Show HN: HealthPlanExplorer.com: Search & parse insurance plan MRF Files | Hacker News Reader

Show HN: HealthPlanExplorer.com: Search & parse insurance plan MRF Files

healthplanexplorer.com

22 pointskunle19 comments

Been exploring the Transparency in Coverage Rule (https://www.cms.gov/healthplan-price-transparency) that forces every health insurer to publish their contract rates with every provider they have in network in a machine readable format (json files)

These published files are massive (about a petabyte of data a month with individual files up to 20GB) and a PITA to dig through. We built HealthPlanExplorer.com to make it easy to search the files. You can:

- choose any in network file (as many as you want) - search by National Provider Identifier (NPI) - search by CPT Code

And we'll generate a CSV with the results. If the items you search for aren't found, we'll give you the full export. Next up, we'll make it queryable so you can do different cuts much more quickly.

Open articleView on HN

It's great to see innovation in this space! Spent some time working on similar solutions both for patients and employers so I understand how big of an undertaking it is.

Patients that I've interviewed generally want a completely abstracted process - send or upload a picture of their bill and their insurance, and then hand it off. This would be similar to the cushion.ai experience.

Employers on the other hand want to ensure that they aren't being overcharged by their current stack. They are always looking for solutions for reducing costs and better understanding the drivers of those costs, which tools like springbuk.com address.

I have less familiarity with the provider side, but the cost of maintaining an insurance billing program has always seemed manual and inefficient. At the SMB scale, I suspect something like Square or Stripe would have a great shot at solving these problems, but I can see data and billing teams at health systems using this to help with negotiations on specific cases and baseline pricing for different CPT codes.

API endpoints that provide clean, abstracted negotiated rates data could be a great infrastructural product for both traditional providers and digital health companies.

Few things

> API endpoints that provide clean, abstracted negotiated rates data could be a great infrastructural product for both traditional providers and digital health companies.

Cascade Health (who I've been working with) have this API on the roadmap

>Patients that I've interviewed generally want a completely abstracted process - send or upload a picture of their bill and their insurance, and then hand it off. This would be similar to the cushion.ai experience.

Mixed on this. I think a way to select your insurance and know what you'd pay out of pocket, and to compare your out of pocket costs between multiple providers, is SUPER valuable. By the time you have a bill, you already owe money. Before the visit, you actually have a shot at making a decision that costs you less.

> Employers on the other hand want to ensure that they aren't being overcharged by their current stack. They are always looking for solutions for reducing costs and better understanding the drivers of those costs, which tools like springbuk.com address.

Cascade is also doing this.

> I have less familiarity with the provider side, but the cost of maintaining an insurance billing program has always seemed manual and inefficient.

Have a writeup coming about this one actually. I believe there's room for "Contract Streaming" which allows you to use the published data to a) ingest your contracted rates with every payor and b)know in advance what the out of network rates are for any payor, so you can tell a patient in advance exactly what their visit will cost them before they walk out the door. (There's also a bunch of other benefits around forecasting and stuff, but this is the start).

Wrote Hacking Healthcare for O'Reilly, long time health/tech exec, yada, yada... I'm really glad this pricing transparency information has become available but unfortunately so far I don't think it has provided a lot tangible value to consumers/patients. When you take into account the following:

- acute care (where there is not a lot of oppourtunity for discrection based on price) - proximity - insurance networks - complexity of care itself

What I have seen for most patients most of the time it is pretty difficult to put this information into practice. Now this information is having a much bigger effect on the relationships between health plans, providers and health systems which is positive.

I'd be really open to discussing with anyone ideas about how this information can be combined with other resources people may not know exist and how it could be possible to make it more useful for patients. Feel free to reach out to me:

du@50km.com

Thanks for this color. I have a ton of thoughts about this. . . A few:

1. The best way I think we can enable patients/consumers to price discriminate is about pushing pricing into the moment when they're making a decision. An example of how to do this is: https://carbonhealth.com/flu/flu-vaccination-centers . .. we created a map, that indexes well, for where to find the lowest cost flu shot (this was done before the price transparency data was available, but I fully intend to a) incorporate the price transparency data for our existing maps and b) create new health maps for basically every service that tells you the relative rate for your plan

2. There's going to be real demand for an API that returns pricing data, for marketplaces that aggregate patients (like Solv, Zocdoc, Sesamecare, healthgrades etc) because pricing is probably the most in-demand decision driver that no one has.

3. Ultimately, I expect Google and Apple to incorporate these prices into maps. Then you'll see it have a real impact on consumer behavior, because most healthcare journeys (whether we like it or not) start with a search.

While I think you should make pricing available for everything (bc after all, there is a price for everything) I think it's most effective to start with diagnostics (mammograms, xrays, STD tests etc) because there's pretty low variance in outcomes (ie it kinda doesn't matter where you get them, you're getting the same thing). When it comes to complex therapeutic searches (eg oncology, chronic care) while pricing is one dimension, patients probably want to see pricing along with a proxy for quality (eg that specialists quality metrics, number of times the MD has done that procedure, etc), bc the healing process for those are not linearly related to price, and the cost of a visit is just a fraction of going down the wrong therapeutic pathway, for both the patient and the system.

