Amanda DiTrolio
💡 HTN | Community Brain Trust | 9/13
🧵TOP THREADS OF THE WEEK
In case you missed them, here are highlights of a few interesting conversations from different channels:
Threads included below:
- Understanding commercial payers' treatment of outpatient facility services
- Exploring the technical lift of pulling claims data from a FHIR repository
- Cracking the code on in-home pediatric care models
- Telehealth options for connecting ER patients to psychiatrists
- Availability of risk adjustment model methodologies in CMS vs Medicaid
1. Understanding commercial payers' treatment of outpatient facility services
Q: Hi there - does anyone understand how commercial payers treat outpatient facility services that fall under the CMS APC (ambulatory payment classification) program? We’re trying to leverage the payer MRF data to negotiate our bundled rates and realizing that the institutional rates don’t make any sense. I’m assuming the commercial payers have a similar APC-type program for outpatient facility services.
For ex: Blue Shield California reported their negotiated rate for an OP knee replacement (27447) at Providence is like $5k… something is missing here.
– Zach Mitchell | via #buildersask
Thread Summary: A few Nerds try to reverse engineer commercial payer MRF data to understand realistic costs that inform how to negotiate bundle rates. While there are a few thoughts on approaches to do so, consensus is ability to do so will vary drastically by the quality of payer data.
Top Response:
Erica Everhart: I think the institutional/facility fees are generally a mess in the MRF data. I have especially noticed it with outpatient facilities like you're pointing out.
I see institutional 27447 rates for Blue Shield California for orgs with providence in the name ranging from $3868 to a couple over $10000, but many in the $4-5K range that you're seeing. Agree it is weird, but also from my experience not uncommon.
I think we all are trying to figure out if there's some way to convert the MRF data into realistic costs. My opinion is that it is a solid maybe, depending on the quality of the payer data which varies drastically by payer/region/etc. we (carejourney) are working on making something akin to recipe books by analyzing a ton of claims data and looking to see what all gets billed together by site of service so that we can help with this sort of analysis. I agree there's no way that is what they are getting paid in practice.
Check out the full HTN Slack convo here!
2. Exploring the technical lift of pulling claims data from a FHIR repository
Q: A prospective client who is a payor asked if we can pull claims data from their FHIR repository directly. I want to explore the technical lift on this. What questions do I need to ask this prospect to further clarify the specs of their repository?
Also would welcome any other input on the complexity of this, the variability of the complexity of it depending on the payor or the specs of their repository, and any general starting points for learning. I know what FHIR is and the use case they are describing, so I am not starting from zero but decently close to it
– Mark Liber | via #buildersask
Thread Summary: The crew jumps in to help build a great list of thoughtful questions to explore the technical lift of a claims data pulling exercise for one startup's payor client.
Top Response:
Matt Pfeffer: You will also want to look at:
- what data elements are necessary and valuable for the intended use case(s)
- are those data elements included in their FHIR data
- is the data consistent (probably not, so really the question is, what kinds of variability are there and how can you mitigate them)
In my experience the data provider doesn't know the answers to these kinds of questions about their data, and you need to get access to the data itself and analyze it to find out.
Check out the full HTN Slack convo here!
3. Cracking the code on in-home pediatric care models
Q: Can someone tell me why there is still no in home pediatric service? Even if it's a premium price?
Is this not a large enough pain point for parents?
(Writing this while waiting 30 mins in the waiting room for my daughter's check up)
– Daniel Yuabov | via #buildersask
Thread Summary: Following one HTNer's delayed waiting time their child's doctor's visit, the group tries to unpack why a model for in-home pediatric care services doesn't exist yet. While it seems there are a handful of reasons why this model hasn't worked yet, one member (Kelsie Brandt) shares how her startup Nest Health is working on exactly this - in-home primary care for families.
Top Response:
Josh Pappas: When I was in the RPM/ H@H space the challenges were:
#1 Reimbursement Medicare covered H@H and RPM but typically no coverage Medicaid/ Peds
#2 RPM devices for peds were limited and challenges with FDA
Check out the full HTN Slack convo here!
4. Telehealth options for connecting ER patients to psychiatrists
Q: At Akin we work with families who's loved one's have SMI in an education and skills building program. one of the excruciating things families are going through is boarding at the ER. It's unclear if there is any actual psychiatric support in this setting. It's heartbreaking to hear about. I was curious, are there any telehealth options that connect ER patients to psychiatrists? Any other recommendations for this type of situation?
