Health insurers just published close to a trillion hospital prices

I'm the author. A question I have is: how did so many prices ever get negotiated in the first place? What kind of systems are in place to do this kind of micro-negotiation?

I&#x27;m author of Hacking Healthcare for O&#x27;Reilly, 20 year health system executive, blah, blah.<p>It&#x27;s very easy for people to forget the scale of the US &quot;health system&quot;, we are talking 1&#x2F;5, maybe more, of the entire US economy. If US healthcare spending were a country, it would have the third largest GDP in the world. Accidents of history and the massive federal beauracracy created the crazy monster of ICD&#x2F;CPT codes that results in the very clumsy way of pricing healthcare services that results in this massive matrix of data.<p>As pointed out elsewhere there is a tremendous amount of cost distribution that goes into the code matrix and this plays a large role in negotiations with health insurers as well. Ground is given in one set of procedures and lost in others.<p>This is a big step in shining light into areas that need it to improve the system overall.

I work in data at <a href="https:&#x2F;&#x2F;www.carrumhealth.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.carrumhealth.com&#x2F;</a>, and I&#x27;ve been parsing this data for weeks. The transparency prices allow us to meaningfully negotiate with providers, and make tangible, incremental progress toward cheaper health care. Providers and existing insurance carriers leverage information asymmetry to control the market otherwise.<p>For context, we bundle the 100&#x27;s of itemized costs into a single, static bill per surgery type. In doing so, we&#x27;ve built a custom virtual-network with the most efficient surgeons. These surgeons are able to meet the volume and quality requirements to allow for lower margins. We&#x27;re able to get negotiated rates that are 10-40% cheaper than traditional insurance contracts when we have data that we trust.<p>Unfortunately, this data alone isn&#x27;t enough to properly determine prices because organizations will spread costs across procedure and billing codes that often occur in aggregate groups. For example, in a joint replacement surgery, some organizations may dump the cost into the billing for the implant itself, while others may put it under the procedure code. You have to gather billing data en masse to see which charges occur together, then combine this pricing data to determine what costs will actually look like for someone experiencing a procedure.<p>It&#x27;s a nightmare!

Someone with the right connections should call up Google Cloud and ask them to ingest the data into BigQuery as an example dataset like the NY taxi trips. It would be a great way for them to show off the capabilities of the engine and helpful for everyone wanting to do analysis on it.<p><a href="https:&#x2F;&#x2F;cloud.google.com&#x2F;bigquery&#x2F;public-data" rel="nofollow">https:&#x2F;&#x2F;cloud.google.com&#x2F;bigquery&#x2F;public-data</a>

To save anyone else similarly curious the trouble, here&#x27;s a sample record from the Humana data set:<p><pre><code> {&#x27;REPORTING_ENTITY_NAME&#x27;: &#x27;Humana Inc&#x27;, &#x27;REPORTING_ENTITY_TYPE&#x27;: &#x27;Health Insurance Issuer&#x27;, &#x27;LAST_UPDATED_ON&#x27;: &#x27;2022-08-24&#x27;, &#x27;VERSION&#x27;: &#x27;1.0.0&#x27;, &#x27;NPI&#x27;: &#x27;1629053517,1659354272&#x27;, &#x27;TIN&#x27;: &#x27;593279318&#x27;, &#x27;TYPE&#x27;: &#x27;ein&#x27;, &#x27;NEGOTIATION_ARRANGEMENT&#x27;: &#x27;ffs&#x27;, &#x27;NAME&#x27;: &#x27;Nasal Prosthesis Replacement See Also Code 21087&#x27;, &#x27;BILLING_CODE_TYPE&#x27;: &#x27;CDT&#x27;, &#x27;BILLING_CODE_TYPE_VERSION&#x27;: &#x27;2022&#x27;, &#x27;BILLING_CODE&#x27;: &#x27;D5926&#x27;, &#x27;DESCRIPTION&#x27;: &#x27;Nasal Prosthesis Replacement See Also Code 21087&#x27;, &#x27;NEGOTIATED_TYPE&#x27;: &#x27;negotiated&#x27;, &#x27;NEGOTIATED_RATE&#x27;: &#x27;906.98&#x27;, &#x27;EXPIRATION_DATE&#x27;: &#x27;9999-12-31&#x27;, &#x27;SERVICE_CODE&#x27;: &#x27;&#x27;, &#x27;BILLING_CLASS&#x27;: &#x27;professional&#x27;, &#x27;BILLING_CODE_MODIFIER&#x27;: &#x27;&#x27;, &#x27;ADDITIONAL_INFO&#x27;: &#x27;&#x27;, &#x27;BUNDLED_BILLING_CODE_TYPE&#x27;: &#x27;&#x27;, &#x27;BUNDLED_BILLING_CODE_VERSION&#x27;: &#x27;&#x27;, &#x27;BUNDLED_BILLING_CODE&#x27;: &#x27;&#x27;, &#x27;BUNDLED_DESCRIPTION&#x27;: &#x27;&#x27;} </code></pre> I think I agree about the negotiation arrangement

Anyone else get the feeling this is malicious compliance on behalf of the insurance companies?<p>&quot;Oh, they&#x27;re going to force us to publish our prices are they? Well we&#x27;ll publish so much data it&#x27;ll take a herculean effort to make it readable to anyone that doesn&#x27;t work in data engineering&quot;

I wonder what percentage of work in the US healthcare system is completely unnecessary from a general perspective but made necessary deliberately to justify the unethical system that allows millions to die unnecessarily.

Here is a golden opportunity for the info&#x2F;data visualization community to show how their tools can handle big datasets to make them comprehensible to the public.

Seems like every week there&#x27;s a new massive scale DB project or company getting announced on HN.<p>If they&#x27;re looking for projects that create public value and demonstrate the power of their products at scale, digitizing this and making it searchable may be a good marketing project that&#x27;s appealing to certain kinds of customers.

I&#x27;m the CEO and cofounder of a health insurer that published its pricing data. See it on GitHub at <a href="https:&#x2F;&#x2F;github.com&#x2F;evryhealth&#x2F;price-transparency" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;evryhealth&#x2F;price-transparency</a> To add some fuel to this discussion, US healthcare has grown 3x as a percentage of GDP in 60 years from approx. 5% to more than 15% [1]. It has done so at the expense of the rest of the economy and communities.<p>Pricing transparency is only one piece of the puzzle. It is a tremendously antiquated industry. Fax is still state of the art -- welcome to the 1980s!<p>[1] CMS. <a href="https:&#x2F;&#x2F;www.cms.gov&#x2F;Research-Statistics-Data-and-Systems&#x2F;Statistics-Trends-and-Reports&#x2F;NationalHealthExpendData&#x2F;NationalHealthAccountsHistorical" rel="nofollow">https:&#x2F;&#x2F;www.cms.gov&#x2F;Research-Statistics-Data-and-Systems&#x2F;Sta...</a>

India is considered a &quot;cheap and poor&quot; country in many ways. It&#x27;s GDP is 1&#x2F;10th that of United States. Even then, India has free health care, and in many states (out of the 30+ states), Govt run healthcare is the best. Woman give birth to babies and walk out paying almost nothing, men break their bones and walk out paying nothing after a month of staying in hospitals and such. Insurance is never heard of or forced to, unless you own a automobile and pays insurance to drive it on roads. No monopoly of any kind.<p>And there are arguments against all these points, I concur. I just said it for the American folks to know.

This &quot;data dump&quot; is step 2 of 4 to deliver &quot;on President Trump’s executive order on Improving Price and Quality Transparency in American Healthcare to Put Patients First.&quot;(1)<p>By January 1, 2023, plans and issuers must make price comparison information available with respect to an initial list of 500 identified items and services. By January 1, 2024, plans and issuers must make price comparison information available with respect to all covered items and services. This information must be made available through an internet-based self-service tool and in paper form, upon request. Typically, consumers receive an Explanation of Benefits after receiving care, which details the prices charged by the provider, the plan’s contracted or negotiated rates, consumer cost-sharing obligations, and other information. Consumers will have access to this type of information <i>before receiving care</i> and can use it to compare prices and better estimate potential out-of-pocket costs.(2)<p>(1)<a href="https:&#x2F;&#x2F;www.cms.gov&#x2F;newsroom&#x2F;fact-sheets&#x2F;transparency-coverage-final-rule-fact-sheet-cms-9915-f" rel="nofollow">https:&#x2F;&#x2F;www.cms.gov&#x2F;newsroom&#x2F;fact-sheets&#x2F;transparency-covera...</a> (2)<a href="https:&#x2F;&#x2F;www.cms.gov&#x2F;healthplan-price-transparency&#x2F;consumers" rel="nofollow">https:&#x2F;&#x2F;www.cms.gov&#x2F;healthplan-price-transparency&#x2F;consumers</a>

