Reflections on Healthcare Economics: The Skewed Incentives of a Third-Party Payer System

The provision and consumption of medical care are strange economic beasts - they defy traditional paradigms of economic behavior.* This is partly related to the fact that we treat our health and well-being differently from other traditional commodities. Traditional supply/demand economic paradigms imply a consumer who understands the utility of what they are purchasing. In medicine, however, clinical knowledge is largely in the hands of the physician and the patient cannot be expected to have a thorough understanding of what their money is buying. It is also affected by the extensively skewed incentives in the healthcare market, which have contributed to the acceleration of costs. The fact that there has been a third-party (i.e. health insurance company) paying for healthcare has removed some of the usual expected downward pressures of rising cost associated with consumption of commodities. This means that physicians (who have been paid per procedure and test ordered) and patients (who want the best, most comprehensive possible care) have both been shielded from the rising costs of healthcare. 

From a patient's perspective, after paying his/her premiums and deductible, any further testing and procedures are effectively "free" in that the insurance company bears the costs thereafter. Why should a given patient not get the most "bang for their buck" and order any tests/procedures that the physician recommends? It's been the same headache he/she has had for years, but why not just get that MRI "just to be sure"? 

From a physician's perspective, there are numerous motives for ordering extra tests/procedures. The first involves the potential ethical/moral pitfalls of a model that has compensated physicians based on the number of procedures/tests ordered (fee-for-service). Of course, our goal is always that we perform the most efficient medicine and do no harm. However, the revelations of Medicare fraud have highlighted the potential for this model to go awry. 

The second motive involves the fact that most physicians are risk-averse and medicine is a field fraught with uncertainty. If an extra test/imaging/procedure can enhance the probability of our diagnostic confidence even slightly, we often feel compelled to act. This is especially true when the extra cost is not borne by the ordering physician or a given patient, but rather by the insurance company. 

The third is our fear of litigation and indulgence in defensive medicine as a protective measure; this goes hand in hand in with our often risk averse tendencies and desire for increased confidence in our diagnoses (more on this later). Throughout medical school, we learn these extensive physical exam maneuvers and classical presenting signs and symptoms. And yet, invariably, we often order the advanced imaging study. For example, if a patient presents with a cough and a doctor is confident the lungs are clear on exam and not affected, why should he or she order an x-ray? Because it increases our diagnostic confidence and also serves the double purpose as protective data in litigation. 

There are models that may address the skewed incentives for both the patient and the physician. For patients, one interesting and potentially revolutionizing idea involves broader implementation of Health Savings Accounts (HSA). This basically a unique savings account where a patient can save their own money to spend on healthcare costs. The idea is that patients will treat healthcare consumption more like a traditional commodity. For example, a patient who has mild symptoms (like the common cold) would theoretically be less likely to seek medical attention and go to the hospital/clinic only if symptoms become severe, resulting in more judicious expenditure. 

More recently, the goal for physicians has been to shift away from a fee-for-service compensation model to one that encourages more responsible ordering of testing and diagnostic studies. In the new paradigm, payments are bundled for a given healthcare encounter. For example, if I am admitted to the hospital for pneumonia, the hospital will receive a fixed sum for my hospitalization, as determined by the expected cost of a hospital stay for this diagnosis. This is expected to bring down how much hospital and doctors get paid. However, on the bright side, doctors will be expected to be more efficient in reaching a diagnosis and order only what is necessary to treat a given patient's illness, since any further testing may not be help the patient and will reduce a hospital/clinics margins. 

One interesting aspect of the new Centers for Medicare and Medicaid Services (CMS) compensation schemes that I have personally experienced is the new regime for readmission. Basically, if a patient originally admitted with a diagnosis of congestive heart failure, pneumonia, or COPD exacerbation (conditions often with a high-risk of re-admission) is re-admitted with the same diagnosis within 30 days of discharge, the hospital has to bear the cost of the second admission. This is effectively CMS telling us to take better care of our patients. I have noticed several positives resulting from this systemic change. First off, patients are now consistently scheduled to follow-up soon after discharge (often within days). I notice my colleagues spending more time at discharge discussing with patients why they were hospitalized and what they can do to take better personal care. More immediate outpatient follow-up for patients with recent hospitalizations as well as better patient education are definitely positive outcomes. 

A concern with the "more efficient" practice of medicine is the possibility that, as we move away from the often broad nets we now set in the search for diagnoses, we may miss other diagnoses. For example, in the aforementioned patient with the cough, a chest x-ray may show evidence of lung cancer as an incidental finding. However, in the bundled payment compensation model, there may be circumstances that the chest x-ray is avoided. While the patient may have their cough effectively treated by the efficient clinician, he/she would not have the benefit of an earlier cancer diagnosis that would be incidentally found on an x-ray.

Given that human life and well-being are at stake, healthcare will never fully behave like traditional commodities and classical supply/demand curves. However, I am hopeful that addressing the skewed incentives and shifting cost to the primary parties (doctors and patients) that matter most in the transaction will help reduce costs in the long run.

*Caveat: even traditional economics may not fully explain human behavior, even in the consumption of non-healthcare related goods and services. See this great article by Derek Thompson in The Atlantic.