This content is for paid members only.

Join for $20/month or $175/year

Amanda DiTrolio

πŸ’‘HTN | Community Brain Trust | 8/1

August 1, 2023
Community Brain Trust

🧡TOP THREADS OF THE WEEK

‍

In case you missed them, here are highlights of a few interesting conversations from different channels:

‍

‍Threads included below:

‍

  1. Medicaid status visibility in the D-SNP market
  2. Patient testimonial no-nos
  3. Data privacy & security in LLM-powered apps
  4. Patient data review timer platforms
  5. Health system EMR finders

‍

1. Medicaid status visibility in the D-SNP market

‍

Q: Does anyone know if MA plans have visibility into the Medicaid status of their members (aka if they are duals) ? Trying to better understand the dual / D-SNP market and would appreciate any insights!​
‍

– Christine Kim | via #buildersask​

‍

Thread Summary: The group offers up several data resources that provide information on Medicaid status of MA plans' members - unsurprisingly, the degree of visibility varies by state.

‍

Top Response:

‍

Prince Baawuah: Hello @Christine Kim - have you checked your plan’s Monthly Membership Reports (MMR) from CMS? Check out the Plan Communication User Guide (PCUG) Version 17.1 Section 6.6.1 for more details.

​

Also in the PCUG, there is a file/report called the β€œMedicare Advantage Medicaid Status Data File” with this description: β€œCMS will send a monthly report to Plans that provides the monthly dual statuses and corresponding dual status codes for their beneficiaries who are full or partial duals. Plans will receive a Medicare Advantage Medicaid Status data file to assist in predicting future revenue impacts under the CMSHCC risk adjustment model, and to assist in benefit coordination. Each report will provide the most recent Medicaid information on plan enrollees, back to the beginning of the payment.” (MAPD Plan Communication User Guide Version v17.1, pg. 7-139).

​

I hope these help! Let me know if you need to chat more

‍

​Check out the full HTN Slack convo here!​

‍

2. Patient testimonial no-nos

‍

Q: How do people here go about getting patient testimonials (anonymized quotes with attribution like "John, California" or "JS, California"). The APA/NASW/ACA codes of ethics frown upon it. Yet I see that pretty much every behavioral heath care delivery company has patient testimonials....

​

Psychologists do not solicit testimonials from current therapy clients/patients or other persons who because of their particular circumstances are vulnerable to undue influence.

‍

Ethical Principles of Psychologists and Code of Conduct, 2010, 5.05

‍

Social workers should not engage in solicitation of testimonial endorsements (including solicitation of consent to use a client’s prior statement as a testimonial endorsement) from current clients or from other people who, because of their particular circumstances, are vulnerable to undue influence.

NASW Code of Ethics, 2017, 4.07.b

​

Counselors who use testimonials do not solicit them from current clients nor former clients nor any other persons who may be vulnerable to undue influence.

‍

ACA Code of Ethics, 2014, C.3.b​
‍

– Gabe Strauss | via #topic-behavioral-health​

‍

Thread Summary: HTNers discuss collection of patient testimonials, with one clinician perspective explaining the risks of using patient testimonials for marketing and advising against the use of them at all.

‍

Top Response:

‍

Owen Muir M.D.: Testimonials from patients are a target on your back--in NYS for example it's a state level medical practice crime, like jail criminal crime. More importantly across states, it's the kind of thing that leaves you wide open to med mal lawsuits. Testimonials from colleagues, etc. fine.

​

I think the issue is that testimonials are weaponize able against PCs without whom the MSO is worthless.
​

So avoid problems by never marketing with patient testimonials.

​

Not theoretical -- happening now in nyc [link]

​

[Other resources here, here, and here]

​

The upside is low. The downside is catastrophic. There are other ways to market and manage downside risk. But as someone who has had a stalker patient family for years--they will come after doctors. Never open up additional avenues of attack. It's bad risk management.
​

Please don't use patient testimonials...even if it might not be a huge problem the reliance on them is not healthy. "Come up with other ways of marketing, this is too high risk."

‍

​Check out the full HTN Slack convo here!​

‍

3. Data privacy & security in LLM-powered apps

‍

Q: User umitmertcakmak posted the following question on twitter:

‍

"What are the best practices for handling data privacy and security in LLM-powered applications?"

​

As I think about it for HIPAA compliance (and responded on twitter), is there anything actually different about it compared to using any other service or API? At the end of the day, you still want: patient data encryption, strong security protocols, employee training, P&Ps, and so on; and with respect to the LLM, you either want to keep patient data within your own secure environment (say with a hosted model) OR have a BAA with an external commercial model (like OpenAI) if you send it to them.

