NOSH ChartingSystem

A new open source health charting system for doctors.

Leave a comment


One of the noticeable changes with NOSH 2.0’s encounter layout is the presence of the SOAP note.  For those not in the medical field, SOAP stands for Subjective, Objective, Assessment, and Plan.  It’s a very simple, clear way of noting an encounter, both in the inpatient and outpatient setting that I, as a medical student and resident many years ago and using paper charts, had learned to utilize to communicate to myself and others on the care team.

Since the advent of EHR’s, documentation has veered farther and farther away from this simple SOAP note.  And to this day, I still don’t know exactly why.  The obvious reason to me is that we’re moving towards structured data, starting with patient demographics and vital signs, but now it’s in the actual note-taking too, thanks to Meaningful Use.

But that is where the line is crossed, probably with most physicians, where usability of the EHR becomes unusable and frustrating.  Physicians don’t think or notate  intuitively in structured data.  Humans are not bits of structured data, especially in the subjective section.  Physicians describe the patient and the encounter as if it is short story.  The patient is a story.  How do you describe skin lesions, and psychological manifestations such as speech, affect, presentation in a structured way?  That is why medicine can’t be like a restaurant where there is a set menu with finite descriptors or items.  However vast the SNOMED CT definitions can be; it still does not give us the full picture; especially those in primary care dealing with patients who have chronic illnesses.

Since NOSH was never Meaningful Use certified to begin with (and proud not to be), I felt I could make things simple for the physician user.  The first iteration of NOSH was better, in my opinion, than current EHR’s, but it still was too regimented for my liking.  As it became clear to me how the new template engine was going to work, I finally felt that NOSH 2.0 will allow me to really simplify the encounter down back to the good ol’ SOAP elements.  Now that brings back good memories.  And happy physicians too!

Screenshot from 2017-02-19 16-21-29

Even on a mobile device, NOSH 2.0’s simplicity really makes it easy for documentation to happen; and if you really want some structure data that is useful and customizable to your needs, the tagging feature of NOSH is clearly the best way to go than structured data fields scattered throughout the encounter.  You can query tags that are created and saved for each item in NOSH.  There will be a time when NOSH will be able to auto-generate tags based on natural language processing of the note so you don’t have to think of the tag you want to use too!

Excited to see it in action?  Try the full featured demo or use NOSH-in-a-Box today!

Leave a comment

The Dark Ages

Mazen Elkurd, DO, a neurology resident at Georgetown University, recent generated a post on KevinMD about how “Medicine is Stuck in the Dark Ages.” He speaks about how technologies are so far behind in the field of medicine and rightly points at some of how HIPAA regulations are creating unintended consequences regarding patient data access and undermining the very security and privacy concerns the legislation was meant to address.  This funny YouTube video accurately highlights the real absurdity of our health care system today.  Believe me, as a practicing physician, the maze that patients go through to get health care services and their information is really embarrassing.

My response to Mazen’s post: WE (meaning patients and physicians) HAVE THE POWER AND THE TOOLS NECESSARY TO CHANGE THIS.  It’s not a dream device or a hope that Google, Microsoft or Apple will create the EHR of our dreams (BTW, reading that article made me laugh so hard…I wish Micro$oft or Apple the best of luck on overthrowing Epic).  The tools that exist today, exist through open source projects that many other industries are utilizing and exploring.  It fosters on the idea of privacy, security, and the idea that data does not and need not be stored in a centralized (or accurately put, a “honey pot”) way.  The current, unacceptable, way health data is stored and collected (by design, and by human choice through Meaningful Use) is the siloed, centralized method that translates to non-reconcilable, inaccurate, and useless healthcare data.  It’s frustrating at best, but dangerous for patient safety at its worst.

What are the tools I’m talking about that already exist?  They are:

  1. An open source EHR that focuses on user interfaces designed for physicians and patients to better health data focused on the patient.  The working code in an open source project is subject to and thrives on peer review so that physicians and patients can continuously improve on it as they see fit without relying on a third party that works against our interests.
  2. OAuth2 for single-sign on so that physicians and patients are not relying on username and passwords that get forgotten, lost, shared, or hacked.
  3. User managed access (UMA), a subset of OAuth2, so that patients can set access for physicians, institutions, caregivers, or applications to their health information
  4. Blockchain for identity verification (which again is to minimize the use of usernames and passwords) and auditing for actions done to the patient’s health related information for data integrity.
  5. FHIR for health information transactions once UMA and the patient that controls it determines appropriate access.  When combined, all leads to:
  6. A distributed network of singular patient data, controlled by UMA, FHIR, OAuth2, and Blockchain, that is not centralized or owned by any particular entity except the patient, so that data protection, security, and integrity are maintained.

