Why Bundling Payments Won't Reduce Costs — Part 3


If you haven’t read parts 1 and 2 of this manifesto, please do so here and here before reading further.

How will bundled payments affect the incentives for each of the players in the medical market?

For patients, a change to bundled payments will probably have little effect upon monetary issues or fears. Demand for medical care will increase. With millions of additional patients being added to Medicaid roles, and with government “paying” the costs, there will still be little disincentive for patients to seek comprehensive care. In addition, patients who are forced to purchase insurance through health care reform will want to get something for their money.

Bundled services will obviously benefit the insurers. Otherwise there would be no incentive to move to such a model. For insurers, bundled payments will increase profits. Much of the uncertainty involving payments for medical care disappears with bundled payments. If a patient with a heart attack develops a complication requiring prolonged hospitalization, in the current system the insurers bear the costs of treating that complication. Switch to a bundled payment model and the providers bear the risk of medical complications or outlier patients. Whether a patient is in a hospital for six hours or six months, the payment to the hospital for one diagnosis will be the same. The theory is that the threat of paying for complications will “encourage” hospitals to take steps to avoid those complications. In other words, a threat of financial liability will improve the quality of care. Kind of like suing our way to better health care … and we all know how well that has worked. For now, the point is that bundled payments increase profits for insurers by decreasing uncertainty in the payments that must be made to providers. As providers decrease costs, then the insurers will gradually decrease the bundled payments while gradually increasing the premiums that every person in the United States will be required to pay under health care reform. Profits go up.
Demand for insurance will go up under health care reform because there is a mandate that everyone purchase insurance. Insurers will encourage people to buy into their plans. More subscribers plus relatively fixed costs equals more profits.
The financial risk that insurers fear in the current medical payment model is largely erased by a bundled payment model. While insurers may be forced to accept all patients – even those with pre-existing conditions – bundled payments diffuse the risk that the insurer must accept. Even though some patients may be hospitalized more often than others, the insurers know that they will only have to pay a fixed cost for the hospitalization.
There will also be a decrease in the legal risks to insurance companies with a bundled payment model. Insurers will be less liable for refusing care. They pay the providers one fee and the providers are then forced to decide what care is and is not “necessary.” Also look for our government to create additional legal protections for insurers as health care reform becomes implemented over the next few years.

Probably the largest effect of bundled payments will be felt by providers of medical care.
For providers, bundled payments will create an incentive to provide less care. Currently, extremely ill patients create profit through utilization of costly medical services. More services = more payments. When providers are paid one price for a given diagnosis, regardless of the severity of the illness, then the incentive will be to accept a large bundled payment and provide the least expensive medical care possible. This will happen in several ways.
First, providers will want to make patients look sicker so that the bundled payments will be larger. Bundled payments for a patient suffering from pneumonia will be much more than a bundled payment for a patient with a chest cold. Patients in respiratory failure will command an even higher bundled payment. Therefore, the incentive will be for providers to label patients with serious illnesses in order to receive higher bundled payments. Just like payments for catheter-related sepsis caused a significant decrease in the reported incidence of catheter-related sepsis (but an increase in other types of sepsis), increase in bundled payments for more serious illnesses will cause an increase in the reported incidence of serious illnesses. The problem is that those serious illnesses will get reported to the Medical Information Bureau and will follow a patient for the rest of the patient’s life.
Second, there will be less utilization of costly medical services. Look for invasive procedures to decrease. Providers will start pointing to medical studies saying that such procedures are not proven effective. Costly antibiotics and other costly medications will be off limits. Consultations will be less available.
Third, providers will begin avoiding patients who are more likely to suffer adverse consequences. Ideally, bundled payments will provide appropriate reimbursement for an “average” patient. Healthy patients will utilize less resources and therefore increase profitability for a given bundled payment. Young healthy patients who may need a day or two in the hospital for their pneumonia will be readily admitted as there will be a high likelihood of profitability with the ensuing bundled payment. Pneumonia patients with diabetes or with HIV who will likely need long admissions and expensive medications will become hot potatoes. Community hospitals will find reasons to transfer high utilizers to other facilities. Perhaps they need an endocrinology consultation. Perhaps they need an infectious disease specialist. Bundled payments will create an incentive to avoid treating obese patients, cancer patients, and other patients with chronic diseases. Financial solvency will be difficult to maintain with bundled payments and chronically or seriously ill patients.
Demand for services from providers will increase, since some patients will not be receiving the level of care to which they are currently accustomed. Patients may go from provider to provider trying to get the care that they desire.
I’m not sure how the fear issue will play out with medical providers. In the current system, fear is mitigated by providing more services. However, in a bundled payment system, providing more services will quickly erase profits and may lead to financial insolvency. How will medical providers adapt? My guess is that there will be less services and more studies and medical testimony showing why providing fewer services is within the standard of care. There will also be a backlash against hospitals if patients die because they didn’t receive what was retrospectively deemed to be “necessary” care. I also think that at some point there will be a revolt against regulatory agencies that create guidelines which increase expense without improving outcomes.

