Live data from Hacker News

Google to use patient data to develop healthcare algorithms for hospital chain

theverge.com

31–40 of 57 posts

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#31
Pretty frequently in these scare-generation articles there's no attempt to differentiate between someone using Google Cloud to do something, and Google itself doing something. Is this just a hospital chain making a deal for Google Cloud services to support their normal IT operations, or is Google actually using this data for something? If so, what is Google using the data for?

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#32

I work as a data engineer in an academic medical research institution. Most electronic medical records (EMR) systems are really about billing. Yes, there are clinically relevant data fields available but a large amount of what we were after as researchers was only available in free-text. Data abstraction is still mostly a human-expert driven activity. It would be fun if that could be better automated but there is sig…

Yep. tl;dr : EHR are mostly about charts. EMR are mostly about billing. The risks of unfettered access to EMR is that conditions can be inferred and future profitabilit of patients predicted so that they can be steered to different types and qualities of care (for good or for evil). For-profit healthcare will overall optimize for profits before patient outcomes. It's as certain as gravity. FPH should be illegal.

Any organization that does not optimize value capture will be short-lived. Neither for-profit, nonprofit nor governmental business models are necessarily aligned towards improving patient outcomes over revenue. However revenue is a critical feedback loop to enforce change in the face of poor outcomes.

The Medicare agency cut off funding for heart transplants at [a nonprofit hospital] last year after the Chronicle-ProPublica investigation documented an outsized number of patient deaths and unusual surgical complications following the procedure in recent years.

https://www.houstonchronicle.com/news/investigations/article...

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#33
post #31

Pretty frequently in these scare-generation articles there's no attempt to differentiate between someone using Google Cloud to do something, and Google itself doing something. Is this just a hospital chain making a deal for Google Cloud services to support their normal IT operations, or is Google actually using this data for something? If so, what is Google using the data for?

> Google has made a deal for access to patient records from HCA, which which operates 181 hospitals and more than 2,000 healthcare sites in 21 states, _so the tech company can develop healthcare algorithms_.

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#34

Earlier quoted context omitted.

As someone who actually works at Google Health (on an unrelated project), it pains me to see folks suggesting that this type of thing could happen. I can tell you with 100% certainty that there is no way in hell that any of the health data for a project like this could possibly ever be used for advertising to patients. Someone on an eng team with access to this data (which is VERY tightly controlled) who have to writ…

project nightingale

What about it? There was definitely not any use of healthcare data for advertising.

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#35

I work as a data engineer in an academic medical research institution. Most electronic medical records (EMR) systems are really about billing. Yes, there are clinically relevant data fields available but a large amount of what we were after as researchers was only available in free-text. Data abstraction is still mostly a human-expert driven activity. It would be fun if that could be better automated but there is sig…

Pretty much all the numerical results in my chart (vitals, lab results etc) are stored numerically, and are importable / graphable in Apple Health.

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#36
post #8

Earlier quoted context omitted.

Even though this happens... Google still makes me nervous because... - Other large companies with other big agendas have violated rules on data sharing to improve performance in other areas. Amazon comes to mind with how they used markeplace data on vendors to produce their own competitive products. Rules are regularly broken. - Will Google let auditors really have enough insight to make sure data is kept in a contro…

As someone who actually works at Google Health (on an unrelated project), it pains me to see folks suggesting that this type of thing could happen. I can tell you with 100% certainty that there is no way in hell that any of the health data for a project like this could possibly ever be used for advertising to patients. Someone on an eng team with access to this data (which is VERY tightly controlled) who have to writ…

Even if we don’t worry about ads, Google doesn’t exactly have a good track record with patient privacy. Look at what you guys did with Ascension, allowing physicians to access medical records from other hospitals within the system based on similarity.

To be clear, just because something represents a common-sense workflow improvement or might be useful to the physician doesn't mean you can just go and do it. While it might be a good idea to pull similar cases (or counterfactual examples) from your vast set of patient records, it doesn't mean it's LEGAL. Medical records are not like case briefs, you can't just crack open someone's record without proper consent.

