Show HN: Review healthcare.gov plan prices in 1 click rather than 16 | Hacker News Reader

Show HN: Review healthcare.gov plan prices in 1 click rather than 16

thehealthsherpa.com

126 pointsmwasser59 comments

http://www.thehealthsherpa.com/insurance_plans?zip_code=2759...

27596 is in multiple counties - you've got it listed only in Wake, but it's also Franklin county and IIRC Granville county too.

Is suspect 27596 is not the only ZIP in the country with this issue, and it can affect the results of this - Coventry was supposedly (as of a few days ago, anyway) not offering ACA plans in Franklin county, but was in Wake.

Thanks for pointing this out!

We currently include the plans from all counties in a particular zipcode rather than limiting to a specific county, but we should probably ask a user to specify their county if there's ambiguity. I'll get on adding this.

As for the data itself, it's current as of two weeks ago which is already slightly stale. This said, the data will only be as updated as the Healthcare.gov data from https://data.healthcare.gov/dataset/QHP-Individual-Medical-L...

NC may be somewhat specific, but per reports a few weeks ago, only BCBS was offering ACA-compliant plans in all 100+ counties - Coventry was the only other option, and they were covering 36 counties. They may have changed that recently, but I'm not 100% sure on that.

EDIT: and thanks for your project - nice use of the data to help people get fast/easy access!

It's worse than that... There is at least one ZIP code that crosses STATE lines!

That blows my mind. I would have expected at least zip codes to be maintained within a state!

No geographical coding system is sane. None of them. Map projections, area codes, zip codes, time zones, nothing. They're all fucked up bonkers.

I hope you got paid $150mm to create that.

Why do I always see people using lowercase "mm" to abbreviate million? That doesn't make any sense to me.

m is 1,000. mm is m * m or 1000 * 1000, which is 1,000,000

MM, sure. mm, however, is millimeters.

its done in finance sometimes and the thought behind it is m = 1000, so multiply 1000 by 1000 and you get 1,000,000.

Personally I hate that notation and use capital M whenever given the opportunity. I feel like it used to show superiority or present another barrier to people outside of finance looking in.

a millimeter is = to 1000 micrometers.

150 mega-megadollars??

Roman number M = 1 thousand

1 thousand 1 thousands = 1 million

Exactly, capital M

But, then that would be 150 * 2000 = 300,000.

Exactly. 1 million in Roman numerals is M with a bar over the top of it.

Yes. If we're talking Roman numerals, MM would be 2000, just like XX is 20 -- not 100. This notation makes no sense to me either.

I've always seen k used to represent one thousand in dollar amounts, so I've never interpreted M as being a Roman numeral but simply as standing for "million."

Disclaimer: I've been in the US my whole life.

We don't use Roman numerals for thousands, why should we do so for millions? The right way is obviously using metric prefixes, so 1k$, 1M$ and 1G$.

And don't forget that 'cents' is short for 'centidollars'.

Is there a special reason the 'k' is lowercase?

Well, M for mega is capitalized since m is already used by milli. And all of the SI prefixes larger than mega are also capital, I guess to denote that they're large...

All of the SI prefixes k and below are lowercase.

Doubling to indicate a plural initialism, like "pp 3-7" for "pages 3-7" or "EE. UU." for "Estados Unidos".

you know how a few years back it became popular to say "kk" in online games instead of "k" or "okay"

I suspect people use MM/mm for the same reason, some guy just made a typo one day and it caught on.

kk=okay, cool

Mille is Latin for one thousand. It's still heavily used in French and perhaps as much in Italian.

I have literally only ever seen this on HN.

(from the UK where m = million)

looks like a hybrid of USD and millimeters

Without information on the deductible this isn't as useful as it looks. Please add it!

The price of plans varies dramatically by the deductible, and for most healthy people the deductible matters WAY more than the nominal plan level. (Since for ordinary services the different plan levels pay an extra $10-$50, but for a catastrophe the only number that matters is the deductible.)

I completely agree!

We'd love to be able to find this data per-plan, but its not currently easily discoverable for the states without exchanges.

We're hoping to find deductible data in time for these plans as well as more data that would help people differentiate plans such as number of providers in-network nearby.

How hard would it be to do it manually? Maybe with Amazon Turk?

Wow I am really in a tough spot. For me to cover my family and my spouse will be $630 a month for bronze i.e 60% coverage.

for zip code 03570

I recently came across http://samaritanministries.org/intro/ and it's much more appealing.

If that's a tough spot, you probably won't be paying the full cost of any of the plans you saw. For a family of four, subsidies don't phase out until something like $90k/year income. The real healthcare.gov site is up right now. The registration process only takes 10-15 minutes to get some real prices.

I just checked.. the deductibles are high plus partial coverage and the subsidies are not as much as I had hoped.. still paying around $480.