What does a distributed network mean?  Most of what we see today is a centralized repository of data (or what I call a node) stored one one large entity (like Google, Apple, Epic, a hospital) with data of millions of people in one server or service.  If a nefarious hacker was wanting to break into any one node (and it only takes one), the hacker could easily get health related information on millions of individuals quickly with very little work.   One has to also assume that there is no bullet proof way to secure any node.  So if a nefarious hacker really wants to get your data, it’s pretty likely that is going to happen especially in a honey pot scenario.  So by spreading the data around to millions of nodes instead one, it would takes a lot of work for the hacker to get your data.  That’s the future of data security that no one in health IT is even remotely addressing.  These recent cyberattacks on hospitals to get data for ransom are just the start and there is no way to really stop them in the future.

The good news about this distributed network solution to healthcare data?  No one “owns” any these technologies that exist today.  So, in essence, patients and physicians HAVE the immense power to harness and utilize them.  We are and can no longer be beholden to EHR companies to give us what we want.  We are no longer shackled by inferior and backdated technologies that hold us back in the dark ages.  Being patient-informed in a distributed network solution calls for a complete, but necessary overhaul of how we currently implement health IT.

Most patients believe that health care should be a simple transaction they have control over  and that there is only one data set for one person as the YouTube video suggests.  But the disconnect between reality and the dream appears to be so wide that we’re just sitting on the sidelines…complaining and dreaming.

But it doesn’t have to be this way.  We can wake up from our health IT nightmare right now if we choose to.  The solution is in our hands.  NOSH and HIE of One, both open source projects that harnesses all of these technologies, aim to unlock the shackles that hold us down the path of the health IT dark ages and is ready to be served.  Are you ready?

Leave a comment


NOSH 2.0 has been redesigned from the ground up, starting with the use of Laravel 5.3 as the PHP framework.  This is a huge step up from Laravel 4.2 which was the foundation for NOSH 1.84 and has allowed better support for modern extensions including making PDF documents for printing (courtesy of TCPDF, which replaced wkhtmltopdf), an updated OpenIDConnect library (for single-sign-on) and the ability to parse YAML, which serves as the foundation of the new template library.

Here’s a snippet of what YAML looks like compared to a CSV (comma separated values) file:




  Color: Yellow
  Taste: Yum
  Color: White
  Taste: Yum
  Color: Orange
  Taste: Yum

So it’s more readable, right?   For templates, this is cool so that even non-technical doctors can create their own templates if they have a text editor.  Soon enough, we can have all these template YAML’s so that they can be shared (since there is no license or restrictions on YAML’s).  There is a pretty good default one already set up for each user for a standard NOSH installation, but one can customize it on the fly or one can copy and paste from one user to another.  It’s super flexible.  See the screenshot below where the templates  (on the right) are tied to a large text box (which is set up on NOSH to be the default).  Each text box has it’s own template groups and each group as a set of items to copy into the text box.

Screenshot from 2017-02-19 16-21-29

YAML’s are also great for another reason in that it gives a lot more flexibility for sections of the patient encounter that can possibly grow (infinitely) without being limited by database field or character size.  For instance, the family history is one such section that needs to have an adaptable database field and is used to build a tree (thanks to sigma.js).  So you get a view of this from a YAML text:

Screenshot from 2017-02-19 16-22-32

Vital signs are another area where YAML makes an entrance on NOSH 2.0.  And you get graphs like this:

Screenshot from 2017-02-19 16-22-10(That poor fake patient grows like a rubber band, but that is beside the point…)

So that’s a brief snippet of YAML on NOSH.  Hope you enjoy and more to come about what else is under the hood of NOSH 2.0.  Check out the demo, if you haven’t already.

Leave a comment

NOSH 2.0

Today, NOSH 2.0 is now pushed out to NOSH-in-a-Box and is ready for launch.  This is the easiest way to install your own NOSH without having to know how to install a Linux operating system, install all the dependencies, and managing databases.  Instructions are here.  If you already have Vagrant installed, here’s the download to the NOSH-in-a-Box Vagrant files.

For those that are technically savvy you can install NOSH on a Linux machine and in the command line,

sudo bash

If you are updating from a previous installation of NOSH,

sudo bash

With the update, you can still use NOSH 1.84 running on nosh-old instead of nosh in your browser’s address bar.

Today I’ll be talking about the most obvious change to NOSH 2 when you look at the demo.

NOSH 2 is now mobile-friendly, meaning that the user interface is optimized for both desktops AND smartphones AND tablets.  No more squinting or zooming to see it.  And since it’s not a dedicated app, it works across most modern browsers irregardless of the device you use.

