Wednesday, March 20, 2013

Exorcising the Ghost of Cost Shifting: Why the Alternative May Be Worse

The cost-shifting ghost!
The Disease Management Care Blog continues to welcome blog posts from outside authors. This is another one, courtesy of Erik Tollefson, who works in the health policy field. He can be reached at erikDOTmDOTtollefsonATgmailDOTcom.

Of all the mythologies in the arcane world of health economics, cost shifting holds a hallowed place. First conjured up by commercial insurers in the 1970s to warn against catastrophic Medicaid cuts on hospitals’ financial positions, the rhetorical phantasm of cost shifting continues to rise from the dead to haunt the public sphere, particularly when politicians propose to reform public insurance reimbursement levels or undertake large-scale reforms.

The theory of cost shifting is fairly straight forward: hospitals raise prices on private insurance customers when public payments are cut in order to make up for lost revenue. For all the importance afforded to cost shifting, however, there still remains a (highly) inconvenient truth: Numerous academic studies over the past 20 years have failed to find systematic evidence of its existence.

A recent National Bureau of Economic Research paper by Dranove, Garthwaite, and Ody examines the phenomenon of cost shifting in a new light. While scholars traditionally have examined hospitals’ pricing responses to planned changes in Medicare and Medicaid reimbursement levels, the financial crisis of 2007 provided a unique opportunity to analyze how they responded to a one-time loss in wealth. That crisis had a substantive impact on most hospitals; Not only did consumer demand for services decline, but many hospitals lost a substantial portion of their endowments due to the ensuing market turmoil. Dranove and his co-authors wanted to explore if hospitals that lost a significant proportion of their endowment would “cost shift” in order to make up for lost wealth, compared to hospitals that did not suffer similar losses.

What the authors found was disconcerting. Only a small sample of hospitals raised prices in the aftermath of the crisis. Many more responded with another strategy: cutting costs. Hospitals axed planned and ongoing capital expenditure projects (e.g.,  electronic health records) and shut down low-profit centers, including resource-intensive trauma and psychiatric centers.

Although the paper’s results cannot necessarily be generalized to all health care markets, it does suggest that hospitals can and will respond to financial downturns by cutting vital services.
Since the concept of cost shifting offends widely held notions of fairness, the further subsidization of baby boomers’ Medicare benefits in the purported era of austerity might not be politically palatable. The paper by Dranove et al, however, shows that a far worse scenario is possible if Medicare payment rates are slashed: cuts to costly but high value clinical programs. That’s ironic, because many of the benefits ascribed to the Affordable Care Act were predicated on increasing access to crucial medical services, particularly in underserved areas.

The only good news is that if hospitals react to changes in reimbursement levels and wealth loss by cutting important services, policy makers will be unable to summon forth the spirit of cost shifting.  While skeptical economists everywhere may rejoice, that will be small comfort to communities that find that their local hospitals are cutting basic services.


Anonymous said...

Jaan, while my information is a bit old as a "former" hospital administrator from decades ago, I can only say that I saw cost shifting first hand through the use of software that would maximize the reimbursement to the hospitals by adjusting prices based on Medicare and Medicaid reimbursement. These increases in price would then flow to the insured who would pay higher rates as most were paying a percent of charges. we knew it.

May not be an issue today, but it sure was used then.

Jeremy Engdahl-Johnson said...

Will cost shifting hold steady, increase, or decrease?