Navigating Medicare

Not so long ago, a healthy patient of mine in her early 70s came to see me. Mrs. Smith, as I will call her, burst into tears, telling me her fear that she was dying of kidney disease. She had written out a bunch of questions pertaining to her kidney health, and had already identified a world-renowned nephrologist with whom she wished to consult. 

I reviewed her chart and assured her that recent lab work was normal, containing no hint of a kidney problem. She affirmed that she felt entirely well and that no new symptoms had emerged since I last saw her. Why then had she come to the conclusion that her kidneys were failing? Because the medical system had inadvertently told her so. 

You Feel Great, but on Paper You Are Dying

Mrs. Smith was the victim of three separate problems that combined to trick her into believing she had a dread illness. 

First off, she was able to log onto my practice’s patient portal and see her medical record. This is a topic that I cover in another essay, but for the purposes of this story, what she saw there was the notation that she had “Stage 3 Chronic Kidney Disease.” 

The second problem that Mrs. Smith encountered was the fact that—by the nomenclature kidney specialists have devised—Stage 1 and Stage 2 Chronic Kidney Disease do not indicate any significant kidney impairment, but rather the risk for such. Hence, even a small drop in kidney function gets labelled as “Stage 3” disease—an understandably frightening label for the uninformed to receive. Moreover, virtually everyone loses some kidney function with age; by their 70s, most people will meet the criteria for “Stage 3 Chronic Kidney Disease,” even though few of these folks will go on to need dialysis or die of kidney failure. 

Mrs. Smith was victim to three separate problems, which implied she had a dread illness.

The third factor that led to her misunderstanding will be the topic of today’s essay.  Namely, the reason why the documentation of “Stage 3 Chronic Kidney Disease” was noted in her chart to begin with, despite it being of no clinical importance. In other words, I will do my best to explain in lay terms the phenomenon of “RAF” scoring, one of the true banes of existence for the modern American primary care doctor. 

The acronym “RAF” should conjure up images of British pilots fighting a gallant battle to defend their homeland from Nazi invasion during World War II. But without any irony, Medicare and the American health insurance industry have morphed this three letter acronym into something much more pedestrian. And while it is not a Hollywood blockbuster topic, you cannot truly appreciate the crisis in American medicine without knowing a little bit about “Risk Adjustment Factor” scoring. 

The Set Up

First, a bit of background. Historically, Medicare (the national health insurance that most Americans get upon retiring or becoming disabled) paid doctors a certain amount for each completed patient visit, with higher rates for more complex services rendered. Thus, a way for doctors to earn more money was to see more patients, bill for more complexity per visit, or both. There was no financial incentive for doctors to either keep costs down or to improve the actual quality of outcomes delivered to patients. 

Over the past decade, Medicare has increasingly looked to rectify this by augmenting the above with a system of bonuses tied to something called “population health management.” Doctors still get paid by the visit, as above, but those rates have been largely frozen, while population health management bonuses have grown. The idea is over time to tie more and more of a physician’s compensation to the earning of these bonuses. 

To understand the concept of population health management, imagine a hypothetical medical practice that cares for 10,000 Medicare patients. Only for simplicity’s sake, let’s assume that Medicare calculates the expected annual cost of caring for each patient to average $1 per person. Therefore, the expected annual cost for this population is $1 times 10,000 people, for a total cost of $10,000 per year. Under a population health management contract, this medical practice will receive financial bonuses at year’s end if the total cost of care for these 10,000 patients comes in at under $10,000. For example, if Medicare spends only $9,000 to care for these 10,000 patients during the year, then Medicare has saved $1,000. Under the terms of the contract, Medicare agrees to give some amount of that back to the medical group in the form of a bonus. 

The above paragraph is vital to understanding what is to follow, so let me restate it once more: Medicare expects to pay $10,000 to take care of a particular group of patients, but instead only pays $9,000 dollars, thus saving $1,000. Medicare then, for example, saves $500 and gives $500 back to the practice caring for these patients. This year-end bonus, plus the revenue earned on a day-to-day basis by billing for each individual patient visit, combines to form the total revenue for the medical practice. 

I have used the absurdly low value of $1 per patient per year just to make the math very simple, but in reality, costs are several thousand dollars per patient per year, and such contracts can thus produce tens of millions of dollars in savings to Medicare and bonuses to medical groups.

This is the basic idea of population health management, and its potential advantages to the system should be obvious. By contracting groups of doctors to care for an entire population (the 10,000 patients in the example above) and promising them a piece of any cost containment, the doctors become incentivized to keep costs down. Now, when a patient calls at 5:00 p.m. on a Friday with an urgent issue, there’s a financial incentive to stay late and see them, rather than referring them for a costly trip to the emergency room. When a cheap generic medication will have the same effect as an expensive new one, there is an incentive to prescribe the cheaper pill. And so on. 

In much the same way that bacteria evolve to resist antibiotics, doctors outwit Medicare regulations, leading to new regulations and an endless cycle of bureaucratic sprawl.

