The ongoing saga of savings estimates for the Community Care of North Carolina (CCNC) patient-centered medical home (PCMH) is finally over. The verdict: no savings. Because the scale and visibility of the CCNC experiment are unparalleled in the Medicaid sector today, it is important that the right policy and delivery system lessons be learned from this dispositive conclusion.
Lesson 1: Enhancements in access do not necessarily create cost reductions, at least in Medicaid.
CCNC is by all accounts an excellent program from the patient’s perspective. Indeed, if I were a Medicaid recipient, I would want to live in North Carolina. The leadership of CCNC is passionate about the program and constantly strives to improve it. However, as was amply observed by J.D. Kleinke on this very blog last week, Medicaid recipients have many lifestyle and economic issues that even the best-intentioned and best-incentivized doctors will never be able to systematically address.
Lesson 2: Perhaps it is time to create an ER co-pay for Medicaid recipients that has more than one digit to the left of the decimal point.
Even as ER co-pays for commercial insurers have soared in the last decade, Medicaid ER co-pays remain virtually non-existent. CCNC created excellent reasons to use primary care but was not permitted to re-price the ER to economically encourage use of primary care. Many Medicaid recipients overuse the ER in part because it is basically free. For the CCNC experiment to truly have a chance to reduce ER visits now that they have created a worthy substitute with their PCMH, it’s only fair to them (and to taxpayers) to reconfigure the financial incentives so that people use their worthy substitute … and then re-measure savings.
Lesson 3: Never assume that if an actuarial consultant says something saved money, they must be right because they are actuaries. Use your own judgment to determine if a claim of savings is plausible.
Milliman is perhaps the best-known name in actuarial consulting and yet they were off by a mile in this case. (They were probably equally off in Illinois, in which they found savings of $180 million in the first six monthsof the Medicaid patient-centered medical home, before many people even had a chance to go to the doctor.) Their failures in their North Carolina analysis were legion, as noted in the article. They “overlooked” data that contradicted their conclusions, they forgot to mention what year the study started, and they came up with conclusions that were quite literally impossible not just one way, but two different independent ways. And in the process of attempting to defend their report, they replaced their only demonstrably true statement — that PCMH savings accrue only in inpatient/ER, while most other costs increase – with the opposite and totally unsupported claim that the majority of savings are somehow found outside the hospital.
Milliman isn’t the only consulting firm not to understand how to measure population-based outcomes. It’s a new field that many if not most consulting firms don’t understand, and as a result routinely report often hilariously impossible findings, as described in my book. The bottom line: if a consultant’s report sounds too good to be true, it probably is, so trust your own judgment when reading it.
Lesson 4: Don’t stop experimenting
Even though savings have been elusive, North Carolina Medicaid should be applauded for trying, and applauded more for persisting. The states are our laboratories for health policy, and if a lab is shut down because we don’t like the result of their experiment, it will discourage other states from attempting novel approaches. Eventually something will work but along the way one must recall Jacob Riis’s observation that “when nothing seems to help, I go and look at a stonecutter hammering away at his rock perhaps a hundred times. On the hundred-and-first time, it will split in two, and I know it was not that blow that did it, but all that had gone before.”
Al Lewis, widely credited with inventing disease management, is author of the forthcoming Why Nobody Believes the Numbers (John Wiley & Sons, June 2012), the introduction to which may now be downloaded gratis from www.dismgmt.com. He also runs the popular course and certification program for Critical Outcomes Report Analysishttp://www.dismgmt.com/certs/cora/self-study and was named the “leading authority on care management outcomes measurement” by the 9th Annual Report on the Disease Management and Wellness Industries (Health Industries Research Co., 2010).
My own politics are severely moderate (believe it or not, that term originated with me) so I usually get hollered at by everybody on this rather politicized forum, so it was nice to see your comments and, yes, sometimes (though not on this particular post) the anonymous comments can be unprofessional.
My niche is actually doing analysis correctly. In my book I take down private-sector and public-sector programs equally.
I agree with what you’re saying, and I also really appreciate your activity in the comments. The analysis they did sounds bush league at best (though I’m taking your peer-reviewed published word for it).
Having worked in managed care I’m aware that IP Admissions are a key indicator, and the easiest indicator, but I don’t think they tell the whole story.
Anyway, I’m a liberal, I love the spirit behind programs like these, I would like to see more like them, and I am wary of arguments that if something in Medicaid costs more it’s necessarily bad. Now, costing more with the same clinical outcomes, that’s definitely not good. I think smoking rates and obesity at least get to the heart of what Medical Homes are supposed to be about, so I definitely agree that those are good clinical measure to choose, I was just curious regarding what else you looked at.
Thanks again for your responses and your professionalism.
a few things. first, I did look at two published measures of outcomes — obesity and smoking rates In both cases NC and SC (control) trended similarly, meaning that NC Medicaid members did not get healthier relative to benchmark.
