FHIR is Fire
https://eccoviasolutions.com/wp-content/uploads/2017/06/FHIR.png

FHIR is Fire

You remember the last time you texted an Android phone (yes, I’m assuming you’re an iPhone user) and a green bubble popped up. Or, worse, you sent an image and it turns up all grainy. The reason for this is due to the encoding formats that differ between androids and iPhones. However, you’re still able to communicate between these two different phone providers. In healthcare, there has been a large emphasis on the ability to communicate between different healthcare providers since the Electronic Medical Record Mandate passed in 2009. At first, various healthcare organizations started digitizing their patient's information, in order to facilitate billing, but there was no way to share this information externally. Now, these organizations are communicating using a standard called F.H.I.R. (Fast Healthcare Interoperability Resources). FHIR is a data format that specifies how to structure information related to patients, medications, observations … etc. Any information that may be related to the health or the care of a patient.

FHIR is the latest iteration in an attempt to have health systems talk to each other (see CDAs, HL7,..). Traditionally, you might have had a primary care physician, a cardiologist for your heart health, your chiropractor … etc. However, all of these people who are looking after you don’t have a way to communicate with each other. Your cardiologist may recommend a new prescription for you to take, but how will your primary care physician know? Or, what if you have an x-ray, how do you share your results without bringing the physical copy? Do you want your doctor to fax your prescription to Walgreens? This is what FHIR is trying to solve; the ability to share all of your health information seamlessly.

No alt text provided for this image

Above, the advantage of FHIR is shown. Any, well almost any, type of clinical information from different producers can be consumed be vastly different consumers (the patient, your doctor, another doctor ...etc).

This unique aspect of healthcare is that the entire industry is consolidating around this standard. Not only does this allow individuals to have a consolidated view of their health information from across multiple providers, but, for providers, it also allows for an aggregate level view of all patients to understand trends in population health as well as assisting with clinical decision support (CDS). For population health, now, not only can public health officials calculate data trends at an aggregate level for an individual hospital, or doctor’s office, but it’s possible to calculate these statistics across all patients for all healthcare systems that have represented their data in the FHIR. format. For example, the CDC is working to release public-facing applications leveraging FHIR to disburse information related to COVID. Of course, this is assuming that the patients have agreed to have their data shared and used in this fashion. In addition, this standard also enables CDS. You can think of CDS as a series of rules that fire given certain data. For example, if a patient has a positive mammogram result that is sent back in a structured format using FHIR, then the CDS system could automatically fire a reminder to the patient to schedule a follow-up appointment. FHIR is paving the way for EHRs to move from being a stale repository of information, to actively taking action to improve health outcomes with patient data.

No alt text provided for this image

Above is a simplified workflow of how CDS works. Given a plethora of information on a patient, what course of action should a doctor take?

From a pure data perspective, this is awesome. The ability to understand the care a patient is receiving across multiple providers is the key to unlocking not only a complete picture of an individual’s health, but can, and will be used to usher in advancements in precision medicine, the ability to recommend specific medication or treatments based upon your unique biometric/genomic/health history, predict disease onset, eg. given enough data on the predispositions of diabetes a doctor could start a patient on a treatment plan to prevent the onset all together, value-based care, the payment structure that incentivizes healthcare providers to follow treatment protocols that avoid costly expenses such as ER visits among other opportunities, and other data projects such as understanding social determinants of health, the impact that your zip code or access to parks has on your health outcomes. In fact, FHIR is coming and is here to stay with almost 80% of hospitals implementing this standard shown below.

No alt text provided for this image

Looking forward, I'm excited about healthcare that not only works to ensure hospitals get paid for my visit, but for care that places my health as the primary focus regardless of where I receive treatments, checkups, procedures, or advice.

Mark Stuenkel

Design Leader @ Vista | Ex-Amazon

4 年

??????

David Hoang

AI @ Atlassian | Investor

4 年

Congrats, John! Looking forward to FHIR Festival!

要查看或添加评论,请登录

Jonathan Hilgart的更多文章

  • Leading, and Pairing on, ML projects

    Leading, and Pairing on, ML projects

    Most machine learning projects sit squarely in the intersection between "spend two years on this and get back to me"…

  • Testing Machine Learning Systems

    Testing Machine Learning Systems

    Testing in software development is almost a science at this point. You’ve probably heard the quip, “Write tests, some…

  • Real-time audio processing on a Pi

    Real-time audio processing on a Pi

    Over the past couple of days, I've been working on visualizing the frequency spectrum of audio. My goal was to wire up…

    2 条评论
  • Good Ol' Algorithms

    Good Ol' Algorithms

    Did you know it is estimated people make 35,000 decisions per day! Most of these decisions are binary, but some of…

    1 条评论
  • Primitive functions and the bias they introduce

    Primitive functions and the bias they introduce

    Noah Chomsky has a theory that humans have an innate ability to learn language. This ability to learn a language is…

    3 条评论
  • How Computers Learn to See

    How Computers Learn to See

    Professional wide receivers have been known to catch a football with their eyes closed. Hockey goalies predict where…

  • Information Security & Social Inclusion

    Information Security & Social Inclusion

    I’m currently reading through the book Computer Security: Principles and Practice which discusses security protocols…

    1 条评论
  • Inductive Bias for Different Machine Learning Approaches

    Inductive Bias for Different Machine Learning Approaches

    Source - https://www.researchgate.

  • How to Teach Computers to be `Intelligent`

    How to Teach Computers to be `Intelligent`

    Does this mean anything to you? If this were a Rorschach test, i.e.

  • Computational Time-Lapse

    Computational Time-Lapse

    Whenever you've seen a time-lapse video, it probably looked fairly 'jumpy' from one frame to the next with objects and…

社区洞察

其他会员也浏览了