– Stephanie Greer | via #topic-behavioral-health
Thread Summary: In a discussion about the challenges faced by families with loved ones with serious mental illness (SMI) who are boarding at the ER, HTN members discuss potential telehealth options for connecting ER patients to psychiatrists. Suggestions include Willow Health, which operates in this space but it's unclear if they focus on ER diversion or supportive care in the ER, and Qler Telepsych, which embeds psychiatrists via telemedicine into ERs.
Top Response:
Arpan Parikh, MD, MBA: It's a really challenging situation. I think about it as two interdependent, but distinct challenges that create these heartbreakingly long lengths of stays in emergency rooms for patients suffering from psychiatric disease:
1. The emergency psychiatric evaluation itself. This is a B2B solution, with many players providing telepsychiatry coverage to hospital emergency rooms (including the big national telehealth players such as AmWell / TDOC as well as some regional BH-specific players).
2. Finding the next level of care. Once the psychiatric evaluation has been completed, it's not always simple to connect the patient to the recommended level of care (particularly if inpatient hospitalization is recommended). I know a few folks exploring this space right now, and there are others who are building the higher-intensity, more accessible (though not in-network, yet) PHP models that can sometimes speed up the disposition planning for psychiatric emergency room patients (such as Willow, as @Ryan Gallagher mentioned above).
Check out the full HTN Slack convo here!
5. Availability of risk adjustment model methodologies in CMS vs Medicaid
Q: Why are CMS and HHS risk adjustment model methodologies publicly available, while the various Medicaid models are paywalled? (ex: CDPS, 3M CRG, Johns Hopkins ACG)
Are license fees for the various Medicaid models just a cost of doing business in that space? The fees are surprisingly high.
– Chris DiBlasi | via #buildersask
Thread Summary: HTNers unpack the reasons behind the public availability of risk adjustment model methodologies in CMS and HHS, while the various Medicaid models are paywalled. The conversation explores the historical development of these models, the role of vendors and payers, and the complexities involved in calibrating the models for Medicaid sub-populations.
Top Response:
Sam Stearns: I think this this is a bit of historical accident...
These groupers were started as academic projects (e.g., ACG at Hopkins, CDPS at UCSD, DxCG at BU) and then organizations/companies emerged around them.
CDPS is the most common grouper in Medicaid and is maintained by UCSD. This paper has some history.
Medicaid may not be freely available due to the unique aspects of Medicaid sub-populations (e.g., TANF, Disability). I can imagine it is complex to calibrate the model weights across different eligibility categories, in addition to keep the codes up to date. And the fragmentation across states would make the business model more complex.
HCCs date back to the 1980s and were selected by CMS in 2000. This paper gives an overview of the history.
Check out the full HTN Slack convo here!
🤖HTN KNOWLEDGE BOT
If you have your own question(s) to ask, don’t forget that a good place to start is our HTN Knowledge Bot. It’s our smart search tool that makes it easier to access the wisdom shared within the HTN powered by ChatGPT. You can log in and use it on the website (here) or see how to use it directly in Slack here.
Check out the example ask below!
Member Question:
I want to learn more about applications of AI in healthcare.
Knowledge Bot Answer:
Certainly! Here are some resources that provide information on applications of AI in healthcare:
1. Article: Synthesizing medical research for HCPs - This article discusses the use of AI to synthesize medical research for healthcare professionals, which can help improve decision-making and patient care.
2. Article: The State of AI in Healthcare - This article provides an overview of the current state of AI in healthcare, covering various applications and trends in the field.
3. Article: AI supporting the future of patient education at scale? - This article discusses the use of generative AI for patient education, highlighting the potential of AI in delivering personalized and scalable educational content.
4. Article: AI Paving the Future of Digital Healthcare - This article explores emerging trends in AI and healthcare, focusing on provider-related tools and discussing the potential of AI in improving healthcare processes.
These resources should provide you with a good starting point to learn more about the applications of AI in healthcare.
✨HIDDEN GEMS
Here we highlight helpful resources from across the community:
- STAT Generative AI Tracker: A guide to the health systems and companies driving adoption – A nice resource from STAT tracking health systems' usage of generative AI tools - detailing applications, target end users, and more.
- GTM Cheat sheet for Medicare Advantage Organizations via Jonathan Wakim – A helpful GTM resource for startups selling to Medicare Advantage organizations, including answers to important questions on determining your co's value prop, plan for addressing barriers to adoption, and more.
- IMPACT Payment & Coding Toolkit from DiMe and ATA via Jim St. Clair – This toolkit offers virtual first care providers a set of resources to better understand healthcare coding strategies for digital first care delivery - including basics of payment, reimbursement pathways, negotiation tactics, and more.