The following originally appeared on The Upshot (copyright 2016, The New York Times Company).
A growing proportion of Medicare beneficiaries are opting out of the government-run insurance program. They are instead choosing a private plan alternative, one of the Medicare Advantage plans. The strength of this trend defies predictions from the Congressional Budget Office, and nobody can fully explain it.
Here’s another mystery. Traditional Medicare spending growth has slowed, bucking historical trends and expectations. Though there are theories, we don’t fully know what’s causing that either.
Pinning down explanations for these two mysteries is important. Doing so could help us understand the structure and cost of Medicare in the future.
The mysteries may be connected by something that appears, at first, to be unrelated: Doctors and hospitals tend to treat insured patients the same way, regardless of what kind of coverage they have. A traditional Medicare patient admitted to the hospital with, say, pneumonia will receive the same standard of care as a similar but privately insured pneumonia patient.
From this, an idea emerges that links the two mysteries. As enrollment in Medicare Advantage plans grows, so too do the plans’ influence over how doctors and hospitals provide care. Unlike the traditional program, Medicare Advantage plans establish networks, covering care provided only by certain doctors and specific hospitals. Often those are the ones with lower cost growth. As doctors and hospitals reduce their cost growth to gain access to Medicare Advantage networks and the increasing number of patients enrolled in the plans, they do so for traditional Medicare patients as well.
So, as Medicare Advantage enrollment swells, the growth in the cost of care for traditional Medicare falls — a spillover effect. That’s the theory, anyway. Does it hold water?
A few studies have examined the question, and all support the spillover theory. The first study, examining the period from 1994 through 2001, found that when the proportion of Medicare beneficiaries enrolled in Medicare H.M.O.s grew by an additional percentage point, per enrollee spending in traditional Medicare fell by one percentage point. Another study, focused on the period from 1999 to 2009, found that a 10-percentage-point increase in Medicare Advantage market share was associated with a 4.5 percent decrease in per enrollee traditional Medicare hospital costs and a commensurate reduction in duration of hospital stays.
“These studies extend on a body of research that demonstrates that health care markets are connected,” said Michael Chernew, a Harvard health economist who participated in both studies. For example, research in the 1990s showed that when H.M.O.s grab enough of a health care market, Medicare spending is reduced.
A study published this year by Kevin Callison, an economist at Grand Valley State University, corroborates the phenomenon. It found that greater Medicare Advantage market penetration is associated with reduced hospital costs for traditional Medicare heart attack patients. But such patients were also more likely to die, a finding at odds with other work by Mr. Chernew and researchers at Harvard and Stanford that showed that the expansion of Medicare Advantage is associated with lower mortality.
Those studies examined periods that predate the Affordable Care Act, which changed how Medicare pays plans and hospitals. Since then, Medicare spending has continued to decelerate, and Medicare Advantage enrollment has continued to grow. So it’s important to look at more recent data to see if the spillover could still be connecting the two phenomena. Two studies published in the last year have done so, and they support previous findings.
A study spanning the 2010 passage of the A.C.A. by Katherine Baicker and Jacob Robbins, health economists at Harvard, found that a 10-percentage-point increase in Medicare Advantage market penetration was associated with a 7.3 percent decline in days that traditional Medicare patients spent in the hospital. Instead, patients receive more care in less expensive settings: Outpatient visits increased 5.5 percent, for example. In total, annual per enrollee traditional Medicare costs were lower by $252 for each 10-percentage-point gain in Medicare Advantage market share.
Finally, a study published last month by researchers with the Harvard T.H. Chan School of Public Health provided the most up-to-date look, using data through 2014. It found that in counties with Medicare Advantage market penetration above 17.2 percent, greater growth in Medicare Advantage was associated with slower growth in traditional Medicare spending. According to the study, the spillover effect accounts for 11 percent of the recent traditional Medicare spending slowdown. Because of the methods used, this may be a conservative estimate — the spillover effect could be even larger, explaining more of the slowdown.
If Medicare Advantage is responsible for slower traditional Medicare spending growth, should policy makers do more to encourage greater Medicare Advantage enrollment? One way to do so would be to coax more plans into the market by paying them more, which is controversial. “There’s almost continuous policy debate about how much we should pay Medicare Advantage plans,” Mr. Chernew said. “In addressing that, we should consider the full impact of Medicare Advantage, not just impact on those who choose to enroll in the program.”
We still don’t know exactly why Medicare Advantage is growing in popularity or why Medicare spending is slowing. But now we know that the two are related. Medicare Advantage growth doesn’t completely explain slower Medicare spending growth, but it is one piece of the puzzle.