Indeed, the measures taken to contain costs can go well beyond the above. My practice, which is incentivized by contracts very similar to the hypothetical described above, has created a weekend sick clinic for our patients to reduce emergency room visits; hired nurses to check on patients following hospital stays to help prevent readmissions; employed pharmacists to call our diabetic patients via telephone to assist them with sugar control; and created a multitude of other services, all designed to improve patient care, keep costs down, and capture lucrative bonus dollars.

If this sounds to you like a “win-win-win” scenario for insurers, doctors, and patients, then you are not alone. Many policy experts expected this system of doing business to be the savior of the healthcare system when it came into vogue about a decade ago, in part due to a push from the “Affordable Care Act” (also known as “Obamacare”). 

But there is a wrinkle.

The Hitch

In the above example, I chose $1 per patient per year as the expected cost to care for our hypothetical population of 10,000 senior citizens. But not all Medicare beneficiaries are going to cost the same to care for, and it turns out that estimating what the cost of care for a particular patient should be is a very tricky business. After all, a Medicare beneficiary might be a newly retired person who is 65 years old, or they might be a 90 year old who has been retired for decades. On average we would expect the former to be healthier than the later, and hence in need of less healthcare. 

And even between 65 year olds, there is a wide range of needs. One 65 year old might be an athlete who has never smoked and takes no medications, while another might be a smoker and a diabetic who has already suffered a heart attack. Clearly, the expected cost of caring for each of these patients is going to be vastly different, and therefore the amount at which a medical group can claim “victory” for keeping costs down is not going to be the same for each. Moreover, things will change over time. The healthy 65 year old who runs marathons today might be the very ill 75 year old a decade hence.

All of this complexity presents a major problem. If Medicare estimates costs too high, practices will easily come in under target, costing Medicare huge sums of money in unwarranted bonuses. If they set the targets too low, practices will never make a bonus, and hence give up on even trying. How then to account for all of this? 

Medicare could simply provide an average cost for all patients across the system, understanding that one beneficiary may cost much more than the average while another costs much less, and that it will all work out when averaged through a large enough population of patients. But this would incentivize medical practices to “cherry pick” the younger and healthier patients, and punish those groups who take the more ethical route of caring for the sick and the elderly. So Medicare needs a way to provide a more  precise method of budgeting for each patient, ensuring that no particular Medicare patient will ever be viewed as a “lost cause” of potential shared savings to their healthcare providers. Enter the RAF score. 

What’s the Score?

Risk Adjustment Factor (RAF) scoring was created as a way to track all of the above. In brief, RAF scores are medical billing codes that influence how expensive a patient is expected to be. 

For example, if a doctor places the code for “Type 2 Diabetes” into at least one note during one visit during a calendar year, then Medicare will run that into its database and use that diagnosis to adjust the expected cost of care upward for that patient, since diabetics on average consume more healthcare dollars than their non-diabetic counterparts. There is a long list of such codes that result in up-regulating budgeted expenses: heart failure, atrial fibrillation, tobacco use, kidney disease, and more than a hundred others. 

Unfortunately, this system incentivizes your doctor (or more typically these days, the company that employs your doctor) to make you look as sick as possible on paper. The ideal patient in this system (from the point of view of a medical practice) is one who looks dreadfully ill on paper—and hence will be budgeted by Medicare for a lot of expenditure—but is in fact reasonably healthy, and will consume a small amount of healthcare, thus earning the coveted bonus money for cost savings. 

Doctors are pressured to raise revenue by making the patient look as sick as possible on paper.

And so, about five years ago, doctors began to experience various pressures to play “the RAF score game,” by entering as many medical problems as we could into our patients’ charts without overtly lying.

At first, my practice required us to attend a few training sessions to learn about RAF scoring.

Then we began to receive “friendly reminders” from a team of coders hired by our company to audit our charts, instructing us to be more aggressive about adding these diagnoses to our notes. “Stage 3 Chronic Kidney Disease,” began to appear in a lot of my notes (sorry, Mrs. Smith) since, as discussed earlier, most senior citizens technically have this condition and it results in a RAF score adjustment. 

Then, because Medicare is perpetually broke and not run by total idiots, we were told it wasn’t just enough to add these codes into our notes—we had to actually write something about each problem, even just the notation that the problem is stable. This led to huge amounts of wasted time documenting things of no real importance, and produced a breakdown in the art of medical charting. 

For example, in the old days, a patient who came to see me for low back pain might lead me to write a note that looked like this:

“Low back pain—most likely due to a muscle spasm. I will prescribe a muscle relaxant and physical therapy. He will call back if not improving in two weeks”

But now, under pressure to add as many RAF score codes to each note as possible—even when they have nothing to do with the reason the patient is coming for an appointment—it looks instead like this:

1. Low back pain—most likely due to a muscle spasm. I will prescribe a muscle relaxant and physical therapy. He will call back if not improving in two weeks
2. Type 2 Diabetes—stable
3. Stage 3 Chronic Kidney Disease—stable
4. Depression—stable
5. Coronary Artery Disease—stable
6. Tobacco use—stable, patient quit 20 years ago and has not resumed smoking since
7. History of Prostate cancer—stable, patient had his prostate removed 10 years ago, cancer has not recurred, he continues to follow-up with his oncologist for this issue

Then things got even worse.