There are a few relatively commodity items you have mentioned, flu shots for one, mammogram possibly another, but for > 65% of patients those are already covered at extremely low/no cost. Yes, there is room to help the remaining 35%.

As you get into xray, clinical laboratory, other imaging things become very complicated where the order from the provider is commonly bound to a particular facility. Trying to get a provider to write an MRI order to a facility they have not worked with before is often a multi-day process fraught with error due to complexity in the order itself. Most patients cannot get their providers office in active communication long enough to even try it. For procedures there is an exponential explosion in complexity. Electives are really, really complex. A birth or a hip replacement is not the same thing across facilities so how do you compare code level pricing?

With drugs GoodRX has produced excellent results for patients. I think that works so well because overwhelmingly drugs are commodity goods.

I do see an oppourtunity because health systems spend a tremendous amount on patient acquisition currently but I am having a difficult time seeing the dots connect into pricing decisions by patients.

So your point on items covered by insurers is right with the caveat that way more patients are now on high deductible plans, so your out of pocket cost still gets impacted by the contract rates even if the provider is in network.

Agree on electives - there’s something to do there price wise but it’s not the most straightforward.

Re: pricing in general, it def matters in the decision but how you present it makes a huge difference. For uninsured people our flu map has actually been super critical (people are more likely to choose an option with a price at all and marginally more likely to choose a lower priced option (the price spreads are lower bc it’s a simple procedure but there are some places where you can pay over $100 for a flu shot - no one chooses those once they know)

The nit here is you have to give a person the ability to a) select their insurance, and b) show them the full out of pocket cost (copay + deductible) and then it makes a difference, which favors a logged in experience (like solv or Zocdoc) over a logged out experience. But I’m convinced that the reason it’s hard to see today is not because people don’t care about price. It’s because they don’t get to price discriminate at all.

There is obviously a huge business opportunity in being able to present this data to consumers / businesses / brokers etc...Interesting to think how something like this could turn into the mint.com or amazon of healthcare. More likely business model might be presenting comparable plan pricing to SMBs and then acting as their broker.

Probably not crazy to bet that The Biggest Company In the World (https://a16z.com/2022/11/11/the-biggest-company-in-the-world...) might start off by solving this kind of problem.

Anyways great early progress and I am excited to see where you guys end up.

Thank you. Lots of directions this could go - right now focusing on making it easy to get access to the data. Later we’ll figure out useful ways to interact with it.

I think you should look at https://turquoise.health/. They are aggregating cash and insurance pricing.

Thanks - seen them and it’s similar but they’re building something much more enterprise scale.

Cascadehealth.ai and serifhealth.com are as well.

Keen to learn more use cases people are interested in. The very next release will be exporting a DB you can query directly. Open to more/other ideas.

Raw data does not tell a story. I would present some of the insightful data using data science and graphs.

Then you can offer to provide the raw underlying data. It would be great if one can browse the aggregate numbers and compare them to say other zip codes / other factors.

Good thought - were planning to do some aggregation but there’s actually literally too much data to be holistic, so we’re going to do it on a few large plans and go from there

As a "regular" consumer, I couldn't figure out how to use this because I don't deal with the kinds of files it was asking for. At the very least it should provide some links to example files you can run through it.

In any case, I'd like to see something where you select a procedure and then get a way to easily drill down to see pricing (both in and out of network) at different locations, probably within a given radius. This assumes, of course, that the system knows my health plan to begin with, which would be a whole other step needed.

Oh hell yeah this is coming. I built healthplanexplorer because the files were too unwieldy, but the ultimate consumer format for this is a map like https://carbonhealth.com/flu/flu-vaccination-centers

With 1. A way to choose your plan 2. A way to see all providers in network with that plan 3. A way to see contract rates 4. A way to estimate your out of pocket costs (copay plus deductible) if you go for a specific service

A good insurance agent should do most of 1-4, but your service should make their job easier. So I see this as more for them than for a consumer. And also because after a plan is selected, copays and deductibles are essentially fixed so consumer would have no need to price shop.

What do you mean insurance agent? The process of picking health insurance in the US doesn't utilize agents.

> And also because after a plan is selected, copays and deductibles are essentially fixed so consumer would have no need to price shop.

This assumes that the service you're picking costs more than the deductible. If the service you're shopping for costs less than your deductible, you're paying for all of it up front...

I am in US and have an idependent insurance agent that offers healthcare plans for every insurance company licensed in our state. They compared my plan with all the others, taking into account the types of service I use, what is covered when travelling, etc. Anytime I need a specialist I call the agent for a list of in-network doctors.

Wow, never knew this was a thing. Do they have a website/can you share?

Are you insured through your employer or via an exchange?

Reply on news.ycombinator.com