Hacker News new | ask | show | jobs
Show HN: Health Scout – An app I wrote for a mother traveling with a sick child (itunes.apple.com)
48 points by rtrivedi 4155 days ago
10 comments

While on vacation, I spoke to a mother who often had to be aware of emergency rooms around her because her child had health problems. I built this simple app for her. It's my first app and I'm looking for suggestions on how to make it better. Right now it only works in the USA but I'm hoping to expand to Canada very soon.
Not sure if it'll be much help, but you might be able to find and scrape data from here: http://data.gov.uk/data/search to do something similar for the UK, I can't seem to find it though.
Except general people should be using pharmacists; in-hours GPs or out of hours doctors and walk-in-centres; and minor injuries units. A&E should be reserved for emergency care. There's a very small percentage of people who need an A&E department and this app would be great for them, but please don't send people who are not in an emergency to A&E.

http://www.nhs.uk/NHSEngland/AboutNHSservices/Pages/NHSservi...

http://www.choosewellglos.nhs.uk/

Thanks for the links. I'd like to help build a world where people have access to useful information as easily as possible. Of course, with ease of access to information people will need to become more responsible. I'm hopeful we can get there :)
Cool! I'll take a look :)
I'm an ER physician, this is a great idea for an app. It would be good to discuss further, I have sent you a connect request on linked in, thanks!
I'm developing something in the ER space, also in LA, will reach out via linkedin
Great!
Cool! Let's talk :)
I think this a really nifty idea. Would love to know where the data comes from - are you aggregating it yourself?

Also, does it include a notification centre widget?

The registered emergency rooms are catalogued on the medicare website (medicare.gov). The government has a list of all the emergency rooms in the USA. They also do a survey of each hospital periodically to keep there data up to date.

Unfortunately, these updates aren't in real time. I'm hoping I can provide people live data through my app through a feedback/reporting mechanism. It took a fair bit of time to extract all the data available and I had to do some aggregation myself.

The app doesn't yet include a notification centre widget but that's a great idea! I'm putting that on the list of potential feature adds :). Thank you!

  > to keep there data up to date
Should be "their data". You should also fix this in the app description: should be "about what their options are", not "about what there options are".
Thanks. Fixed in the app description.
Where are you keeping the data that you've compiled? Are you using Parse?
I've got a medically sensitive daughter and this is a cool idea. Where does the data come from? Are you protected in situations where someone might drive to a now-closed ER?
The data comes from http://medicare.gov/hospitalcompare/search.html

It has a pretty good amount of useful data if you're interested in searching around :).

To answer your second question: I'm not. For the V1 of the app, I'm relying on the government data being valid. However, I'm currently building an in-app feedback form that will allow users to (a) report real-time wait times if they are in the hospital, and (b) report closed ERs / incorrect addresses etc.

rtrivedi - do you have an email I can reach you on? I work for the company that helps Medicare.gov (and a lot of other govt agencies) publish their data online in machine readable formats and I'd love to get feedback from you and understand how you're using the data and potentially make the data publishers aware of this use case as well. You can reach me at mark.silverberg@socrata.com if you prefer not to post your email publicly.

Great job!

that's a very insightful comment on potential liability. it could indeed prove to be a problem in the event that the app misguides a user to a closed ER with a tragic outcome, especially in a litigious society.
I totally agree. One thing I made sure to do was make it very clear in the app (via a disclaimer) that the data is not real time and the app should not be used in emergency situations. I even tell them to call 911 whenever in doubt.

I think Health Scout is very useful but definitely shouldn't be used in life and death situations. Maybe I should make the disclaimer even more prominent...

One thing you could do AND SHOULD DO now is post your disclaimer in your app's description. LIKE RIGHT NOW!! Don't even read the rest of what I've posted. Do not pass go. Just go append your disclaimer to your App Page.

Okay, now that you've done that, good. Second, you need to put up a disclaimer screen or modal that happens on launch or viewDidLoad that requires user to agree to your terms of service/disclaimer/privacy policy in order to use the app. If they don't agree, they can't use the app. Period.

As it is now, I can access your app without seeing any messages or disclaimers on the accuracy of the information.

People can be litigious jackasses and may want to find somebody to blame for their grief or problems. With the false perception that every app makes millions of dollars, make sure you are not a target for lawsuits by at least doing the above steps ASAP.

Other than that, great job on the app!

Thanks! Will do asap and push out a new build.
Remember that you do not need to push a new build for your description changes to get updated. Those do not require approval and will go live soon after you hit save in iTunesConnect.
Your app to find Emergency Rooms nearby shouldn't be used in emergencies? I'm not convinced a disclaimer like that it's going to cover you.
I imagine the use would be, for the caregiver of a medically sensitive individual, to familiarize oneself with a new local area on arrival rather than waiting for an emergency situation.

For example: You've arrived in a new city. Open the app, see locations for ERs. Drive to each and familiarize yourself with the surroundings.