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#37

Earlier quoted context omitted.

Way to oversimplify. Some healthcare is vital (emergency intervention for heart attacks), some healthcare is grey-area quality of life (do I need glasses vs. LASIK?), and some healthcare is pure luxury (cosmetic). These are not all created equal. Additionally, removing the profit motive has its own consequences. Ever notice how the front-desk staff that serves as your docs' connection to the rest of the healthcare se…

so let's clarify a few things. All the things you described that suck (e.g. preauth, drug renewal, etc) are made to suck, on purpose, by ... drumroll ... you insurance company which, believe it or not is usually a for profit entity (USA USA USA) and therefore is incentivized to take your money and not pay any out in premium. Paying for your care to them is a loss, and what they do is try to minimize loss ratio. (reme…

Part of the problem is informational. Payors, which may include those backed by nonprofit (like BCBS) and government (CMS), are tasked with reducing un-necessary and duplicative services. This is a critical duty in bending the high cost of healthcare.

Payors identify specific procedures or treatments which represent things they want more information about before they commit to paying for it. It is difficult work on the healthcare provider side to understand what requires pre-auth and provide the payor's decisional information. There is a HL7 group that is working on the problem, the DaVinci Project:

Interoperability challenges have limited many stakeholders in the healthcare community from achieving better care at lower cost. The dual challenges of data standardization and easy information access are compromising the ability of both payers and providers to create efficient care delivery solutions and effective care management models. The goal of the Da Vinci project is to help payers and providers to positively impact clinical, quality, cost and care management outcomes.

http://www.hl7.org/about/davinci/index.cfm?ref=common

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#38

Earlier quoted context omitted.

With existing tech? No, not really. Most of those lab values are more or less worthless outside of context - which is why so much time is spent on training docs on physio, pathophys, and history taking. I know that it's frustrating to hear, but it's why every "helpful" tech solution to date has resulted in increasing doc griping and burnout.

Even simple charts for numerical variables would be helpful though. For example, evaluating growth over time is important in pediatrics. In my country, mothers receive a booklet with all the charts and pediatricians are supposed to fill it with data during consults. When I open one of these, I usually find only two or three data points. Surely EMR software could do the same thing automatically. https://www.cdc.gov/gr…

In the US, growth charts for peds are already built into every EMR I know of. But even that isn't particularly nuanced: I want different charts at different ages based on, for instance, whether the kid is bottle or breast-fed. The difference will impact an interpretation of being under- or over-weight, and using the wrong ones can create the appearance that the kid is dropping off the growth curve at around two years old (which is to say, without context you can't even interpret just the trend itself). The EMRs aren't even at the "IF feeding==bottle, display.curveA; ELSE display.curveB" level of sophistication.

Which is another way of saying "even where it looks like the numbers matter unto themselves without context, nope, you still can't meaningfully interpret them without context."

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#39
post #24

Earlier quoted context omitted.

> a large amount of what we were after as researchers was only available in free-text Is there no way to parse this? Surely the EMR systems could have a simple key-value syntax. Patient history is too complex to be expressed that way but there is a lot of other valuable data that could. Blood pressure, for example. Ideally, things like lab results would already be stored as structured data.

I mean they kinda do since lots and lots of systems support HL7 which can best be described as "JSON but so so so much worse" or "CSV with '|' kinda."

"Lots of systems" is a stretch, only a few do. In fact, "support" is a stretch. Most EMRs (Cerner, Epic etc) only support HL7 as much as it gets them their CMMI / Promoting Interoperability certifications which are not relevant for actual use.

That being said, it is JSON but much worse

Re: Google to use patient data to develop healthcare algorithms for hospital chain

#40

Earlier quoted context omitted.

project nightingale

What about it? There was definitely not any use of healthcare data for advertising.

The concern wasn't about ads, it was about patient privacy. You guys basically put together a search engine for medical encounters, allowing physicians to access medical records they shouldn't be able to: https://www.wsj.com/articles/behind-googles-project-nighting...
Post reply on HN