Time to challenge ROI of medical homes

Almost every initiative in healthcare costs money. But proponents invariably promise that the particular initiative in question will ultimately save more than it costs, and the subsequent outcomes reports almost invariably demonstrate a high return on investment.

Yet healthcare expenses continue to rise. How can this be? The savings claimed for these initiatives are illusory, but usually all that is lost is the cost of the initiative itself.

Not so with patient-centered medical homes. Medical homes are the first innovation whose cost is high and whose implementation actually can increase costs far beyond the program itself. This is exactly what happened in the Community Care of North Carolina Medicaid medical home program.

A consulting agency estimated a savings for the state approaching $300 million from the program. No one in North Carolina—or in states where similar programs are being proposed based largely on the North Carolina experience—demonstrated any health outcomes numeracy when reviewing this result. Instead, in North Carolina and other states, people just accepted it.

REALITY CHECK

No one applied a “plausibility check” to curious findings of savings—findings at odds with common sense:

• Physician costs (11% savings). Wouldn’t North Carolina’s medical society be complaining if doctors were providing more care but getting less reimbursement— instead of being among the program’s biggest supporters?
• Drug expense (5% savings). Wouldn’t more drugs have been prescribed if more preventive care were being provided?
• Inpatient care (46% savings). Since a large percentage of admissions are simply not preventable, a 46% savings requires that virtually every preventable admission was avoided. A specific piece of data was used for another plausibility check: For the diagnosis of asthma—perhaps the most preventable reason for inpatient admission and therefore the diagnosis where one would expect the greatest percentage of savings—admissions fell by only one quarter.
• Outpatient services (24% savings). If the care isn’t being provided in the office or with drugs or inpatient care, shouldn’t outpatient costs be rising?
• The greatest savings was found in infants (54%), and yet the relative rate of the only major source of preventable discharges in that age group—neonates—barely budged statewide. Why didn’t anyone check that actual result against the reported 54% savings?

What could explain the difference between the actual results and the report? Actuaries rely heavily on trend assumptions, changes in which can swing outcomes. A journal article from the actuary indicates that “choice of trend has a large impact on savings.” A valid analysis requires an understanding that trend is not a “choice.”

A specific level of savings was achieved, and it’s up to the evaluator to find it. Reliance on variables that depend on personal choice automatically invalidates any result, as any outcomes-numerate analyst knows.

Meanwhile, if initiatives and access decisions on this scale continue to be considered generally, the entire healthcare sector needs to learn how to analyze reported outcomes more critically in order to distinguish true cost savings from fallacious results.

If you enjoyed this post, make sure you subscribe to my RSS feed!