Yup, exactly. Recently I traveled to California and used the app to familiarize myself. I noticed that the closest hospital was 30+ minutes away. It was just good info even though I never really needed it.
It isn't reasonable to expect an app to replace 911.

That might not be what plays out in court, but the problem there is people aren't reasonable.

Hmm, this is a tough one. Given that it's just a different visualization on publicly accessible data, does the app have to safeguard against this in a special way? I used a disclaimer in the app, and went into detail in the privacy policy about how the app claims no responsibility to how the data is used. Do you have any suggestions as to a good way to handle these types of situations in general?
It's so people can plan out their visits and know what their options are in a new area.
I am also very interested to know where the data comes from. Great idea and great execution!
I want to be as open as possible about where the data comes from. It's pretty darn important to understand potential flaws in the system and I'm hoping to identify and (help) fix it.

The data comes from http://www.medicare.gov/. If you dig deep enough you'll see that the government does extensive surveys on each hospital and tries to catalog everything.

Two important deficiencies in the data that I am thinking about:

(a) Data is updated in real time. (b) There is no existing mechanism for Health Scout to help updating the gov data.

I agree the above will take some work and I intend to try through Health Scout :)

I'd suggest looking at openstreetmap for hospitals with emergency rooms. This might also give you an easy way to allow people to update the data.

OSM has http://wiki.openstreetmap.org/wiki/Tag:amenity%3Dhospital

  amenity=hospital
  name=Royal London Hospital
  emergency=yes
I've been working on this over the weekend. Due to imported data, the quality of hospitals in OSM (for the US) is pretty low. There are hundreds of doctors offices, clinics and nursing homes that are listed as hospitals.

(I've been correcting the data in my region. I have the idea to do some analysis for the entire country and incite people to clean up their areas. The Medicare dataset mentioned in this thread really makes that easier.)

Here's a query for the OSM tagged emergency rooms near me: http://overpass-turbo.eu/s/7pL (move the map and press "run" and it will update)

https://www.mapbox.com/blog/coffee-directions/ has a "coffee finder" that will give directions, but replace "coffee" with "emergency room" and it suddenly has a much greater impact.

That query works better than just looking for hospitals.

At a glance the rest look okay, but I converted one of the results into a 24 hour clinic (which is a better semantic match for urgent care than hospital+emergency=yes).

Thanks so much for this awesome app :) Multiple times in my past I've needed an app like this. If I may offer two suggestions:

- Can you show hospitals without average wait times? I just tried the app and it doesn't show me my preferred ER because medicare.gov doesn't have any wait time listed for it. That might weed out good ERs for no reason. (It's my HMO's hospital. I would travel 2 extra miles in nearly all cases for ER staff that have direct access to my medical records, the easy insurance situation, and the super easy followup with my doctors.)

- A way to search ERs by manually entering a location, not just showing the nearest. Unfortunately, my parents have a habit of calling me rather than calling 911 or just going to the nearest ER. While 99% of the time it's the ER near their home, sometimes I need to look it up. They're averse to learning any new app :(

Great ideas.

(1) I will add in the functionality to show nearby hospitals without average wait times. It seems easy enough and very useful without search functionality. FWIW, if the government doesn't report wait times, I default to the average ER wait time for your state. If there is no average ER wait time for your state, then I use the national ER wait time reported by the gov. May I ask where in the country you are located?

One thing to note, I explicitly query for hospitals within a 10 mile radius. I figure if it's more than 10 miles out, it's probably not worth the drive. Maybe I should rethink that assumption...

(2) The search functionality is definitely coming in an update. I've heard that it would be useful from numerous people.

> I default to the average ER wait time for your state.

Ahh, I did see that in the app after I wrote my comment, but that was the only thing I could think of that might affect the search results :/ I'm in San Francisco. 10 miles would cover most ERs in the city ;) but in more specific terms, the particular hospital I can't find is Kaiser SF, only ~2 miles away from me. I see hospitals 5+ miles further away in my list... so.

I wouldn't mind a view that opts out of wait times altogether too, or sort by travel time only. Most times I go it's serious enough that I rarely wait more than a few minutes, so I don't pay attention to wait times. And I definitely do a mental sort when presented with the ER a couple blocks away with the 20 minute wait, or the 20 minute away hospital with the super short wait time...

One minor nitpick, isn't the listed average delay more a measure of the rate of non-ER traffic in the ER?

I mean if you're in allergic shock or diabetic coma they'll treat you immediately, but if it takes an uninsured person 8 hours on average to be treated for a simple cold, then that doesn't really fit the theoretical use case for the app?

Good point. The times that are reported are what the government has captured as time until you get to see a doctor. I haven't confirmed but I think you're right. If it's a life threatening situation the hospital will probably treat you right away (I sure hope they will, anyway). I'm also not sure who makes the call whether or not you should get treated right away.
It looks pretty slick, nice job. I echo other people's questions about the source of the data etc.

