Untangling Knots: How a Fee-For-Service Model Complicates the Work of an AF Doctor

John Mandrola


August 26, 2013

Let's be bold.

I believe a major change in healthcare policy could improve the care of a large group of patients with heart disease.

Some leeway, please . . .

A patient presents with atrial fibrillation. Imagine giving caregivers—hospitals and doctors—a chunk of money to care for this problem. Then, that's it. Use it up on extraneous stuff, and the rest is on the house.

It's true; I often wish that atrial fibrillation would register on the radar screen of policy makers and quality folks. Patients with pneumonia, MI, stroke, and venous clots have benefited from such attention. For some reason, AF therapy hasn't reached this level of surveillance. It's not as if caring for the millions of inflamed Americans with chaotic atria is inexpensive.

Perhaps this lean to the left has something to do with burnout. Not the normal kind of physician burnout; I've never been more excited about doctoring. It's more the fatigue that comes from allaying fear and undoing excess care. Currently, an AF specialist in the US spends too much time managing fear and untying knots. I stop far more drugs than I prescribe.

There are many reasons why patients with AF get tangled up. One is that they are susceptible. Tranquil, well-scheduled folks rarely get AF. This intriguing phenotype, however, will not be the focus of this post.

What I want to consider is how our healthcare delivery system, particularly the fee-for-service model, makes my job as an AF doctor harder.

The Study

An article published this spring in the Journal of the American Medical Association gets to the essence of the problem. Influential surgeon and writer Dr Atul Gawande and colleagues studied the relationship between surgical complications and hospital finances. Their findings were stunning: The average US hospital makes money when surgical patients suffer complications. This is because patients who suffer complications get more care, and the added services increase reimbursement. In short, the more a hospital does, the more it makes. The corollary is worse: some hospitals are financially penalized for improving surgical quality.

If this doesn't shock you, you aren't paying enough attention to policy. It's not only an embarrassment; this sort of model fosters overtreatment of almost all ailments, especially AF. Treating patients with AF requires nuance, intense patient education, and large doses of patient-caregiver teamwork. AF care must not ever be worse than the disease. Now you can see the problem: nuance, education, and teamwork do not pay as well as procedures or procedure-related complications.

Three common scenarios involving the in-hospital care of patients with AF support my thesis:

The IV-Diltiazem Debacle:

Patients with AF are too often treated aggressively with the IV calcium-channel blocker diltiazem. There is no evidence that acute rate control with IV-diltiazem lowers the risk of stroke, heart attack, or death. I've researched it. The (ancient) studies that led to approval of this drug looked only at short-term rate control. The pharmacokinetics of the drug don't lend itself well to long-term infusions; the risk of hypotension is real; the resulting bradycardia leads to many unnecessary pacemakers; and negative inotropic agents are contraindicated in patients with systolic dysfunction. The use of IV-diltiazem is easy. It has become accepted practice. It constitutes doing something. And . . . the reflexive use of this drug is a metaphor for what is wrong with so much of US healthcare.

There are no financial penalties for the extra work created from IV diltiazem. Rather, it's the opposite. Excessive bradycardia for which a pacemaker is implanted is reimbursed. A complication from the unnecessary pacemaker is also reimbursed. Hypotension requiring an ICU stay or pressor drug—yes, this too is reimbursed.

I'm not the only one negative on IV diltiazem. Here is the opinion of a clinical pharmacist from a prestigious university.

Excessive Inpatient Workups:

The symptoms of AF overlap with ischemic heart disease and/or congestive heart failure. Chest pain, dyspnea, and near syncope are common symptoms of all three disease states. This fact gives caregivers carte blanche to order scores of scans and tests. In a fee-for-service model, both doctor and hospital are rewarded by such "thorough" workups. Nuclear stress tests are done. Nuclear stress tests pick up asymptomatic coronary disease. This leads to cardiac catheterization. Stents get deployed, and now the patient with AF plus an incidental 70% circumflex lesion is stuck taking an antiplatelet and an anticoagulant. All this is reimbursed. And, if there's a complication of the unnecessary cath or stent; well, you guessed correctly—it's reimbursed.

Then there is the echocardiogram and its visually estimated ejection fraction. All agree that an echo is a useful test in patients with AF. Knowing the systolic function of the left ventricle, for instance, is important. But let me ask two questions: 1) Does a shadowy image of a tachycardic left ventricle bathed in a negative inotrope constitute a reliable assessment of systolic function? 2) Could LV systolic function be better assessed when the patient is adequately rate controlled or in sinus rhythm?

Yet there is a value in making the diagnosis of "CHF." It bumps up disease complexity, which then brings higher payments to the institution.

Adding the diagnosis of CHF or CAD to patients with AF creates knots. Now rate control is harder to establish because the protocol-mandated ACE inhibitor has lowered the blood pressure. Or, in the case of stents, the blood-tinged urine in the presence of multiple clot-inhibiting drugs has delayed cardioversion—and/or necessitated a GU workup.

A children's book comes to mind—something about giving cookies to a mouse.

Overzealous Use of Nonoral Anticoagulation:

How often is a patient with AF kept in the hospital on IV heparin or low-molecular-weight heparin until warfarin moves the INR into therapeutic range? This practice is entrenched. Yet, excluding periprocedural patients, there are no randomized controlled trials that have even considered the use of nonoral anticoagulants for stroke prevention in AF. No matter, dogma calls us to extrapolate long-term data on oral anticoagulation to the IV and sc varieties. Do you ever wonder why AF patients are treated as if they had acute coronary syndrome? I do. Consider that a CHADS-VASc score of 3 confers an annual stroke risk of 3.3%. What's that per day? And yes again: all this non–evidenced-based care is reimbursed.

I have more examples. In the interest of brevity, I'll leave out the folly of treating asymptomatic AF patients with rhythm-control drugs. I also won't mention the use of TEE-guided cardioversion merely for convenience. One good thing about the labor-intensive nature of AF ablation is that it discourages overuse.


The care of patients with AF costs billions. As Americans grow larger, less mobile, and more inflamed, costs will continue to rise. We waste tremendous amounts of time and dollars treating patients with AF.

I can't think of a disease (except perhaps old age) that begs for a more nuanced, less disruptive approach to its care.

AF is a serious but not immediately life-threatening disease. We can do a lot of good; but we also can do way too much. It's very easy to entangle AF patients in well-reimbursed care. These knots can be tough to undo.

This is why I believe capitated care for AF would force us to get to the heart of the matter. Such a model would surely save money. It would likely improve outcomes. Best of all, it would value nuance.

This can't come soon enough for me.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: