tag:blogger.com,1999:blog-4384692836709903146.post8709281091122964254..comments2024-03-18T04:38:01.678-07:00Comments on Dispatch from the Digital Health Frontier: Brainstorming About the Future of Clinical Documentation John Halamkahttp://www.blogger.com/profile/04550236129132159307noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-4384692836709903146.post-23476825129159799032013-02-26T12:25:43.943-08:002013-02-26T12:25:43.943-08:00Have you detailed an ROI on all this? Including ac...Have you detailed an ROI on all this? Including activity costs?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-4384692836709903146.post-70185700531731747682012-12-22T17:13:12.989-08:002012-12-22T17:13:12.989-08:00RE: Module 1 - disease specific templates. Since ...RE: Module 1 - disease specific templates. Since there are thousands of diseases, how does one expect to generate the entire SNOMED-CT vocabulary list of such and to include the required quality indicators?<br /><br />RE: Module 2 - I have blogged often about, and agree with Commenter Nick that a Facebook-type user interface be standard for the next generation of all EMRs. Highly unlikely that will occur due to having to rewrite all the code (but all the code had to be re-written with the introduction of the WWW). After all, the medical record is THE most collaborative tool in healthcare!<br /><br />RE: Module 3 - A good NLP or NLU engine should be able to codify SNOMED-CT concepts, and your team is working with a few companies that have good engines. (There are others, BTW.) But most of the current NLP/NLU engines work best using I-9! Seversal years ago, AMIA and AHIMA proposed solutions to this challenge, and currently a group of HIM professionals (including yours truly) is working to resurrect this proposal. In addition, to answer your Challenge #1, you are correct, Dr. H, that SNOMED-CT MUST be embedded underneath EMR plain text descriptions so that no clinicians become coders.<br /><br />RE: Module 4 - Yes, the NLM's mapping of SNOMED-CT to ICD-10 should assist with this. However, your challenge #4 is the big unknown at this time. Because today all clinical coding is performed using I-9 (for billing, reimbursement) and few EMRS embed SNOMED-CT, we don't know. <br /><br />RE: Module 5 - Ensuring documentation is complete and of excellent quality to justify I-10 codes should be realized by "sister" Clinical Documentation Improvement (CDI) systems to Computer-Assisted Coding (CAC) systems. Several CAC vendors are beginning to introduce and / or interface to existing CDI systems. It is unclear from your post if you have considered CDI systems, too.<br /><br />RE: How clinical documentation could be performed by the rounding teams at academic medical centers - it would behoove EMR vendors to explore Electronic Content Management System features and functions, such as document check-in and check-out. This valuable function has been around for a long time and fits into your scenario perfectly. Deborah Kohnhttp://www.daksystemsconsulting.comnoreply@blogger.comtag:blogger.com,1999:blog-4384692836709903146.post-65528761370759178092012-12-21T10:34:23.615-08:002012-12-21T10:34:23.615-08:00As always, thanks for an interesting post. It rem...As always, thanks for an interesting post. It reminds me of a concept I touched upon a few years back, <br /><a href="http://drnic.posterous.com/a-facebook-medical-record" rel="nofollow">Why Facebook should be a template for electronic medical records</a>. Also, your post from 2010, <a href="http://drvoice.blogspot.com/2010/04/clinical-documentation-is-at-core-of.html" rel="nofollow">"Rethinking Clinical Documentation,"</a> was along a similar vein to a blog I wrote around the same time, Clinical documentation is at the core of healthcare reform. <br /> <br />Today, the narrative remains essential but the copy forward capability is problematic at best. In fact, I heard recent guidance that suggests this capability should be switched off. Going digital with our existing infrastructure does not mean losing the narrative, nor the meaning of clinical thinking. Technology, in particular Clinical Language Understanding (aka NLP), is the “Rosetta Stone” for medical notes. CLU unlocks the data in the narrative and makes it available in SNOMED-CT & ICD-10 tagged format – amongst many other coding systems. Still, while technologies like CLU and computer-assisted coding can help ensure the specificity of the clinical documentation created by the care team in line with ICD-10, we need to focus first and foremost on improving physician documentation at the point-of-care. While physicians don’t need to be trained on the intricacies of coding, they do need to have a basic understanding of requirements specific to their area of practice to ensure “right coding” (and this is based on having the right medical information in the documentation) the first time around. In Module 5 you briefly touch upon this but I’d actually move Module 5 to Module 1 because instilling documentation “rules” or CDI insight with physicians must be the first step in this process. Regardless of how advanced technology may become, if you put garbage into this system, you’re going to continue to get garbage out. In the coming year, I think the overall demand to draw data out of physicians' heads and into computerized form will elevate the need for CDI as step-one in an intelligent, streamlined clinical documentation strategy that will support value-based purchasing and quality-of-care measurement, as well as ICD-10.Anonymoushttps://www.blogger.com/profile/15795210188567921398noreply@blogger.comtag:blogger.com,1999:blog-4384692836709903146.post-16739007591860729922012-12-18T18:18:59.753-08:002012-12-18T18:18:59.753-08:00Love the re-engineering of clinical documentation ...Love the re-engineering of clinical documentation to leverage the data and collaboration potential of HIT. Both the process and the output can and should be simpler and more effective than what our EHRs support today. I'm sure the brainstorming sessions discussed the role of the Patient and their family as contributors to the record and an important audience for the documentation produced. Where did the group land on this consideration?<br /><br />And the corollary question of how the inpatient progress note "wiki" becomes or integrates with the patient-centered record for the ambulatory clinic, specialists office, care manager, home health team and patient/family self-care.<br /><br />Thanks for taking on this important work. There are many technical, cultural and regulatory challenges along the way, but it's a path we must travel.<br /><br />GregAnonymoushttps://www.blogger.com/profile/00381668313532393093noreply@blogger.comtag:blogger.com,1999:blog-4384692836709903146.post-14068364567761914652012-12-18T13:29:34.714-08:002012-12-18T13:29:34.714-08:00Sparrow EDIS uses speech recognition and tappable ...<a href="http://www.sparrowedis.com/features/" rel="nofollow">Sparrow EDIS</a> uses speech recognition and tappable templates, combining input from multiple users into a visit summary. All done using iPads. It doesn't do the Wikipedia-type editing of others' notes though. I would think it could be modified for module number 2—you could have the primary author make changes, which are then routed to original authors on their devices for approval or comment. And since Nuance is the speech engine, there is some movement toward clinical language understanding, which could accomplish #3.Brianhttp://www.sparrowedis.comnoreply@blogger.comtag:blogger.com,1999:blog-4384692836709903146.post-6998524684385046952012-12-18T11:54:26.164-08:002012-12-18T11:54:26.164-08:00Re Replacement of all Laboratory Information Syste...Re <i>Replacement of all Laboratory Information Systems</i> John, is there a link to a listing of those LISs?Trevor3130https://www.blogger.com/profile/08148555743063226957noreply@blogger.com