Also, how do the emergency services work in the US? If you dial 911 and ask for an ambulance, will you not be taken to the nearest ER anyway? (Asking as an ignorant Brit!)

Ambulances do take you to nearest ER but more and more health insurance plans do not cover the cost of the ride (which is $1000-$2000) or have a $500 deductible for such.

Unless you're in real trouble, its often cheaper and simpler to have someone drive you.

Yes, the ambulance will take you to the nearest ER. I guess there are conditions where treatment is urgent but not so urgent you need an ambulance. Also, it can often be faster to drive in than to wait.
I found this out the hard way. Even though I live within walking distance of an emergency room, the 911 operator told me to wait for an ambulance to arrive (even though it wasn't an emergency). The net result was me waiting 20 minutes for an ambulance to arrive and take me to the hospital, instead of walking to the hospital in 10 minutes. Luckily my insurance covered the ambulance ride to the hospital!
When I have free time, I spend some time working on ALS ambulances to maintain my certification level.

How EMS works in the US generally follows this path:

You get sick.

You then dial 911.

A dispatcher talks with you, routes your call to an ambulance company, and provides us with some information on what's going on. If the environment is no danger to you, they should generally tell you to stay put. Depending on the nature of the emergency, they'll have some first-responder/stopgap measures they'll want you (or bystanders) to do. If you have crushing chest pain for example, they'll explain how to place yourself into a recovery position (in case you vomit while unconscious), they'll ask for a nearby bystander (in case they have to instruct them how to perform CPR), and they'll do their best to keep you calm and aware of your situation. What they don't want you to do (even if you're 10 minutes away from a hospital) is try to move yourself there because:

If you become unconscious on the way to the hospital, it becomes so much harder to find you (rather than if you had stayed in a static location that you've described).

By moving, you might move from a safe environment to a more dangerous environment (sidewalk to middle of the street).

When the ambulance arrives, the medic will perform their own assessment of your condition. If you're alert, they'll ask you questions, ask you to chew some aspirin, place you on a heart monitor. If you've become unresponsive at this point, then slightly more drastic measures are taken.

When they load you into the ambulance, they'll make a determination (based upon their assessment) of where to take you. The nearest ER may not have the proper resources to deal with or handle your condition.

I'm unsure how emergency departments in hospitals outside of the US work, but here, not all EDs can provide the same level of care. If you're having an MI (heart attack), and you need cardiac care (a facility that can crack your chest open), it's far faster for the medic to assess you then immediately take you to the correct place.

As a contrary example:

If you happen to make it into an ED on your own foot-power and manage to explain what's going on to the triage nurse, then they still have to call in an ambulance to transport you to a hospital which can provide an adequate level of care. In some hospitals, they won't even begin to treat you (because of liability issues...it's horrible).

TL;DR. Some ERs cannot provide a certain level of care (Trauma Centers, Cardiac Centers, Psych, etc) and thus the closest ER may not be the correct ER to go to.

As a medical student and EMS provider, I agree with all of the above except "In some hospitals, they won't even begin to treat you". Unless you end up at the rare hospital that doesn't accept Medicare, the ED's are required (by EMTALA) to provide a screening examination and treat any patient with an emergency medical condition. For example, any ED should be able to provide you initial care if you're complaining of chest pain (EKG, Chest X-Ray, Nitroglycerin, Aspirin etc). That said, if you're actually having a cardiac event, you'll need to be at a cardiac center to actually get definitive treatment, so the smaller outlying hospital will call an ambulance to transport you to the appropriate facility.
Wow, this is incredibly useful info. Thank you! It still baffles me that in some cases hospitals won't begin to treat you because of liability issues. I guess that's a different discussion though...

In general, if you're ever in doubt, it's best to call 911. Let the professionals handle it.

As a parent, I can say this is a STELLAR idea. Thank you for taking the time to make it.

I'm downloading it right now.

This is very relevant to me, my wife and I are planning a 3 week Europe trip for next summer with our two small children.

This is really awesome, and WELL DESIGNED! Looks very easy to use, and such a great purpose.

Have you thought about the legal ramifications if someone depended on your app and the info was outdated causing a lot of ache?

Thanks for the praise. I definitely have considered the legal ramifications. I'm going to do my best to keep the data up to date. However, I've attached a disclaimer in the app that explains the data may not always be accurate. It should be used for informational purposes only.
Would be helpful to see user reviews for each hospital... and if the hospital is good for out of country/state travelers.
Awesome idea. Good work with the clean looking UI. Looking forward to when it's available in Canada.
Thanks! Hopefully in the next update.
This is a really neat app and brilliantly executed! Good luck with it. Would love a UK version!
Be gender neutral, some fathers might find title a bit offensive.
The title isn't offensive whatsoever, it's a brief origin story.
Fair enough, i had not noticed a singular form.
As a father I don't find this offensive at all.