Tuesday, May 26, 2015

Captivated by captives

One of the educational benefits of being CEO of one of the Harvard-affiliated teaching hospitals was that I also served on the board of our captive insurance company, CRICO, and its subsidiary, the Risk Management Foundation.  Like many people, I had not previously understood the ins and outs of risk retention groups, medical malpractice insurance, and risk management, and so this was a great experience.

As noted on its website:

In 1976, the crisis of insurance availability led the Harvard Medical Institutions to form a “captive” medical professional liability insurer, CRICO, an offshore insurance company operating and domiciled in the Cayman Islands. Harvard's model of creating an offshore captive—an insurer owned by the facility or health care system it exclusively insures—was later replicated by other health systems across the U.S. 

Risk Management Foundation of the Harvard Medical Institutions, Inc. (RMF) was established in 1979 to apply a data-driven approach to claims management and patient safety on behalf of our insured organizations. 

This has been a very successful organization on many fronts, not only in defending unwarranted malpractice asserts, but, more importantly, looking at the risk profile of the affiliated institutions and designing training programs for clinicians to help provide safer care.  Those programs are worth a look and might be informative to other health systems.

But my topic today is not CRICO.  It's about the panoply of captive insurance companies that have been set up across the country, and it's about probing what we might learn about safety and quality improvement from their results.

This paper by ProMutual Group contains some basics about the business model that must be followed by captives.  To a great extent, they must mimic other insurers:  Without a proper actuarial base, good financial management, and good claims management, they will fail.

Fundamentally, from an economic and operational perspective, captives are similar to traditional insurance companies, to the extent that a captive collects premiums from selected individuals and entities, invests its assets, provides services, and, eventually, pays claims. Any surplus or deficiency is either paid to or paid by the owners of the captive insurer, which are, for the most part, the insureds. 

Where the captive differs from a commercial insurer is that the owners--because they are also the insured--should have a strategic interest in reducing the likelihood of claims over time.  Indeed, as noted above, the RMF function of CRICO was focused on that purpose.

This got me wondering.  Could we measure the success of hospitals and physician groups in improving the safety of care by the degree to which malpractice asserts for those organizations declined over time? After all, everything else being equal, if safety has improved, the chance of doing harm and being sued should drop.

Further, to the extent a captive is successful at helping its members carry out risk management programs, shouldn't actuarial assumptions be modified, and shouldn't premiums go down?  Also, since previous premiums were based on old actuarial assumptions, shouldn't an effective captive generate a surplus in earnings that should be returned to its members?

(By the way, if there is a surplus to be returned, the question of who should get it--the hospitals, the doctors in general, or certain specialists in particular--should prompt a very interesting negotiation!  We'll leave that topic for another day.)

So, here's my challenge to those of you in hospitals and physician groups served by a captive medical malpactice insurance company:  What do the financial characteristics of your captive tell you about the change in safety performance of your hospital or your physician group?  Or turning the question around, if you know that the safety of your hospital or your physician group has improved, how has your captive dealt with that improvement?  Have premiums gone down? Have surpluses been returned?

Since your captive is owned solely for the benefit of, and is governed by, people from your organizations, you have a right to know the answers to these questions.  Maybe it's time to ask.

3 comments:

Barry Carol said...

Two issues that could complicate the analysis you are asking for are (1) any changes, for better or worse, in the local litigation environment over the relevant time period and (2) any changes in patient expectations that might make them more or less likely to sue over a given outcome at different points in time. Separately, if the hospital later adopted a policy like “sorry works” where it admits a mistake and apologizes to the patient and provides appropriate compensation, there could be more payouts than before but each one could be smaller on average, in part, because the patient appreciated the apology and also because the associated legal costs were a lot lower.

Paul Levy said...

True, Barry, that why I cleverly said "everything else being equal." (The economist's savior and catch-all!)

Seriously, over time, those factors could be expected to even out, but you are of course right about their confounding effect.

Bob said...


Your point is actually a subtle one on several fronts. Importantly, since stand alone insurance companies make a percentage of their total premiums, it is in their interest for premiums to rise, reflecting increased risk, moderated only by competition, but there can also be a tacit understanding one would think among the insurers not to do too much with intervention. Also, a captive can be much more active within a company than an external company. Here is one instance where vertical integration is quite a good thing.