Is there anything more frightening than the image of a young toddler running while holding a pair of scissors? From the time we were old enough to pull ourselves up on the edge of the couch and take our first staggering steps, our mothers warned us to never run with scissors. Some health care providers are determined to take on incidence risk in the form of global spending targets for defined populations, often in the form of accountable care organizations (ACOs), a concept that emerged from the Affordable Care Act of 2010. We would be wise to remember the lessons we learned from Mom.
Prospective payment risk, for example inpatient DRGs or episode-of-care bundled payments, arises from pre-determined payments based on a patient’s specific illness or procedure. Providers are tasked with fitting resource consumption within the limits of the prospective payment. Incidence risk, by contrast, involves the unknown prevalence of various illnesses or impending health care events within a defined population. How many people in a population of 5,000 will eventually develop cancer, or suffer from a chronic disease, or fall from a ladder while painting the garage, is completely outside of the control of health care providers. Plus, the economic risk for the incidence of illness or accidents is best left to heavily capitalized insurance companies, who aggregate millions of covered lives to arrive at tolerably predictable incidence rates.
Claims costs for a group of 5,000 attributed lives randomly fall within the risk-free corridor of plus or minus 2 percent of target (the typical ACO measurement standard) less than 50 percent of the time. If an ACO had no impact whatsoever on the claims costs of a group of 5,000 individuals, an unearned bonus would randomly accrue more than 25 percent of the time while undeserved penalties would be incurred more than 26 percent of the time. There is a greater chance of an “incentive error” than the likelihood that claims costs would fall within the expected corridor. From a statistical standpoint, a group of providers would improve their chance of winning by taking $1 million to Las Vegas and betting it all on red at the roulette table.
Forty percent of ACOs with 10,000 attributed lives would see random claims fluctuation sufficient to generate at least $500,000 in unwarranted penalties at least once over a three-year period while having had no adverse effect on population claims costs whatsoever. Undeserved penalties of $1 million to $2 million would be most common, but the worst case would be a loss of almost $5 million that was unrelated to anything that the ACO did or did not do. Similarly, 30 percent of ACOs with 25,000 attributed lives would incur unearned penalties ranging from $2 million to $4 million due solely to random claims fluctuations for such a small population. The least fortunate ACOs in that scenario would face $7 million in unwarranted penalties.
As troubling as the prospect of physicians and hospitals wrestling with incidence risk may be, even more concerning is the realization that such risk-taking has virtually no chance of solving the affordability crisis for the American middle class. The typical working family of four spent approximately $9,200 on the combination of insurance premiums and out-of-pocket medical expenditures in 2001. By 2016, that figure had grown to more than $26,000. By 2030, a middle class family will face $50,000 per year in health care spending–more than 30 percent higher than would be the case if health care costs increased at the same rate as wages. With spending on pace to be 30 percent higher than the middle class can afford, ACOs have failed to consistently demonstrate even 2 percent savings, while adding enormous administrative overhead costs to create the infrastructure to share risk.
Prospective payment rates work. DRG payments give providers a financial incentive to reduce resource consumption during the course of a hospital confinement and to shorten lengths of stay when clinically appropriate. Bundled payments for joint replacements have spurred providers to examine more closely the discharge disposition of post-surgical patients; the use of post-acute care facilities by patients who were able to be discharged home dropped significantly with the introduction of bundled payments.
Prospective payments for a specific set of diagnoses and comorbidities, or anchored to a procedure or other identifiable event, create financial incentives for efficiency without introducing the highly volatile incidence risk that is entirely outside of the providers’ control. Population health, to the extent that it shifts incidence risk to providers for populations smaller than 100,000 beneficiaries, is like handing a pair of scissors to a toddler. We can tell the toddler to hold the scissors pointing downward, and to walk and never run, but it would be wiser not to give them scissors at all.
About the author and the Vizient Research Institute™. As executive director of the Vizient Research Institute, Tom Robertson and his team have conducted strategic research on clinical enterprise challenges for 20 years. The groundbreaking work at the Vizient Research Institute drives exceptional member value using a systematic, integrated approach. The investigations quickly uncover practical, tested results that lead to measurable improvement in clinical and economic performance.