In this post:
"I, for one, welcome our new computer overlords" - Ken Jennings, Jeopardy! legend
They lost. In 2011 Ken Jennings and Brad Rutter, who are the best Jeopardy! contestants of all time (with combined winnings of over $5 million) played against a computer… and lost . By winning, IBM's "Watson" became a Jeopardy! champion and the start of a wild plan to change how cancer is treated.
IBM does not make their $7.3 billion in annual revenue by winning game shows. They needed a commercial application for this machine that could be trained on clues (symptoms) that are then matched to answers (diagnosis). Oncology was selected, partnership were formed, and Watson was connected to the internet so that it could consume publications like Nature and the British Medical Journal. IBM made plans for Watson to pass the medical board exams .
But as the future looked brightest for Watson Health, a widening gap between the marketing promises and the reality of it's capability were giving IBM's credibility a black eye , the new Watson Health team at IBM wasn't working well together- and cultural differences led to someone mounting computer monitors in the middle of the whiteboards… in every conference room.
2021 has been the year of the technology-cloaked health insurance startup going public with an IPO (traditional or SPAC). Leading the field are Oscar ($9.5BN valuation), Clover ($3.7BN valuation), and Alignment ($3.6BN valuation).
For comparison, Humana has a market cap of $35BN and covers 20MM lives- $1,750 per life. OCA together are valued at $16.8BN and cover 527,870 lives- an eye popping $31,867 per life, 18x more than Humana. That large multiple? It's because OCA brand themselves more as technology first and insurance business. In their own words:
My expectation was that OCA would have been the leaders in coming to market with their patient access APIs. I was… disappointed. None of them have developer sandboxes, or even documentation. I made attempts to contact Oscar and Clover by phone and email; maybe I just wasn't Googling right. Neither was responsive by email and phone calls were unproductive.
(Update 09-April-2021) A friend contacted me and shared that Oscar is actively building their APIs with plans to meet the July deadline, likely using a managed solution.
I also found it interesting that all of these companies filed S-1s in 2021, after the CMS rule was released, each one includes a lengthy discussion of regulatory risks, but not a single mention of patient access, interoperability, or CMS 9115-F.
Because we can't review the APIs, this post will be about the penalty for noncompliance with the CMS 9115-F Interoperability and Patient Access rule instead.
OCA primarily offer Medicare Advantage plans. Oscar is the exception, covering 400,000 people on the Obamacare exchanges. The table below summarizes insurance plan types that are in-scope for the CMS 9115-F Patient Access and Interoperability rule, and how many members OCA have under each:
The CMS rule doesn't explicitly list a penalty for noncompliance. The most clarity we get is from the "Applicability and Timing" section:
MA organizations and QHP issuers on the FFEs generally are subject to rules regarding bid and application submissions to CMS in advance of the coverage period ... to ensure that these requirements for MA organizations and QHP issuers on the FFEs are enforceable and reflected in the bids
To me, this says that if the plan submits a bid, it's assumed that compliance with this rule is factored into that bid. If the plan is noncompliant, then they falsified their bid. Penalties for falsified bids are:
So Oscar would be looking at up to ~$40MM per day in possible fines for their ACA Exchange plans. Clover and Alignment would each be looking at $750MM in fines. These are very steep penalties and I'm sure the regulators will be gracious at first. Enforcement starts 1-July-2021.
Note: I am not a lawyer. If you have better insights on how the rules will be enforced, please email me.
It was all made up
“It was all made up,” one former employee said of the marketing without robust data behind it. “They were hellbent on putting [advertisements] out on health care. But we didn’t have the clinical proof or evidence to put anything out there that a clinician or oncologist would believe. It was a constant struggle.”
In 2015 IBM set a goal of generating $5 billion in revenue from Watson Health by 2020. To get there IBM purchased Explorys, Phytel, and Truven Health Analytics for a combined $4 billion dollars. These acquisitions would supply the training data needed by an AI like Watson Health.
IBM struggled to integrate both the culture and the data from their acquisitions. Former employees said that IBM failed to fully comprehend the differences between its traditional tech business lines and the complex world of health care. As the Watson Health team failed to collaborate waves of lays-offs and departing leaders portended the economic reality of Watson Health. In 2020 revenues were much closer to $1 billion than the $5 billion target. Watson Health operated like a billion dollar research project, that was also burdened with the financial expectations of a public company.
A year later, 2021 may be the year that reality catches up with the hype of Watson Health. IBM is rumored to be looking for a private equity buyer for the business unit.
It's important to know what you are, even if you're just a computer that wins at Jeopardy!
Do you know someone with an Aetna Medicare Advantage plan? Please send them to the HealthSouse home page. We may also be able to save your friend some money.
Still here? Thanks for reading! You seem like a nice person, who should send me an email: firstname.lastname@example.org. Encouragement and criticism are both welcome!