
Fascinating article, but there's NO WAY that Reuters had that much contact with state public health folks without being aware that they track lab results. Misleading! This shouldn't be surprising. Imagine if Reuters wrote an article that said: - Everyone acknowledges that superbugs are an issue - Hospitals are doing their best with their limited resources, but overall are capturing a good deal of data - The data systems are old, but the CDC can get the data and with some manual work process it, but it takes time - Everyone agrees that more could be done, but they also agree it's complex and nothing is perfect. No one would read the article!!
spoiledtechie 10y ago
>>but there's NO WAY that Reuters had that much contact with state public health folks Why Not? They said it was a YEAR LONG investigation...
- Confusion 10y ago
  Read that sentence again. It makes an argument that hinges upon Reuters having had a lot of contact.

daveloyall 10y ago

Update... I asked around the office and learned more about this. TL;DR: I'm less right than I thought. Specifically regarding tracking cause of death: Not every disease or condition is "reportable". Each state maintains a list of reportable diseases. When a hospital/lab (hereafter: facility) encounters a disease on the list, then they are required to report it. State health departments will open investigations in various circumstances. You've seen events like e. coli outbreaks in the news--those always precipitate an investigation. There are also investigations associated with individual infections, i.e. every case of tuberculosis. If a patient involved in an investigation dies, public health will find out. But, if a patient dies from (because of) an infection that wasn't on the reportable list, then public health won't know the cause of death unless it is on the death certificate. Just like the article says! Regarding tracking the susceptibility of bacteria to antibiotics (superbugs): Any sample that is tested for antibiotic susceptibility is reportable. The results of that testing generally include the specimen source, but don't include why the test was ordered (i.e. because some person has sepsis and might die) and certainly there's no (deliberate) relation between this data and cause of death.

pg_is_a_butt 10y ago

Hello. I read the lies published by Y Combinator. Misleading!

thr0waway1239 10y ago

From the subhead: "government agencies remain unwilling or unable to impose reporting requirements on a healthcare industry that often hides the problem" I ask as someone who is quite a novice when it comes to knowledge of what is the expected reporting requirement: That subhead is a blanket assertion, but when applied to the death-by-MRSA issue, do the public have access to accurate, up-to-date information? If not, isn't that assertion partially correct?

dmix 10y ago

Strange, so that basically refutes the a pretty core point of the article (that it isn't being tracked). The reason they said doctors don't put it on death records is because they dont always want to wait the 2-3 days it takes to get results back from the labs. So it makes way more sense that the labs record the bugs. Instead of doctors in the field. Those are the people who are trained to detect the bugs... so they should be responsible for tracking them. I presume they could get indications from the doctors on whether the patients died as well. Couldn't the doctors update death records after-the-fact? If they found out afterwards what the actual cause of death was, they should really being recording that.

toss1 10y ago

OK, so somebody's tracking it in secret, doing the 'antibiotic susceptibilities' tests and logging the results, and the state health dept receives the data, but the info is being kept from the patients' relatives and from the public in general? This creates two obvious results. 1) suppression of legitimate claims from relatives, and 2) suppression of valid concern and pressure for govt action such as increased research funding. How is this good?

dv_dt 10y ago
In particular, a potential patient might want to know MRSA case rates when selecting hospitals or other medical treatment facilities.
- apathy 10y ago
  Or (perhaps more likely) Medicare/Medicaid and/or private insurers might start penalizing the ones that are screwing it up (by renegotiating contracts and reimbursements). I know of one center where executive laziness led to a $2 million fine (completely unjustified) for a line infection in an immunosuppressed patient. Which was cleared up in hours, probably didn't come from the line to begin with, and got said executive fired (because if there's one thing hospital "leadership" understands, it's money). Actually I have all sorts of examples of the above, but the point is that oftentimes the data is "corrupted" on the way to the collection point(s). Anyways, the executive (VP of some bullshit) got his walking papers and funnily enough, no more reportable line infections that didn't come from lines. So at least sometimes there are happy endings, sort of...
  - thr0waway1239 10y ago
    Sorry, I am a little confused. Why was it a happy ending? Because the VP was just a titular post? Also, looking at your comment, it reads like the problem went away not because the hospital improved, but because the data stopped being reported. This does not look like a happy ending either. And if the line infection was cleared up in hours, how did it also "probably didn't come from the line to begin with"?

kendallpark 10y ago

Daveloyall is correct. I'll add that pretty much every local hospital system has their own database of antibiotic susceptibilities that is updated regularly so that doctors can prescribe the best antibiotic for the microbe populations specific to their region. Pretty nifty.

nkrumm 10y ago
If you are curious as to what this might look like, here is the University of Washington 'antibiogram' for 2015: http://depts.washington.edu/labweb/Divisions/Micro/antibiograms/2015_antibiogram.pdf http://depts.washington.edu/labweb/Divisions/Micro/antibiogr...

refurb 10y ago

Fascinating article, but there's NO WAY that Reuters had that much contact with state public health folks without being aware that they track lab results. Misleading! This shouldn't be surprising. Imagine if Reuters wrote an article that said: - Everyone acknowledges that superbugs are an issue - Hospitals are doing their best with their limited resources, but overall are capturing a good deal of data - The data systems are old, but the CDC can get the data and with some manual work process it, but it takes time - Everyone agrees that more could be done, but they also agree it's complex and nothing is perfect. No one would read the article!!

spoiledtechie 10y ago

>>but there's NO WAY that Reuters had that much contact with state public health folks Why Not? They said it was a YEAR LONG investigation...

Confusion 10y ago
Read that sentence again. It makes an argument that hinges upon Reuters having had a lot of contact.