Hacker News Reader

Comment by MDNukem | Hacker News Reader

I'm surprised I don't see any comments about the FHIR api here. It's a project seeking to standardize a RESTful api for clinical encounter data that's gaining a fair bit of traction.

If the existing EMR companies don't manage to subvert its goals, the problem of data interoperability will be largely solved within the next 5-7 years.

View on HN

While FHIR is certainly making health records development more approachable and sane for developers, it isn't going to fix the large-scale data interoperability problem. Even if we completely suspended disbelief and assumed it was perfectly and consistently adopted across all EHR products, it still is just a structured way of exchanging data like HL7- the much more complicated issues preventing semantic interoperability of clinical data still exist.

That's the best news I've heard in a while. I was just complaining earlier today, that I hate the way I have to login to so many different EMR systems to look at my medical records, and there's no interop between them, or (mostly) any convenient API to download data.

I want record from my GP? FollowMyHealth.com

I want records from my cardiologist? UNC's EMR system.

I want records from my bariatric doctor? Duke's EMR system.

I want records from my dermatologist? Yet another system.

I want records from my othopaedist? You guessed it, yet another system.

What I want is to be able to put ALL of my medical records, AND my "fitness" information (stuff from Strava, Fitbit, etc.) into one freaking place. This current setup is horrible. :-(

There are a few startups that do this. gliimpse picknick health

How would that work? The existing provider would have to provide an API or export feature, or else you'd be back to re-entering shit by hand. My problem is that most (if not all) of these existing systems seem to lack any kind of API or ability to dump data in a machine readable format. :-(

They are using the view, download and transmit aspects of meaningful use 2. So they either scrape the portal or parse the downloaded file. The downloaded file is often a variant of c-CDa. However each portal can be different so I assume they are doing the epic and cerner (large dominant players) first.

The more irritating detail to me is that very few places want your records. Even if you carried your up to date medical records on you in some kind rugged secure flash drive, you're likely to be told "thanks but no thanks" if you suggest any medical professionals should look at them.

9/10 they don't want to look. I know there's lots of valid reasons why so I understand that part, but there's a bigger problem it speaks to. To me the greatest failure of modern medicine is that we have a word full of doctors (specifically, general practitioners) who have no time to stay current in their medical knowledge or to learn anything beyond the superficial details of a patient before pronouncing their diagnosis and moving on.

I truly miss having a GP that not only dealt with my medical matters for the first 24 years of my life, but 26 years of both my parents, and close to 20 and 15 years for my siblings as well. There's no way to replace this knowledge now he retired.

Hmm... I haven't encountered that, at least not to that degree. In fact, my cardiologist was happy to get some lipid profile numbers from me, that I read off my phone to him (I had emailed them to myself for safe-keeping) based on my most recent bloodwork done at my GP's office.

Likewise, I just started visiting a new internist who specializes in diet/lifestyle who supervises people going on a ketogenic diet. He asked if I had recent bloodwork, and as it happened, I'd just had a physical about two weeks before at my GP's office. So I fired up the "Follow My Health" app on my phone, pulled up the relevant test results and handed him the phone. He took the notes and says "Great, you just avoided having to have blood drawn today".

Ditto. My current GP was delighted when I lent him my copies of my lab work going back to 1996 for him to copy, and when in every visit I give him copies of updated spreadsheets showing the history of the relevant results going back then, one each for PSA and lipids and the like. Asked me to add liver enzymes to the latter after my previous visit revealed they'd spiked, so that he'd have a better picture of what was going on after the two month later followup to that.

Any doctor who's not interested in this sort of thing when relevant should be fired on the spot; remember, whatever the "God complex" or the like they require to stay sane making decisions that will inevitably sometimes be wrong and harm or kill a few of their patients, they still work for you.

See my comment below; to that I add, have you tried delivering these as paper? I.e. general history on one page (although in my experience that's normally that's covered in a single page of checkboxes), and lab results in both raw and condensed spreadsheet form for the relevant ones?

Note that absent a really difficult medical history, no doctor has time go back through a long history by (many) other doctors trying to find what's relevant and what's not. A "just the facts, mam" summation of your history is a much better start.

I fact, that's exactly what my current psychiatrist got from my previous one who'd seen me for a decade before I moved back to/retried to my home town. Probably took the former a few extra minutes. but he preferred that approach and I'm sure it was easily condensed to a page or so, even a paragraph or two would be sufficient.

That said, my mother was an RN and my father managed the business affairs of a bunch of doctors while I was in middle and high school, and one was our primary hunting partner. And I was on a biology track before I discovered how much better I liked chemistry. So I'm on top of this sort of stuff a lot more than most patient will be, but I suspect this sort of "just the facts, mam" approach could be more generally applied to our current system with benefit.

And if I were you, I'd fire those 9/10ths of ... overly focused doctors until I found a good one like your old one. Did just that after I found a very good one who then went to work for the man in the ER room before the multi-doctor practice he was in failed, then I had to fire two doctors who would not stop forcefully pushing statins, and finally found one who was willing to rationally discuss them, e.g. respect that I had enough of a biology background to balance the picture presented to him by the usual suspects.

Epic actually had a FHIR mini-conference of sorts with a couple organizations this week. Prototypes have already been available to test with for a few months for customers to make their own apps with.

Reply on news.ycombinator.com