6000 hospitals x say around 10000 priced items, on average, per hospital x 100 different negotiated pricing formulas = 6 trillion unique prices.<p>Of course many hospitals negotiate en bloc as part of a healthcare network, and there probably are more than 100 different organizations that negotiated unique healthcare pricing but the ballpark number seems to make sense.

Seems crazy that SEC rules require structured data but these health accounting rules did not. I guess health care has better lobbyists.

This data is already available cleaned here.<p><a href="https:&#x2F;&#x2F;turquoise.health&#x2F;" rel="nofollow">https:&#x2F;&#x2F;turquoise.health&#x2F;</a>

Unrelated: “dolt” vs “doIt” is one of my standard font competence tests

I would <i>love</i> to hear from an insider on what the difference is between these published files and the internal databases - I&#x27;m sure the difference between the two is striking. Malicious compliance, indeed.

Where is the raw 50-100TB of compressed data available for download?<p>Is it fully public or does it require registration to access?

This is the real solution to healthcare costs and quality. Instead of the government handing hospitals a blank check for low quality services in a single payer scenario, we allow price transparency, competition and the free market to drive down costs and increase quality. Hate him if you want, but this was a huge accomplishment by Trump that has just started to take effect.<p>If you think the government can bring down the cost of anything please see education and NASA for great examples.

Price transparency is an essential part of a healthy market. Or so says the theory. I hope they release it. I wonder what I can do to help? I’m capable in db design, sql, etc being a data engineer by day perhaps I can help this effort.

I started a healthcare cost transparency platform in 2014 [1] and struggled to find costs by CPT code. The best we had was medicare reimbursement rates, which were hardly a good litmus for out of pocket costs. It&#x27;s nice that after 8 years, things are finally shifting. Just shows how important timing is in entrepreneurship.<p>[1] <a href="https:&#x2F;&#x2F;ethansteininger.com&#x2F;portfolio&#x2F;comparedcare" rel="nofollow">https:&#x2F;&#x2F;ethansteininger.com&#x2F;portfolio&#x2F;comparedcare</a>

Say someone ingests this data and clean it up make it usable, who&#x27;s the customer for that service? What would they want to know from it?

Maybe good, maybe bad. I suspect the good will be lower prices as we resolve major conflicting prices for the same service. Maybe bad as we find that nurse Jackie is spending too much time taking care of your sick husband and that needs to cut back as the prices the hospital is negotiating drops. The service will become more standardized and robotic.

FYI, if you want to view the data in a reasonable format replace ~ with comma, but first make sure to replace any commas with something like a semicolon. I guess csv delimiters can be anything and they chose ~! Good thing there are no insurance company names with a <i>~</i> in it.

From my understanding though, these &#x27;prices&#x27; are outdated nearly as quickly as they are published. I.e. sure you have a set, but everything is dynamic and changing. Seems like it would require a ML approach to &#x27;understand&#x27; such a dataset going forward.

What happens if we mandate that hospitals cannot charge patients more than the lowest negotiated price they have for any procedure? That is to say, if you have offered to perform a procedure for $X, you can not charge anyone more than $X for the same procedure.

A common complaint here is that the single payer systems &#x2F; public run systems have very long wait times compared to private US systems (this is true where I live as well. A simple MRI is a five month wait and that requires traveling to a different city).

Is it possible that they intentionally bloated the data to make it difficult to use effectively?

This seems very much like the episode of Veep&#x27;Boxes of lies&#x27; where they try to hide their nefarious deeds alongside the real day to day inter-workings of the vice presidents office but inundating the public with data.