You’ll also see that some workflows have been re-arranged but simplified and the template engine has been re-engineered with YAML, so a physician can edit the templates through NOSH or on your own with any text editor without causing inadvertent thermonuclear war.  Since each user has their own template, they can share their templates with others through a simple YAML text file (which I hope can be shared through a GitHub-like repository).  I call it NOSH on YAMl (or yams are tasty).


A lot of design emphasis for NOSH 2 was on being able to quickly get to the tasks that physicians and patients use regularly.  Less clicks to get work done.  A patient timeline is the first thing a user sees when they go to a patient’s chart.  This new version of the timeline fits in with the new user interface.  Everything is seamless and integrated.  Nothing looks out of place with clean lines and consistent and clear buttons.  No need for a user manual.  No hidden tricks, which is just the way I like it as a practicing physician. For patient’s the patient portal has been completely reworked with more power to the patients with medication/problem list/medical history editing and reconciliation with your physician.  Next week, I’ll get to indulge a little of how NOSH 2 was built as well as my companion project, HIE of One.  Have a great weekend!


Leave a comment

Crying over spilled milk (personal)

Lots of important and exciting changes are coming to NOSH soon.  But before I present more posts about this, I’d like to take a moment to vent based on the recent developments in the ONC/Meaningful Use/MACRA/Primary Care/ACA world.  In my view, these negative developments started to take shape in 2009, which was the beginning of the end of my solo family practice, which was finally laid to rest in 2012.  And now in 2017, I can’t help but feel disgusted.

The big news in the Health IT world last year was the doing away of the Meaningful Use program (even before it was supposed to be finished).  This is what I had anticipated way back when and I reasoned that it was going to be an expensive, meaningless endeavor that was going to lead to 2 things: 1) consolidation of the EHR market so that the big market leaders are going to grab everything and leave nothing left for innovation and 2) consolidation of primary care physicians and the death of the independent medical practice for those who are still seeing Medicare patients.  I was secretly hoping none of the 2 things were going to happen but I’m sad to say that after nearly 7 years, both have come to fruition.

And the architects of this legislation are now crying over spilled milk.

They talk about unintended consequences, but I think that is a very poor excuse.  I was just a low-paid primary care physician with a little solo practice, and I was not a full time policy wonk.  But somehow I predicted this was going to happen because I could see it through the lens of my micropractice.

Instead, I think they know exactly what was going to happen and turned a blind eye towards the consequences.

They can’t turn back time and say to the primary care folks, “Oops we made a mistake, please let us continue this torture because we can’t turn back because if we stop, it would be, yeh – crazy.” or “Oops, please forgive us, but we need you back somehow but please let us torture you some more”.

It’s too late to cry over spilled milk.

And one wonders why we still don’t have interoperability between EHR’s when in fact standards were developed and the largest EHR vendor decides to break them?  Shouldn’t there be certifications or legislation for playing by the rules?

And one wonders why EHR certification is a solution looking for a problem and why the wrong kind of certification like this one is onerous to innovation?

And one wonders why physicians are outraged because we’ve been treated condescendingly by this legislation and being forced to use broken tools and are told, “give us better care!”

And one wonders why physicians are told to buy these expensive, broken, and useless tools and are told, “give us better care for less cost!” and if we can’t give them whatever it is they want, we get paid less anyways.

And one wonders why physicians are told to buy these expensive, broken, and useless tools with the intention to recoup these costs and then snatching the program away and build a new one, never recouping the cost anyways.

Do they think this is funny?!  Do they think these unintended consequences are academic?

These are real lives, real people, real physicians, and real livelihoods.  The toil and trouble for primary care physicians is not academic.  It’s been going on for decades and with this legislation, the house of primary care has been turned to ashes.

I know sometimes I should not even think about it and walk away.  But the curiosity inside me and me being trained as a trauma-informed medical provider requires that part of my healing process is to know what happened, a post-mortem if you will, even though I’m going to re-experience the pain.  I felt that a great opportunity was lost and the damage that has been permanently done in the name of crony capitalism, monopolization at the cost of cooperation, innovation, and evidence-based approaches are now coming home to roost.

But like the great Elton John once sang…”I’m still standing”.

As a physician.

As a healer.

As a hacker.

As a coder.

As a husband and father.