Bundled payments will also have several other effects:
First, the system will get gamed. Big time. If insurers are going to make a large bundled payment for a given diagnosis, expect more of those diagnoses to be made. Patients who previously were sent home with “walking pneumonia” will be admitted because admissions for “pneumonia” generate more money. The admission may only be “overnight,” but it will still generate that bundled payment. Outpatient diseases will suddenly require inpatient management – if that inpatient management is what generates the bundled payments.
Second, bundled payments will allow insurers to vilify medical providers. In the current system, insurers are the bad guys when they refuse to authorize or to pay for medical care. By bundling payments, insurers will be able to blame medical providers for not providing more services because those services are included in the bundled payment. Patients will then direct their anger toward providers when the patients don’t get the medical services that they want.
Finally, bundling payments will also cause fighting between providers. How do physician consultants get paid when the hospital receives the bundled payment for the patient’s illness? The pie is only so big and anyone that provides services is going to want a piece. Hospitals are already trying to minimize this problem by purchasing physician medical practices. When physicians are employees and paid by the hospitals, the hospitals get to keep the bundled payments. Otherwise, let the fights begin.
What happens if a patient goes to an emergency department with a pneumonia and needs to be transferred? Who gets the bundled payment? What if a patient is hospitalized for a hip fracture and then develops a pneumonia while in the hospital? Who gets the bundled payment? Will the payments be split? If so, how much? I posed these questions to a friend who works at CMS. Her response was that the providers would have to create agreements regarding payments for services. Of course, providing a prospective division of payments for every possible type of care would be impossible, so the providers will be left fighting over who gets what payments and how much. When providers fight with each other, nothing good happens. Divide and conquer.

Bundling payments will protect insurers, increase insurer profits, and decrease the willingness of providers to care for seriously ill patients. When the only variable for payments from insurers is how many times a diagnosis is made, the diagnoses will be made more frequently and will result in an increase in the number of “bundled” payments.
Bundling payments will also cause rifts between medical providers that will ultimately detract from the medical care provided to patients.

Stay tuned for Part 4 where I discuss solutions that will reduce costs.


  1. Hmm, not sure if its relevant, as the Australian & USA systems are quite different, but in Victoria (one of the States in Australia), we have had a ‘casemix’ system in acute care for nearly 20 years now (so have several other States in various forms). The system, while mainly government funded, operates a bit like the bundled payment model you’ve outlined – hospitals are payed $X to deliver care for patients with diagnosis Y.

    The idea is that hospitals will be able to treat some patients for less than Y, some for more than Y, but that it should roughly even out.

    Payments are set with a base level of $X per diagnosis, with inlier and outlier payments as well – so if a patient is in only overnight, when usually they’d be 2 days, there will be less payment. Equally, if complications occur and suddenly the patient is in ICU for a month, that is compensated. There are co-payments for certain things that are known to cost more – such as ICU treatment, or the patient having certain co-morbidities such as HIV.

    Where a patient is treated at one emergency department and then transferred to another hospital, both hospitals get paid for the relevant portion of the care provided.

    While the system isn’t perfect by any means, patients do not seem to be being routinely over-diagnosed or under-treated…

    I should probably also mention that while the hospitals are not required to make a profit, they do have to remain financially solvent, and *generally*, they are. Regarding payment though – generally Australian doctors are employees of the hospital, so tend to recieve a salary, rather than direct per-procedure payments, so that might also be a factor in the bundling working reasonably.

    Sorry to ramble & love the blog 🙂

  2. I believe bundled payments should be limited to elective procedures like knee replacements or perhaps even non-emergency coronary bypass. No emergency or unscheduled services. Also, need to be risk adjusted. That might solve some of the issues you have raised (though not all). Great effort.

    • Agree that non-emergent services are less subject to some of the things that I mentioned, but then some of the higher risk patients will still get vetted and refused care. For example, an obese patient with diabetes and a history of blood clots might not find someone willing to do an elective bypass operation due to the high possibility of complications. And a healthy patient with a glucose of 130 may get called “diabetic” to increase payments under risk adjustment without truly being “diabetic.”
      That’s a good starting point, though.

Leave A Reply