A bit off tangent, is there something that can ingest unstructured data and output something meaningful? Something like GPT-3 but huge amount of data.

Aren&#x27;t the costs of the US healthcare system largely driven by the cost of the healthcare workforce - basically, not enough trained personnel leading to the absence of competition, leading to doctors getting paid much more than they should? Of course it&#x27;s going to be more complicated, with doctors&#x27; insurance issues and the price of meds into the mix, but still, how far does this factor goes into explaining the extraordinary cost of getting a basic treatment (let alone surgery) in the US?

this disclosure was spurned by recent federal legislation that required it. Im a full cynic on the disclosure, so be warned.<p>- these prices, as negotiated between insurers and providers, were already well known inside the industry. so much so that many procedures could be declined coverage well in advance of a customer ever needing one. this insider knowledge formed the core of many earnings reports for insurers and hospitals alike,<p>- Disclosure is meaningless if the customer has no alternative. most health services that bankrupt are emergency medicine, and as such youll pay anything to save your own life. thrusting a stack of price sheets at a faceless national healthcare monopoly and demanding a fair price is a laughable if not sad idea. Healthcare is not something capitalism is equipped to competently support.<p>- hospitals have zero incentive to work with you on any price for any service, and no federal state or local law will compel them to do so by virtue of a combination of bureaucratic deadlock and regulatory capture. is it, for them, more profitable to sell your arbitrary debt to a credit collection agency? shove you into a debt counseling service they get kickbacks from? work a long and grueling payment plan through their own financial services division to bolster quarterly profit long-term in a recession? or just ignore your pleas entirely? what they charge is not up for debate by <i>you.</i>

I wonder what their egress bill will be if a large number of people are interesting in parsing this data.

This is unacceptable. Perhaps it meets the letter of the law, but certainly not the spirit. This doesn&#x27;t begin to make healthcare anything like a &quot;free market&quot;. It&#x27;s dishonest in the same way a restaurant might be forced to release a nutritional analysis of their food, and so they release a molecule-by-molecule description. Despicable.

I wonder if our entire health care problems in the US stem from the medical cartel itself. You have medical schools limiting seats, you have physicians and head physicians and medical groups lobbying for changes, and you have inflated doctor salaries…<p>A sane approach might be to let more people graduate medical school and increase the number of applicants…

I don&#x27;t think he meant largesse did he? He seems to mean &quot;large size.&quot;

A worthy cause. Thank you so much for your work. Please keep it up.

Anyone knows if this dump contains drug insurance coverage?

I believe this was a Trump policy

This is clearly obfuscation. There is no way they have &quot;negotiated&quot; half a trillion of anything.

De-reg. De-reg. De-reg.

Could we rename the title to: &quot;Health insurers just published close to a trillion hospital prices&quot;<p>This post is lot more interesting and important than the current short title would suggest.

USA Healthcare is 19.5% of GDP. Canada + Australia + UK + (a few similar countries) average 10.5% of GDP.<p>The USA healthcare price gouges via run away prices. Healthcare corporations corrupting congress is the fuel that forces this on people (and breaks a free market).

My current assumption is that private healthcare&#x2F;insurance is to blame, because countries without that or with less generally have better outcomes at less cost. Looking for evidence to the contrary

Nothing enrages me like US healthcare.<p>Physicians have a monopoly on treatment and restrict supply through the ACGME.<p>Hospitals and physicians have spent 1Billion on lobbying.<p>I could go on, I want a science based healthcare as an alternative to our authority based system.<p>It would be one thing if our system healed everyone, but the number of mistakes is enough to make me lose faith completely.

I know that most people don&#x27;t like to think twice when it&#x27;s about their own health but as a society we have to ask if it&#x27;s ethical to heavily indebt or tax the young working age population to extend the lives of a few old peoples who often never regain their full health.<p>In the actual system good lifetime is traded for bad one.

The issue is that states are largely arbitrary historical accidents. You&#x27;ll need this for a variety of things. &quot;Vice&quot; states are located next to more conservative states. Consider the states of New Hampshire, Delaware, Pennsylvania, and South Carolina.<p>In the case of healthcare, states have a tendency to shift costs elsewhere. As in, send the sick person to New York City.