​

Is this a naΓ―ve take? Is there anything specifically different about it because it's GenAI?

‍

@Michael Herrick, any thoughts?​
‍

– Nitin Karandikar | via #topic-ai-ml​

‍

Thread Summary: The AI-ML crew chats about HIPAA compliance in relation to generative AI and LLMs, highlighting several key risks and technical limitations to consider.

‍

Top Response:

‍

Cody Crumrine: I think you're spot on here. For the most part we can think of LLM providers like any other third party data processor in regards to risks & protections that should be in place. Existing standards will probably evolve to address them more explicitly as a sub class.

​

Michael is right that when you get into training or fine-tuning models you do need to be very careful about the data you use. Predicting how an individual piece of training data will affect the model's output is incredibly difficult. As a rule of thumb: only train a model with data you're okay with any user of the model having access too.

​

That doesn't mean you can't build LLM powered applications that work with PHI - just that you should avoid using it to train.

​

‍Check out the full HTN Slack convo here!​

‍

4. Patient data review timer platforms

‍

Q: Hi all! Our commercial team really wants us to be able to have a timer on our diabetes data platform to calculate time spent on patient data review in order for health system partners to be able to justify RPM code billing.

​

1. Anyone know of a partnership that would be able to allow us to offer the option

‍

2. Alternatively, any product solution/widget we could integrate (quickly)?
‍

​– Kelly Watson | via #buildersask​

‍

Thread Summary: The group offers up several potential RTM and CCM/CoCM solutions, including HumHealth, MD Revolution, HealthViewX, and more.

‍

Top Response:

‍

Christopher Young: we've looked at a number of solutions on timing (first for RTM and then again for CCM/CoCM) and we never found a great fit that was easy to integrate and fit our use case well enough. A lot of the tools are full-suite solutions so they don't give point options on just a timer. For RTM we made a build decision on a timer, with automatic timing on some of the most common actions.

​

If helpful here is a list of all the vendors we reviewed:

‍

​https://www.healthviewx.com/healthcare-orchestration-platform-engine/#​

​https://mdrevolution.com/​

​https://www.humhealth.com/#pricing​

​https://www.healthrecoverysolutions.com/​

​https://www.prevounce.com/​

​https://www.thoroughcare.net/​

​

Of note that it was for RTM (much newer and limited adoption compared to RPM) but we got push back back on self-recorded timing and documenting in our platform. Mostly that it was extra work, dual documentation, etc.

​

‍Check out the full HTN Slack convo here!​

‍

5. Health system EMR finders

‍

Q: Is there a database available that shows which EMR a given health system operates on?​
‍

– Rachel Menon, PA-C | via #buildersask​

‍

Thread Summary: The group shares a handful of data resources to identify which EMR various health systems operate on.

‍

Top Response:

‍

Nick Neral: Definitive Healthcare

​

You can also be scrappy and just go to health systems websites and they normally give it away via their patient portal url or other tools.

​

Epic gives away all their customers at this URL. If you click on the state, you’ll see all the orgs where you can log into mychart. [Link]

‍

Check out the full HTN Slack convo here!​

‍

❓UNANSWERED ASK

‍

Here we highlight a question from the Slack that needs some additional community insights - if you have a helpful thought, jump in below!

‍

Q: Anyone know a partner for IV therapy that has a broad set of infusion and injection options? All the ones we work with can’t support all the options we need. We need iron, B12, Glutathione to start with. We previously worked with IVX and SAGE.

‍

– Luka Ivicevic | via #buildersask​
​
​

‍⭐Jump in with some insights here!​
​

πŸ€–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!

‍

‍

✨HIDDEN GEMS

‍

Here we highlight helpful resources from across the community:

‍

  1. ​Early thoughts on proposed Medicare Physician Fee Schedule changes: CMS proposing further MSSP rule changes aiming to grow the program via Deana Bell – Milliman published some thoughts on ACOs and participating in MSSP.
  2. ​Part 2: Should you open new clinics to fuel growth? by Zach Miller – The continuation of an interesting thought experiment walking through various considerations and financials projections of a cash-pay healthcare services business organic growth strategy.
  3. ​Mental Health Parity: What do Shaggy and Big Insurance Have in Common by Owen Muir M.D. – Part 2 of a multi-part series on mental health parity, the MPHAEA enforcement, and more.
  4. ​Layoffs.fyi - Tech Layoff Tracked and Startup Layoff Lists via Margarette Jung – A comprehensive and real time tech layoff tracker.