Risk Adjustment is a process that has to occur for any insurance program where there are multiple insurers and guaranteed issue. It is needed to prevent or at least minimize the insurer’s incentives to get very good at cherry picking only healthy people while being ugly to sick people.
Risk adjustment is a process of sending extra money to insurers that have a sicker than average population. There are a couple of ways to do risk adjustment. The Exchanges, some Medicaid and some CHIP programs use revenue neutral risk adjustment where money comes from insurers with healthy covered populations and money then goes to insurers with sick populations. This is two sided risk adjustment. It can either be revenue neutral or non-neutral.
Medicare uses one sided risk adjustment. Medicare uses a disease category model (HCC) that ties certain diangosis and demographic factors to the base capitation rate Medicare Advantage plans receive bonus payments. Someone with diabetes is attributed a larger risk bonus than someone with a mild skin condition. CMS receives the data from their encounter process which is where all claims by Medicare Advantage insurers are sent for analysis. They have a scraper program that pulls the diagnosis codes from claims and match it against individuals. Each additional unique diagnosis on a claim that is medically supportable then receives a bonus. Medicare Fee for Service (FFS) does not have risk adjustment built into insurer payments as the insurer is the Federal Government.
As you can see, the incentive for individual Medicare Advantage insurers is to make their patients look as sick as possible to maximize their bonus. This can be done by making sure every possible supportable diagnosis is entered on a claim at least once a year. The NBER has an interesting paper that looked at how much this upcoding could cost the government:
Empirically, we examine how coding intensity in Medicare differs between the traditional fee-for-service option, in which coding incentives are weak, and Medicare Advantage, in which insurers receive diagnosis-based subsidies. Our estimates imply that enrollees in private Medicare Advantage plans generate 6% to 16% higher diagnosis-based risk scores than the same enrollees would generate under fee-for-service Medicare. Consistent with a principal-agent problem faced by insurers attempting to induce their providers to upcode, we find that coding intensity increases with the level of vertical integration between insurers and the physicians with whom they contract. Absent a coding inflation correction, our findings imply excess public payments to Medicare Advantage plans of around $10 billion annually.
This excessive upcoding is the source of most of the cuts to Medicare Advantage from PPACA as the goal of the cuts is to bring back the revenue and risk neutrality to Medicare Advantage programs and FFS.
Any risk adjustment program will have red queen race incentives unless it is very carefully designed and recalibrated against unadjusted/non-intervention populations every few years.
HinTN
Are there data yet to see if this has had success?
gvg
What is ” red queen race incentives”? Thank you
Richard Mayhew
@gvg: Got to run harder to stay in place
Or in this case, got to code harder to not lose any additional money as everyone else is coding hard… it is a nasty but stable equilibrium
Jim
It’s complicated, isn’t it? The GOP doesn’t understand this (probably choosing not to, as it doesn’t help their simplistic talking points). As a result, they’re running into real problems trying to figure out how to decimate the ACA without raising deficits.
kinda
Well timed, Richard. one of our providers got a call just last week from a medical director of a small MA plan asking to please make sure he was more rigorous in coding diabetic complications. This doctor’s been practicing for 12 years so he’s pretty thorough. He definitely thought there was an implication of improved reimbursement in the call, and now I guess we know.
Bruce Webb
Well I knew Medicare Advantage was a scam but had no idea they had a business model cleverly based on four words:
Recruit healthier, report sicker.
Good to know
muddy
I wonder if this is why my last printout from the cardiologist had me listed for several diseases I don’t have? For instance a medication I took years ago had the side effect of lowering the thyroid. So I took a (minimal) thyroid supplement. When I stopped taking the med, the levels were normal right away. No more supplement necessary.
Then in the printout years later it suddenly started saying I had Hashimoto’s Disease. Looked it up and said, No I don’t. There was something else that I can’t remember right this minute.
Richard mayhew
@Bruce Webb: the first two words are the business model of all casualty insurers
Deecarda
Every encounter our providers document in our EMR is accompanied by flashing icons to encourage the addition of HCC coding.
Dennis Byron
FYI, according to Modern Healthcare the authors have now retracted this claim and their new estimate of “excess” risk-related Part C cost is $2 billion annually, not $10 billion (or just over 1% of the total Part C cost). The authors’ estimate is now in line as a percent of total with estimates made by MedPAC concerning 2015 and the GAO concerning 2010.
I personally do not believe that the government should be at all involved in this picking of winners and losers based on total bureaucratic nuttiness. Give all of us on Medicare except low income beneficiaries the same exact amount of premium support. Let the insurance companies figure out how risk affects the price we pay. That is what insurance companies are supposed to do.