And from the ashes, a phoenix will rise.  Stay tuned…

Leave a comment

Forget Metrics for EHR adoption…INTEROPERABILITY is where we NEED to go, but has it left the station

A lot has been discussed about MACRA lately and the potential impact on physicians and the EHR marketplace.  The post-mortem analysis about the Meaningful Use legislation has primarily focused on the critique that there were too many metrics for physicians to manage to prove they were using the EHR’s in a meaningful way.  Although this is a reasonable conclusion, I would argue stepping back even further and seeing what was really the whole point of Meaningful Use in the first place to get a better assessment of the damage done.

The good intention, I believed, was to harness the technology of EHRs (by being digital, as opposed to paper), so that these EHR systems were to be able to talk to each other.  This is not a foreign concept as there are so many examples today of how digital has transformed the ability for data to go from one container to another (e-mail, websites, social media) and even now with single sign-on solutions, current technologies (RESTful APIs, HTTPS, oAUTH) makes data transfers relatively trivial and secure.

It is laughable to me now that we are nowhere near where we should be to get true interoperability given the existence of these well documented and proven technologies.  The framework of the current crop of EHR’s with the largest market share is not focused about interoperability, but it’s about a monolithic, siloed entity that does not care whether you practice in a hospital, outpatient clinic, mental health office, child abuse evaluation center, nursing home, dentist office, or a direct primary care office.  Never mind that these EHR’s has 95% of everything a health care provider doesn’t need to use or see and that he or she has to use extra mental capacity to filter all of it out just to make it work way that works for them.  Never mind that these EHR’s will still not have the ability to automatically get the most up-to-date information about the patient you are taking care of in front of you because there is a pretty good chance that this patient may have gone to a different clinic or hospital, added or deleted a medication, allergy, or diagnosis without your knowledge, and it has not been reconciled with your EHR.

Effectively, these EHR’s are just an expensive, digital form of paper except it doesn’t work any better than the old-fashioned paper.  At least with paper, you can organize your chart YOUR way.  At least with paper prescriptions, patients can provide a signed physician prescription to different pharmacies if they want to price shop.  No wonder most doctors are disgusted and frustrated.  Doctors and patients are now left to choose and use only a handful of non-useful tools now that the EHR marketplace has been decimated to only a few big players thanks to Meaningful Use.

What’s next?  In general, these few big EHR players aim to increase their profit and market share by doing the walled-garden, monolithic, sliloed approach to interoperability.  Think about it…it wouldn’t make any sense for them from a business standpoint to share a standard interface to talk to another big vendor EHR if they can do anything about it.  They will delay, delay, and delay to prevent interoperability from ever seeing the light of day.  In the meantime, they hope to convince hapless hospitals and clinics that they will benefit to having their product because everyone else in this city/county/state/nation uses their EHR so if you want to talk to them, you’d better use the same one as every one else…even if it costs an arm and a leg…even if it doesn’t work well for the doctors and patients.

In other words, the train we needed so desperately has left the station because we wasted time going the rabbit hole instead.  And it’s likely not to ever come back in the form we expected from several years ago.  So what is a doctor to do if they are forced to use an EHR they don’t like AND they don’t have any leverage to convince the EHR company to improve their product so that it works for them?  Stay tuned…

Leave a comment


Another year has come and gone and although it’s been quiet on these blog pages, I will say with absolute certainty that NOSH is not by any means quiet.  I’ll be making a variety of blog posts in the next few weeks about what NOSH has been doing under the radar, but first, I’m most happy to share that NOSH ChartingSystem is now going Vagrant!  Don’t worry – NOSH is not begging for anything and it’s not living in the streets.  Vagrant is the super cool technology that allows one to build a full-working virtual machine (a computer within a computer for the non-techies), customized to the needs of NOSH.  What that means to you is that NOSH can be installed on any machine (Windows, Mac, Linux, or even in the cloud such as DigitalOcean) as long as it’s constantly online.  The heart of Vagrant technology is the concept of Boxes and that’s where NOSH-in-a-Box got its name.

Because NOSH-in-a-Box makes it super easy for interested physicians or clinics to try and actually use it in a real-live environment, it will now be the recommended installation method going forward.  Since it is a virtual machine, it is sand boxed from the host computer operating system so it doesn’t mess with it in any way.  You can even port it to another machine quickly if your main machine dies and if you have made a backup of the virtual machine.

The current NOSH-in-a-Box and how to install it on your system is right here.   If you’re too excited to try, may I ask that you wait for a few more days?   As of right now, the installation program is using NOSH 1.84.  But in a few days, we’ll be moving to NOSH 2.0.  Yes..a jump to 2.0 means a whole new NOSH.  A NOSH that is mobile-optimized so you don’t have to squint your smartphone or table to use it.  And even though NOSH was already simple to use compared to most EHR’s, NOSH 2.0 is easier and faster than ever to use.  Go munch on that!