Why Quality Assurance Doesn’t Save Much Money

Today’s Managing Health Care Costs Indicator is 30%

A perspective e-published by the New England Journal yesterday asks a challenging question:

Why haven't nearly two decades of work on improving health care quality had a measurable effect on health care costs?

The article answers the question, too. 

The authors note that truly variable costs like supplies and medications – which are saved if utilization decreases – are pretty small.  While economists and accountants frequently say “there are no fixed costs,” that is over a long time horizon.   Today and tomorrow, there are many costs that are fixed in health care.  We can’t close hospital real estate or fire clinical and nonclinical personnel based on just a LITTLE less utilization – we need to have a LOT less utilization to get rid of many of the costs of our system. Quality improvement can lead to small incremental decreases in utilization, but these are often not enough to let us diminish the underlying cost of delivering care.

Let me give two examples.

If a quality assurance program keeps a patient out of the emergency department – what cost savings would you expect to receive? All numbers are illustrative only.

-        $1000 (likely charges for an emergency department visit)
-        $600  (average allowable cost of an emergency department visit)
-        $400 (average allowable cost of a lower intensity visit that is preventable)
-        $100 (cost of the time that the nurses and physicians spend with the patient)
-        <$5 (cost of the electricity used, tylenol dispensed and the wax paper from the exam table)

The emergency department would have been there, as would the doctors and nurses, regardless of whether that patient was seen unnecessarily.  So it’s only fair to credit the quality assurance program with a small amount of “real” savings.

Here’s another example.  Many have suggested that electronic medical records will decrease the duplication of laboratory tests.  This is a good idea – people shouldn’t be stuck with needles because we’re bad at keeping track of lab tests!  But if a physician doesn’t order an unnecessary blood count, what will be the savings?

-        $50 charge
-        $12 allowable
-        <$0.50 cost of reagent to perform the test.
The real savings are probably under a dollar - unless a LOT of blood counts are eliminated. 

I’m told by my friends in hospitals that the marginal cost to deliver additional hospital services is generally about 30%.  (This is assuming there is some surplus capacity in the system, which is usually the case.)  That means that preventing overutilization might decrease billings and decrease the amount paid by health insurers. But the nasty secret is that lowering this utilization only lowers the actual resource cost of hospitalization by about 1/3 of the amount of saved billings.

The authors of the article are skeptical that the time-defined activity based costing suggested by Kaplan and Porter can significantly lower health care costs.  I'm more optimistic that engineering studies of health care can save dollars. If TD-ABC is really effective, organizations could diminish more of the formerly-fixed costs.  But that’s a hypothesis – and we need to measure true cost savings carefully before we prematurely declare victory over the high cost of medical care.