Wednesday, June 16, 2010

More on the Success of Blue Cross Blue Shield of Michigan's Success With the Patient Centered Medical Home (PCMH)

The Disease Management Care Blog heard back from the folks at Blue Cross Blue Shield of Michigan (BCBSM) about their positive experience with Patient Centered Medical Homes (PCMHs). As outlined in this prior DMCB post, it may be that this Blues plan has finally found the PCMH Holy Grail: genuine savings in a commercial real world, i.e., non-academic, non-integrated delivery and non-government insurance setting.

Questions that the DMCB asked and the gently edited BCBSM responses are below. The DMCB, because it cannot help itself, has some additional comments in italics.

Question: It isn’t clear how the medical homes were certified by BCBSM: did you do onsite visits, use NCQA criteria, or simply rely on attestation?

"BCBSM works with more than 100 physician organizations across the state on the PCMH program. Physician organizations nominate practices from within their organizations for designation. Physician organizations work with their practices to self-report data on PCMH capabilities into BCBSM. BCBSM also reviews quality data on key metrics, such as generic dispensing rate, ED visits for non-emergent conditions, evidence based care, and imaging use. BCBSM then conducts a site visit at a sampling of nominated practices to validate the data and to see the PCMH capabilities in place.

Designation decisions are made according to how much progress the practice has made in implementing the PCMH capabilities, and on quality and use data. Practices are analyzed and scored using a weighted measurement model."

This tells the DMCB that cooperation between a regional insurer and physician groups can be an important asset in promoting the PCMH. Imagine that: real partnerships between and insurer and physicians. What's more, the PCMH is not operating in a vacuum: there's a link with pay for performance and managing the pharmacy benefit.

Question: The PCMH certification you use itself seems pretty generic. What was the secret sauce(s)?

"Designations are conducted annually, and each designation lasts for one year. Designation is based on 50% progress on adopting PCMH capabilities (124 capabilities organized around 12 domains of function) and 50% quality and utilization performance.

Our “secret sauce” is the voluntary engagement of the physicians themselves. We developed the PCMH model together with the physicians. We are not dictating that primary care physicians become PCMH-designated. There are 5,800 physicians in Michigan who are eagerly and voluntarily choosing to transform their practices to earn designation."

This tells the DMCB that BCBSM didn't assume that all of the State's primary care sites would automatically want to become PCMHs. Rather, the model was jointly developed and remains voluntary. Some physicians are apparently not turning their clinics into PCMHs and BCBSM is not only fine with that, it has enough of a critical mass to work around it.

Question: Do the savings in your press release match your administrative costs as well as any P4P/gainshare or other economic incentives to the PCP? In other words, did you bend the total cost curve?

"The data that we shared on savings should be considered “early findings.” A complete cost analysis will be completed in the fall. However, our early findings indicate that the PCMH model is helping to avoid costs, such as ER visits, hospital admissions and radiology usage. BCBSM’s “spend” on the extra 10% payment to PCMH doctors in 2009 was approximately $8 million. These are dollars that were reallocated from other areas of the annual fee schedule. In other words, no extra or additional dollars were expended. PCMH doctors earned a higher fee, while some specialists earned slightly less.

BCBSM worked in partnership with its physician organizations to help provide administrative support to physician practices. As part of our Physician Group Incentive Program, BCBSM allocated approximately $30 million in 2009 from that reward pool to go toward PCMH capabilities and performance. Those reward dollars go to the physician organizations. The physician organizations determine how they wish to use those dollars. Many have used the funds to help pay for care management nurses, disease registries, electronic medical records, and physician training."

Let's be realists: unlike the Feds, regional insurers like BCBSM cannot print money and it has to come out of somewhere. It appears to the DMCB that money that would have been used to pay for year-to-year trend was used to support the PCMH. They apparently did not assume that "savings" this year would pay for next year.

Question: The numbers in your press release are percents. Any preliminary dollar numbers you can share?

We do plan to share dollar numbers later in the year, after the cost analysis is complete. BCBSM actually began developing the PCMH program with our partner physician organizations in 2004, and in 2009 we left the “pilot” phase and launched the full program. One of the first initiatives we launched with our physician organization partners was the generic dispensing effort – we estimate that since this program began in 2001, generic use has saved more than $700 million.

We are working with researchers from the University of Michigan School of Public Health on a comprehensive evaluation of the PGIP program, which includes the PCMH program. Funding for the evaluation has been received from Robert Wood Johnson Foundation, Commonwealth Fund and Agency for Healthcare Research and Quality (AHRQ). We have some articles in development for submission to health care publications and industry journals.

The DMCB agrees that if you are in the health care industry, you have an obligation to advance the science by sharing your knowledge in a transparent fashion that allows peers to review your results. That may involve traditional publications but can also involve other venues. However it is done, other workers in the PCMH field need to learn what works and why.
The DMCB hopes we learn more soon.


Anonymous said...

The reason for BCBSM's lack of report about savings is due to the fact that the model may be good for physicians but does not provide cost savings. Millions have been spent on providing incentives to physicians without showing any returns. The program has been reporting the same savings from generic drugs year after year. Incentives have not been passed on to patients by the physician groups or to their staff, except on an exception basis. How long should the model be allowed to continue by the BCBSM board without "proof of concept?" The physician groups do not object since they are still earning additional funds. There is little accountability to the BCBSM members, except self-promotion of the program. Has the state ever audited the program? Self-report on success is not a credible measure.

Jaan Sidorov said...

Anonymous makes some very good points. "Savings" consist of something that didn't happen, which is very hard to mention. In addition, assuming savings can be translated into hard cash somewhere in the system, it's not clear just who gets the cash PLUS it doesn't appear to be the patients/BCBS subscribers.

On the other hand, I read that BCBS is relying on the University of MI to analyze the results and there is a commitment to share the data in a peer reviewed forum.

As for the State auditing the program, I don't know if a typical State bureaucreacy has the expertise to do it well. On the other hand, the State does have regulatory oversight and they haven't stepped in and said the program is the rip off that's being implied in Anon's comments.

Last but not least, this is a nonprofit company, which should help give comfort to the idea that a savings > expense (which is, admittedly, where the physicians are) will ultimately work to slow down premium rate increases for purchasers and subscrbers. THAT scenario is more reasonable than a Medicare scenario, where the accountability is so diffuse and the time horizon is limitless.

coda: DMAA has an excellent methodology to measure savings which is very credible. The challenge is to make it transparent AND publish this, the sooner the better.