Second, what you’re saying about other QOL measures is possible and I hope someone looks at it, and I hope you are right.
Third, generally speaking, admissions for ambulatory-care-sensitive-conditions are considered a bad thing, something that one tries to avoid. But that wasn’t really the point of the article. The point was that Milliman was incredibly wrong about the savings, and should probably refund their money in NC taxpayers. Rookie mistakes are only excusable if you are drawing a rookie paycheck. They got $500,000, which is ten times what I charge, to do it right.
if MIlliman refunded their money, the state could use it to look into the things you suggest, since at this point the cost savings issue is settled.
I saw from your article that Admissions and Days per thousand were essentially the same for the two groups, but it seems like those are crude measures to evaluate whether people in this program are healthier or not. Granted, they’re often used because they can be measured fairly cheaply, which makes them better than things that are either too costly to measure or would have required an actual experiment to measure.
But I’d be more curious to learn about attendance at school for those in this program and those not, or attendance at work, and quality of life measures, mortality scores, etc.
IP Admissions per thousand could just be a result of general increased access to care and could be a good thing.
The entire analysis is based upon outcomes, which is how it differed from Milliman’s. Their actuaries did what actuaries always do–make a bunch of assumptions and run models — for $500,000, you’d think they would have bothered to ask the question, what happened to outcomes? Were enhanced primary care access and other PCMH tools effective at reducing the number of members who “crash” and end up in the hospital?
It turns out that there was essentially no change in that number, overall and for ambulatory care-sensitive medical conditions, and for every high-volme ambulatory-care sensitive medical condition. This conclusion was true both on an absolute basis and relative to bordering states.
The entire analysis is in the book, Why Nobody Believes the Numbers.
Once again, though, I want to congratulate Community Care of North Carolina for their efforts. And I think they would be more successful if financial incentives were aligned with care availability, a policy which is out of their control.
Is there any data on clinical outcomes? It seems like all this is around the cost. At this point the main conclusion seems to be that patients in a conventional setting under utilize due to lack of access.
Appreciated the article knowing how so many are quick to tout the “proven value” of the North Carolina Model. Have also just finished reading Al Lewis’ new book. Just ordered 5 additional copies to send to send to Medicaid disease management staffers around the country who truly need to improve their discernment skills around what constitutes an effective care management/disease management program for the populations they serve.
Al and MD,
People go to the ED because they have no other place to go or because they are scared and think the hospital is the best place to go.
If you want folks to go somewhere else, than make sure there is such place and that the place is open “after hours”. What is needed is infrastructure and education, not financial barriers.
$20 is a lot of money for too many Americans, and I don’t think this is the appropriate modality to ensure that “they learn the first thing about taking care of anything”. And it’s not “they” – it’s us.
Are you talking about $20 being too much for someone on medicaid. Because I for one have seen medicaid pts on iphones and have ipads,tattoos,jewelry,,nails done,hair done and dresses to the nines. But they can’t contribute to their healthcare? Come on, we all know better than that. They are not made to work for anything,and there are no consequences. So while we are worried to death about how we are going to pay our 20% copay they have no worries. We are used as a pmd ,.when asked the question have you called your doctor the answer is most often…no.
Try this version. Instead of just raising the co-pay on ER visits, perhaps a state, such as North Carolina, that has achieved certain thresholds for access, should be allowed to do the following: allow members to choose between the standard near-zero co-pay for ER and standard drug co-pays (also near zero, but drugs are boght much more), and a $20 ER co-pay but no co-pays for generics. Two different benefits plans.
This would be the member option but on an opt-out basis following contact. Most people will go along with it, since it is opt-out.
This would encourage compliance while also encouraging people to use their PCP…and save the state a ton of money on ER. (Yes, some have extra capacity but they still send the states big bills.)
I don’t see how making emergency care unaffordable solves any problems.
Using MD as HELL’s logic, make PCMH free and see if it works that way.
Besides hospitals seem to have tons of unused ED capacity. Otherwise why would they advertise on every highway and byway billboard, or encourage you to schedule ED visits from the comfort of your home? Or is that capacity not really for everybody?
It does not make it unaffordable. It makes it valuable. Right now the Medicaid bunch does not value healthcare. It is free and limitless. Why should they worry about not wasting it? Why should they learn the first thing about taking care of anything? They can just call EMS and come to the ER.
Yes, make primary care free…no co-pay. As for the billboards, they are bait and switch.
Hey…we agree on something… 🙂
We have a BINGO!!. I know nothing about CCNC, but as a North Carolina Emergency Physician I can affirm all the above assertions about NC Medicaid as it relates to the ER.
Nothing will shift the cost curve when the alternative for the patient remains totally free.
There must be a Medicaid co-pay. It must be collected from the beneficiary by the program itself . Do not ask the hospital to bill and rebill and send them to collections. Failure to pay the co=pay to the program should makke them ineligible for further benefits.