American hospitals are expensive. Reams of data show that hospital-based services are more expensive than the same services provided in other settings. Moreover, the cost of hospital services has grown faster than costs in other parts of our health care system; from 1997 to 2012, for instance, the cost of hospital services grew a spectacular 149 percent, while the cost of physician services grew only 55 percent.
The explosive growth of hospital costs is one of the key culprits in the nation’s high health care spending. Nonetheless, attempts to reform hospital payment methodologies are usually greeted with fierce criticism from the industry. The latest clash, which stems from a CMS proposal to consolidate facility fees, offers an opportunity to review why hospitals are so expensive, to detail some of the larger issues with hospital billing practices and how they contribute to increasing health care costs, and to explore some ideas for reform.
The Facility Fee Conundrum
When you buy anything — a watch, a car, even groceries — you pay a single price for the goods. The Walgreens down the street doesn’t add a separate charge to cover its rent, utilities, or the cost of refrigeration units. Basic microeconomics teaches us that companies accept these as “sunk costs” and continue operating as any profit-maximizing firm is supposed to. In the long-run, of course, this changes somewhat; all costs are variable and companies go out of business if they can’t cover their rent or employees’ salaries.
In many respects, a hospital can be thought of as a profit-maximizing firm. Yet, when it comes to sunk costs, hospitals behave much differently than other actors in the economy. When someone comes to a hospital or emergency room, they (or more appropriately, their insurance provider) aren’t just paying physician fees — they’re also paying what’s become known as “facility fees,” which are intended to help offset certain overhead costs. Hospitals defend these fees as necessary to pay for the acquisition and maintenance of expensive technology — such as MRI machines — that lets them perform similarly expensive diagnostics and procedures. The actual facility fee is determined by the actual procedures being conducted.
To be blunt, the sicker the patient is the more money hospitals get from their facility fees.
Controversy over these fees focuses mainly on two points. First, these fees are often high relative to the value of the actual services being provided. A 2005 case best exemplifies this. The Virginia Mason Medical Center charged a patient $1,100 for a 30-second procedure to check if she had fungus under her toe. The facility fee portion of the bill turned out to be a whopping $418, nearly 40 percent of the entire bill.
The second focus of the controversy, and perhaps a more important concern, is the incentives that facility fees create for hospitals to consolidate. Buying up free-standing physician practices allows hospitals to turn these practices into outpatient clinics that, under Medicare’s rules, are allowed to charge facility fees. Moreover, one can imagine that hospitals also face incentives to engage in “upcoding,” an illegal practice that artificially makes patients appear sicker than they are by using higher-costing reimbursement codes. The advent of electronic health records has facilitated upcoding by making it easier for hospitals and physicians to “bundle” more diagnostic codes, making patients appear sicker than they may be and also increasing facility fee payments.
The CMS response to these perverse incentives is to consolidate facility fees into one flat rate for all facilities. The idea behind this is simple and well-intentioned: by paring back some of the exorbitant fees charged by hospitals, Medicare can save some money and stop bad apples from gaming the system. But this isn’t the first time that hospitals’ billing practices have come under scrutiny. Services provided at hospitals are generally reimbursed at much higher rates than the same services provided at free-standing clinics or physicians’ offices. In its 2012 report to Congress, the Medicare Payment Advisory Commission (MedPac) noted that “[t]he combined fees paid for visits to hospital-based practices can be 80 percent greater than rates paid to freestanding practices.”
Indeed, CMS’s proposal is misguided not necessarily because it’s bad policy per se. (That’s not to say that it’s good policy either.) Rather, the proposal doesn’t even begin to address the important and costly financial preference for hospital-based care embedded in Medicare payment rates.
The only reason for higher reimbursements to hospitals over free-standing clinics or independent physicians should be higher quality of provided services. As of yet, evidence supporting this claim appears to be lacking. This isn’t to say, however, that hospitals deliver a poor quality of care, or that the inpatient setting is unnecessary. Indeed, the latest data from MedPac show that important measures of quality continue to improve for hospitals: In-hospital and 30-day post-discharge mortality rates fell for myocardial infarction, congestive heart failure, stroke, and pneumonia.
