How to make +30bn Dollar (fast)!
How to make +30bn Dollar (fast)!
Don't sell drugs, buy health care claims!
In July, MSP Recovery went public via a SPAC on a $32.6 billion valuation - the second largest valuation ever for a SPAC deal. The company was founded in 2014 by class action lawyer John Ruiz, who is estimated to retain more than 70% of the shares, making him one of the richest people in the world. It took Mark Zuckerberg longer to reach that level of net worth. Quite a run...
MSP Recovery's business model is based on a certain tweak in the US health care system: If a patient has dual eligibility, i.e. if they are covered by more than one insurance, the insurers first have to figure out who is responsible to pay. This process - also called "Coordination of Benefits" - often contains errors and often government programs like Medicare and Medicaid have to foot the bill, even though another payer would be responsible. MSP Recovery builds on this problem, and is buying the rights from government plan health insurers, to recovery payments that the insurer made.
Medicare Advantage Organizations, i.e. privately administered health plans that are sponsored by medicare, are covered under the Medicare Secondary Payer Act. This act (and a few related court rulings) provides the basis for MSP Recovery to buy claims from the insurance companies, identify which claims fall under the Medicare-Second-Payer category and then recover payment from the payer who should be responsible for it. They will then keep a cut of the recovered value. Even better: often they will be able to recover a multiple of the claim value from the other payer for these reasons:
The primary payer has to pay the market rate (or contract rate) for the claim to the provider and not the Medicare rate, which often is lower.
The payer has to pay interest on the not-paid claim and a late fee which can be up to $1000/ day
The payer has to pay double damages, if they have not reimbursed Medicare in good faith (i.e. withheld payments).
To identify these claims where a second payer is responsible, MSP recovery is using "advanced data technology" to determine for which claims a primary payer might exist, according to their investor pitch.
I find this a fascinating business model, as this company potentially found a way to align incentives that will force health insurances to improve their coordination of benefits systems. This will eventually help the government to avoid paying for claims that a private insurance should have paid. That being said: so far MSP Recovery has recovered $0 of claims... We will see how well their hypothesis will play out.
In this post, I want to look a bit deeper into the coordination of benefit problem. It's hard for me to understand why on the one hand MSP Recovery claims that about 5% of Medicare claims can be recovered from another payer, but on the other hand there are multiple billion dollar market cap vendors that are helping with coordination of benefits for providers and payers.
Having two health insurance? Why would you?
First of all, let me quickly review why there even can be dual eligibility. Here are some common situations, where Medicare is a secondary payer:
Claims from patients that are Medicare eligible (+65 years old) but still covered by an employer health plan (their own or from their spouse)
Claims that come from a job related illness that is covered under workers-compensation
Claims related to an accident that is covered by liability insurance, mostly car accidents
The Medicare Secondary Payer Act states that Medicare will not care in certain situations, if the health care should be covered by another payer (primary payer). In the cases above, the claims need to be first send to the primary payer like the group health plan or the auto liability insurance, before Medicare will pay.
Coordination of Benefits
This coordination of benefits (COB) is done two times in the claims process: first by the provider, who needs to figure out who to send the claim to. The primary source of information here is the patient (via the insurance card). Medicare providers are also required to have the patient fill out a questionnaire to identify whether there are any possible primary payers. More paperwork at the doctor to fill out when you turn 65 - wohoo. Actually, providers have an incentive to get this right, because in general private plans or liability plans pay a higher rate for the service than Medicare.
The second coordination of benefits happens at the health plan, who will check whether the member is eligible and whether any other insurance could be the primary payer. There are several incumbent software vendors providing solutions for this, like Cotiviti, Gainwell/ HMS and Multiplan. Health plans also have an obvious incentive to identify the payer, as they would not have to pay the claim.
So if both payers & health plans should actively try to identify primary payers, why does a company like MSP Recovery even have a business model? I don't have an answer to this question, but I have a few guesses!
Why does coordination of benefits fail
It's a problem I often come across in my health care research: money is wasted left and right, and at the same time there are a lot of vendors tackling this problem, but they don't seem to address the problem all the way. Here are some guesses why this is the case for coordination of benefits:
Lack of Liability Insurance crosschecks: In their investor deck MSP recovery indicates that a typical recoverable claim comes from a patient that was involved in a car accident. I guess identifying whether a patient was covered by third-party liability insurance is notoriously difficult to determine by payers and providers. If you're getting brought to the hospital after a crash and you have a Medicare card with you, providers probably don't try to investigate the counter party of the accident and just bill Medicare. Insurers could crosscheck the case with the collision database - but they apparently don't put enough effort into these cases.
Long-tail death: Fixing the first 80% of a problem is often easy, fixing the last 20% is incomparably harder. This is probably especially true in health care administration as there are so many edge cases of medical conditions and cases as well as payer situations. Vendors might capture the easy 80/90/95%. However, it could still be that MSP Recovery found another chunk of claims that are systematically missed by the COB vendors.
Small Enterprise: The long tail also exists for the payers - even though there has been quite some consolidation in the health insurance industry, there are still many small health plans that maybe don't have the capabilities to conduct a coordination of benefits beyond simple crosschecking with an eligibility database. It's probably worth checking whether this hypothesis is true.
How could Coordination of Benefits be improved?
I could think of several general approaches on how to improve coordination of benefits:
Better data about member eligibility: The easiest way to avoid interference with workers compensation or group health plan coverage would be, if there exists a database that checks the enrollment status of a person to a certain health plan. Similar to a credit check, an insurance would check this database first to see whether there are any other plans. While it seems that this database is provided by incumbent coordination of benefits vendors, these vendors may not take into account coverage by liability or auto insurances.
Smart claim detection: There are certain patterns that can be detected in claims that have a high likelihood for dual coverage. A mix of good heuristics combined with machine learning models can probably get very good in detecting accident claims (→ check for liability insurance) or claims that could be caused by a work related illness. (→ check for workers comp). This detection could even be combined with a member interface, to cross check the findings with the patient. Giving incentives to patients to help determine the right coverage could be a way to go.
Cross correlation with external data: The detection strategies can be further enriched with data that is not on the claim - like patient questionnaires (there should be a better way to collect this information than paper forms), accident reports, court filings, employment records, etc.
Help Auto-Insurance to comply with MSP: If MSP recovery's business takes off, and they successfully start to recoup multiples of the payments from liability insurers, they will probably have to respond with better detection rules. They want to know whether their members are involved in a crash and whether they have to pay a medical bill to Medicare. In the end it will be cheaper to pay in the first place, than to wait for a recovery business to knock at their door.
I have done a quick search on startups tackling this space, but I could not find a whole lot. I came across Anomaly who is tackling the payment integrity space, but could not find out too much about them. If you know companies tackling this problem, let me know - I would love to learn about them.
All opinions in this article are my own and do not represent the opinions of my employer. None of the information above should be seen as investment advice.