“Reference pricing” is yet another intrusive diversion from real reform

Reference Pricing: A Small Piece of the Health Care Price and Quality Puzzle

By Chapin White, Megan Eguchi
National Institute for Health Care Reform, NIHCR Research Brief No. 18, October 2014

As purchasers seek strategies to reduce high health care provider prices, interest in reference pricing—or capping payment for a particular medical service—has grown significantly. However, potential savings to health plans and purchasers from reference pricing for medical services are modest, according to a new analysis by researchers at the former Center for Studying Health System Change (HSC) using 2011 private insurance claims data for about 528,000 active and retired nonelderly autoworkers and their dependents. In 2011, the California Public Employees’ Retirement System (CalPERS) adopted reference pricing for inpatient knee and hip replacements. Using quality and price information, CalPERS set an upper limit of $30,000—the reference price—for hospital facility services for a knee or hip replacement. CalPERS designated certain in-network hospitals as meeting the reference price, and patients using designated hospitals are responsible only for the health plan’s usual cost-sharing amounts. However, if patients use a non-designated hospital, they are responsible for both usual cost sharing and any amount beyond the $30,000 reference price. While reference pricing for inpatient services has some potential to steer patients to hospitals with better quality metrics, only limited savings—a few tenths of a percent of total spending—are possible from applying a similarly narrow reference pricing to other privately insured populations. The potential savings from reference pricing are modest for two reasons: Shoppable services only account for about a third of total spending, and reference pricing only directly affects prices at the high end of the price distribution. When considering reference pricing, employers and health plans need to weigh potential savings against increased plan complexity and financial risk to enrollees, along with the analytical and financial resources needed to create and manage the program.

From the Implications

The CalPERS reference pricing experience tells two different but equally true stories—a dramatic percentage decline in prices and spending on knee and hip replacements and an extremely small percentage decline in total spending. To significantly impact spending among the privately insured, reference pricing would have to be applied quite broadly. And, even using a very inclusive list of shoppable services, the potential savings are relatively modest.

Both conventional network-based plans—preferred provider organizations and health maintenance organizations—and reference pricing suffer some of the same limitations. Both types of plans rely on patient cost-sharing differentials to steer patients to certain providers, but the higher cost sharing cannot reasonably be applied in emergencies when patients can’t choose their provider. Also, both types of plans are vulnerable to the demands of dominant “must-have” providers, either to be in network or to be both in network and designated.

Compared to a limited-network plan, reference pricing faces at least three additional logistical hurdles. First, the health plan must have reliable price data for specific providers for specific services so that it can set the reference price and designate providers. Even very large plans will lack the historical data to accurately measure the prices they typically pay to smaller hospitals. Second, the plan would ideally have provider-specific quality metrics on hand that can be used to assure patients that they are not being steered to low-quality providers. Although hospital quality metrics and rankings abound, the methodologies behind those rankings are still under development. Third, reference pricing requires new customer-service tools to support shopping by patients and to deal with inevitable member complaints. Implementation would require commitment of significant resources by the plan, potentially offsetting some or all of the savings from reductions in payments to high-price providers.

The main disadvantage of reference pricing is that it adds a new layer of complexity for plan administrators and enrollees. Rather than facilities simply either being in or out of the network, there are now three types of facilities: in-network designated, in-network non-designated and out of network. Even more confusingly, a single facility might be designated for one type of service—for example, an inpatient hospital providing a knee replacement—but not designated for another—that same hospital providing a colonoscopy in an outpatient department. Additional complexity raises significant concerns, given that the basic elements of conventional benefit design are already beyond the grasp of many consumers.

One question is whether a reference pricing program can steer patients to lower-price, adequate-quality providers. The answer, based on the CalPERS experience, appears to be yes. But, that may not be the right question. A better question may be why private health plans would ever pay negotiated prices over $30,000 for inpatient knee and hip replacements. The CalPERS reference pricing program seemingly took a hard line against hospitals charging unreasonably high prices—$30,000 or more—for knee and hip replacements. But, is $30,000 really a reasonable price for an inpatient knee or hip replacement? To put that amount in perspective, the Medicare program on average paid $14,324 for inpatient knee and hip replacements in 2011.



By Don McCanne, MD

Our policy wonks continue to look for methods of controlling health care costs that will protect the role of private insurers, and, above all, prevent us from drawing the inevitable conclusion that we desperately need a single payer national health program. Reference pricing is one more supposed cost-saving tool that has garnered much interest. What is it and will it work?

Reference pricing is a process in which a maximum price is set for a given medical service. The patient then either goes to a designated provider who has agreed to accept that amount, along with contracted patient cost sharing, as payment in full, or, if the patient chooses a provider with higher prices, the patient must pay the full difference.

California Public Employees’ Retirement System (CalPERS) experimented with reference pricing by establishing $30,000 as a reference price for the hospital services for a knee or hip replacement. The experiment was considered to be a success based on the fact that 15 to 30 percent of patients who would have gone to a non-designated hospital switched and went to a designated hospital instead. But how do you define success?

The package fee that the private market was able to negotiate was $30,000. Medicare on average paid $14,324 for the same services. Is it a success when private insurers are paying double the fees set by a public insurer?

Some might consider it to be a success when the patient did not have to pay the difference between the reference price and the price that would have been charged in a non-designated hospital, though what is the patient giving up?

Consider how a decision is usually made to have a joint replacement. The patient typically goes to his or her primary care professional where initial evaluation indicates that a joint replacement may be indicated. A referral is made to a specialist qualified to do joint replacements. If the decision is made to go ahead with the procedure, arrangements are made with the hospital the specialist is associated with. But then the insurer intervenes and threatens the patient with severe financial penalties (all costs above the reference price) unless the patient agrees to give up the specialist and hospital that is part of the team of her primary care professional, either as a formal integrated health system or as an informal team of community professionals working together. To escape the financial penalty, the patient pays the penalty of disruption in care.

So what did this successful experiment that everyone wants to emulate actually accomplish? Care of 15 to 30 percent of patients was disrupted in exchange for establishing a cap on payment that was twice what Medicare pays, and yet netting only “an extremely small percentage decline in total spending.”

Another important consideration discussed in this report is that reference pricing significantly increases administrative complexity - just what we need in a system that is unique in the world for all of its profound administrative waste. But very telling is the comment in this report that some of this administrative increase will be in the form of new customer-service tools “to deal with inevitable member complaints.”

Great - a disruptive program that doesn’t work very well but ticks everyone off! We’ll say it again - single payer.