But broadly, the evidence reviewed by MedPac indicates that the difference in Medicare payment rates itself is resulting in a shift to hospital-based services. Basic services — like evaluation and management (E&M) visits — grew by 7 percent and 8 percent in 2009-10 and 2010-11, respectively; meanwhile, E&M visits at physicians’ offices fell by 1 percent over the same period. And while hospital bed occupancy rates fell from 64 to 62 percent in 2006-10, MedPac reports that hospitals are both investing in new and expensive technology (like robotic surgery) and consolidating. This all points to a health care system becoming ever more hospital intensive, thanks to perverse financial incentives.
Managed Care vs. Fee-For-Service: Thinking About Alternative Payment Reforms
Quality-based payment rates, used commonly in managed care settings, can work as an effective alternative. In a recent analysis, researchers from the Boston Consulting Group found that Medicare Advantage’s (MA) capitated HMOs had the best health care outcomes when compared with Medicare fee-for-service. On the other hand, a 2012 study assessing the effects of Medicare’s Premier Hospital Quality Incentive Demonstration — the precursor to the current Accountable Care Organization (ACO) model under the Affordable Care Act — found little improvement in 30-day mortality rates.
So what are health care researchers to make of these seemingly conflicting data? The first thing to remember is that MA is a very different animal than Medicare fee-for-service. The managed care approaches and quality-focused payments under MA were (and are) pioneered by firms in the private sector. (The MA population is less transient than the rest of the insurance market allowing insurers to recoup the high fixed costs managed care.) While the way that payments are calculated for MA insurers can, of course, influence their behavior in unanticipated ways, managed care as practiced by MA is very different than the tightly defined, bureaucratic approach being attempted with Accountable Care Organizations (ACOs).
A variety of problems plague the Pioneer ACO model under the ACA. As Jeff Goldsmith points out, our experience with the ACO model has already made clear that adverse selection (requiring aggressive coding) along with a limited ability to identify high-risk patients likely makes it a non-viable solution to Medicare’s woes. Moreover, the “sticks” (meaning “shared losses”) are relatively weak in the ACO model (as compared to the very blunt, but effective, capitated HMO model in MA), ensuring that any savings to the system as a whole will be limited.
There’s a bigger problem when it comes to the Pioneer ACO model, however – it’s trying to reinvent the wheel. While the research into this area isn’t as robust as we might like, recent literature indicates that MA penetration is associated with lower hospital costs and shorter hospital stays for the fee-for-service and the commercially insured population. So perhaps, if we consider hospitals an expensive source of care relative to other settings, but are still interested in controlling quality, scrapping the ACO model and instead focusing on expanding MA would be a more prudent approach.
Indeed, what’s most distressing about the ACA’s push for ACOs is that it comes at the expense of MA. Under the law, payments to MA plans are reduced by tying the administrative benchmark to FFS costs. Despite these cuts, however, enrollment in MA has continued unimpeded despite projections by CMS and CBO to the contrary. But that doesn’t mean that payment cuts have no effect; according to a recent report by Avalere Health, there will be 142 fewer plans being offered across the country — the first drop since 2011.
The aggregate shift, however, belies an interesting trend – it appears that PPO and other “less managed” plans are being shifted out in favor of more HMO offerings. On the one hand, this might be good news (and unrelated to payment changes) as MA plans move further towards the successful managed care model. At the same time, “special-needs plan” offerings, which cater to those with chronic conditions, are falling 13 percent. The main takeaway, however, is that despite payment cuts MA is still thriving.
Going even further and beyond MA, the core formula used for reimbursement under Medicare – the Resource-Based Relative Value Scale (RBRVS) – is ripe for reform (this is a topic that deserves its own discussion). So while CMS’s proposal isn’t necessarily a bad idea, it only nods in the direction of the underlying problem — the dysfunction of Medicare’s expensive hospital-centric payment system.