Nice Job! I'd suggest adding a page with information on what each plan level (Gold, Silver, Bronze) covers at minimum.

Very nice! I was thinking of creating something similar as I'm sure I'm not the only one who just quickly wants to see how much option X would cost. Where did you guys get the data from? I was considering screen scraping the main site but I gave up after I couldn't get it to load at all.

Would be nice if something like this was on the official site but my guess as to why it's not (besides the general mismanagement of the site creation...) is that by default it shows higher prices for plans than people would be eligible for (ex: no income based subsidies). That would make them "look bad" and they'd like to show lower overall costs.

Bug report too: When I entered a zip code the first two pages of the results show up blank. Clicking page 3 and onward show the data though. If there's only one page of data nothing shows up at all (no paging links either).

Data comes from https://data.healthcare.gov/dataset/QHP-Individual-Medical-L...

Thanks for the bug report- In theory I just pushed a fix for it. Could you confirm it's working for you now? If it's still there, could you let me know what zipcode you are using?

First page issue is fixed now.

I was about to report one more about funky displaying of plan names but looks like the source data itself is like that (in the set you linked too). Search for bronze plans in 07302 and you'll see: AmeriHealth NJ Tier 1 Advantage_Ã

Same here. But more of a parody site with some 'bugs' and 'outages'

Here is their official Developer API

To send a request: https://data.healthcare.gov/resource/qhp-iml-dev.json?state=...

Good stuff. Would you open source this ? The prices are a great starting point. A nice addition will be to easily see details like deductibles, co-pays, out of pocket limits, exclusions etc, the price just by itself won't be a perfect measure.

I wrote the same thing, with a much crappier user interface, last week. Mine includes Idaho and New Mexico as well: http://healthcare.brianjcohen.com

My notes on that page explain why this is really not a great dataset. These aren't really price quotes, they're averages they've generated for various age ranges. And since deductibles, coinsurance rates, and copays aren't included, there's really no way to compare them on a level playing field (although this is a problem that healthcare.gov has as well).

Looks like the actual healthcare.gov website is not that bad anymore when it comes to just checking on prices in your area. Just use the following URL and replace the state and county with your choice. My example uses NJ/Warren county.

https://www.healthcare.gov/find-premium-estimates/#results/&...

This is what we saw as well. The main idea was that it still takes a number of clicks to get there at all. We wanted to provide an interface that significantly lowered the time to seeing the data and then provided the other details.

Great timing! It's unfortunate that 2/3 states that I tried have "their own healthcare exchange." Kind of defeats the point of having centralized healthcare and also a complete pain the ass for the consumer.

On a bright side, I bet a ton of people got their pockets full by working on all of these exchanges.

Yeah -- the story behind it is intriguing. Also, the states that have their own exchanges use central services provided by the federal gov for subsidy eligibility.

I do think we will see states release data back to the federal gov for ease of use in time. NM and ID already has done this and other states seem to send their raw data to individuals when asked -- suggesting the data can be centralized (e.g. when we reached out to CA for data, they quickly provided their data to us).

This is very interesting to me, I live in KY and we have our own exchange. Do to not having all my data (my fault not theirs) on hand the process of submitting an application has spanned a few days and their interface is really bad. I'd like to play with raw data and throw something together to allow for quick and easy searching. Who did you contact in CA for the data? I'd like to find the KY equivalent and reach out. Thanks and great job on the site!

the actual healthcare.gov (if you can get in) actually does a decent job of letting you compare plans in detail... there are far more factors to consider than just monthly deductible. various copays, out of network coverage, whether dental and vision are included, etc. this site does none of that

Do you have to sign to see this? When I just browse the site all I can see is the name of the plan, the insurance provider, and the estimated cost.

Not robust yet. Try: http://www.thehealthsherpa.com/insurance_plans?zip_code=9404... and see an error.

Washington DC zip codes don't work (e.g., 20001, 20002, 20011, etc.).

Washington DC is running its own exchange. Apps like this that use data.healthcare.gov as their backing will only work for states that are on the Federal exchange.

FYI, tried to enter zipcode 85502 and received a "We're sorry, something went wrong" message. Worked fine for other zipcodes I was interested in. Thanks for creating this!

I got the same for 94906. Made me think it was a joke site actually. ;-)

When I entered my zip code, the site stated that my state, Arkansas, is running its own exchange. In fact, Arkansas defaulted to the federal exchange.

Thanks for pointing this out, we've fixed the bug and you should be able to find Arkansas plans now.

Any plans to list the link to the summary of benefits as well? That would be super nice.

Was really excited about this, but no functionality for New York. :/

I was waiting for someone to do this...

22314 doesn't work at all.

Reply on news.ycombinator.com