A workaround for the workaround was to make the diagnosis codes increasingly elaborate.

As a workaround for the above workaround, Medicare started making diagnosis codes increasingly elaborate. Initially, it was enough to document that a patient has type 2 diabetes. Then we had to document for example that they have Type 2 Diabetes with resultant eye disease. Then we had to document the type of eye problem, how advanced it was, and which eye it was in. It’s now common for me to document patients with “Type 2 Diabetes Mellitus with stable proliferative diabetic retinopathy, right eye,” even in a visit which has nothing to do with the patient’s diabetes nor their eye problems. Which takes up even more time, further clogs up the note, and does nothing to improve patient care. 

(As an aside, uncluttered and concise notes are important. Notes are how doctors communicate their thoughts to each other, as well as to their future selves. A well-written note can prove vital if another physician needs to assume your care in an emergency situation. Notes cluttered with useless information aren’t just a nuisance, but potentially a patient safety issue.)

None of this should be that surprising to an observer of the American medical system. In much the same way that scientists design antibiotics to kill bacteria, only to see those bacteria evolve antibiotic resistance, Medicare has been proclaiming regulations designed to contain costs almost since its inception, only to see doctors outwit those regulations. This in turn leads Medicare to proclaim new regulations, in a cycle of endless bureaucratic sprawl. 

In sum, RAF scoring is wreaking havoc on the system. It is leading to doctor burnout, patient anxiety, a breakdown in medical charting as a mechanism of patient safety, and an army of new employees in the health system whose role is not to deliver better care but to help organizations capture every dollar they possibly can from Medicare. While I do not know of any study that has examined the problem, I suspect that these costs are eroding much, if not all, of any savings that “Population Health Management” has managed to achieve. 

Is There a Better Way?

What can be done about this? Unfortunately there aren’t any easy answers. In another essay here, I suggest we replace our insurance system with a different model of care. But since that’s unlikely to occur anytime in the near future, let me close by offering a few less revolutionary proposals that can help. 

First off, if Medicare feels that RAF scoring is vital to its interests, then they should at least take it off of the doctor’s plate. A way to do this would be to hire third-party vendors that are considered neutral arbiters and who are trusted by both Medicare and by doctors to act fairly. These vendors would audit charts and create RAF scores for patients in a Medicare database that would then be used to calculate cost savings and bonuses. Doctors would no longer be expected to serve this function, and non-relevant items would stop taking up large chunks of space in patient’s notes. 

A second, more bold proposal, would be to eliminate RAF scoring and population health management contracts altogether. Of course, this system was put into place in an effort to contain costs, so simply going back to the “old ways” wouldn’t be a great option either. I would propose instead that Medicare try creating simplicity and transparency. 

One way would be to replace the current way that billing for visits and co-pays are handled. In the current system, payments from population health management and RAF scoring are as yet only 20-30% of payments to doctors, with the bulk still coming from Medicare paying doctors piecemeal for each patient visit they do. As it stands now, this is a complex process (it would take another whole essay to describe the requirements to doctors for billing an office visit) and leaves 20% of the bill for each visit to the patient. Medicare could instead tie billing to the amount of time that a patient takes, and have co-pays kick in only after a certain amount of time for a visit is reached. 

For example, a computer program could be turned on the moment the doctor enters the room and turned off at the end of the visit. The program would keep track of how many minutes the visit lasted, and the doctor would be paid accordingly. In order to incentivize both patient and doctor to be as efficient as possible, perhaps the first ten minutes of the visit could be paid for 100% by Medicare at a high rate, but every minute thereafter would incur an increasing amount of co-pay to the patient and reimburse the doctor at a slightly lower rate than the first ten minutes did. 

Since lengthy visits would cost the patient increasingly more, and since a doctor would always earn more by going on to the next patient than by spending more time with the current patient, this would incentivize a real efficiency in the system, namely encouraging both the patient and the doctor to use time as wisely as possible. By contrast, our current system incentivizes doctors to waste a lot of time documenting information that is of dubious benefit to the patient, and does not give the patient any “skin in the game” for using healthcare dollars efficiently. 

Finally, Medicare needs to adopt the physicians credo “first, do no harm.” In other words, Medicare needs to scrutinize any new policies not only for how they might impact costs, but also through the lens of whether or not they might have unintended consequences such as creating greater distractions for doctors, increasing the complexity of the system, or in other ways making it more difficult for actual medical care to be delivered. 

Doctors should be free to focus on caring for patients, not writing endless reams of documentation for the purposes of navigating an overly complex billing system. Until Medicare places this concept at the center of its philosophy, it will continue to bounce from one bad “innovation” to another.

This is one of an ongoing series from Dr. Bittermilk, offering solutions to a variety of issues surrounding the U.S. medical system.

Sign up for our free newsletter, find us on instagram and linkedin, and tell us what you think.

Reginald Bittermilk, M.D.