A front-page article in the April 16, 2013 issue of the New York Times announced that “[h]ospitals make money from their own mistakes because insurers pay them for the longer stays and extra care that patients need to treat surgical complications that could have been prevented.” That conclusion came from a study by Sunil Eappen and his co-authors, published in the April 17 issue of The Journal of the American Medical Association. Eappen and his colleagues were not alone in reporting this finding. Last year, we along with William Weeks came to the same conclusion. The two studies demonstrate that hospitals that set out to compete by reducing complication rates face a negative financial impact, and thus raise a policy challenge: how to create effective incentives to improve hospital quality.
Well before the two articles appeared, hospitals appeared to act as though they lacked incentives to compete on low complication and mortality rates. Consider the following evidence. (1) Fewer than 10 percent of the nation’s hospitals participate in the American College of Surgeons’ National Surgical Quality Program (NSQIP). Yet, the program allows them to compare their risk-adjusted results to their peers’ and to learn from the best practitioners in the program. (2) Fewer than 45 percent of programs that received three stars – statistically above average in a rating scheme based largely on outcomes – for performing coronary artery bypass grafting (CABG) from the Society of Thoracic Surgeons mention the rating on their web sites; even fewer emphasize its significance. (3) Arguably the best known effort to reduce central line associated blood stream infections (CLABSIs) in acute care settings is the development of a standardized protocol for inserting central venous catheters. The protocol, developed and pioneered at Johns Hopkins University in 2001, was subsequently introduced with great success to Michigan statewide at the invitation of its hospitals. Yet, fewer than 5 percent of Michigan hospitals mention their CLABSI rates on their web sites.
Fortunately, a consensus is emerging that improvement in hospital quality and cost will be achieved through a combination of bundled payments and transparency of outcomes and costs. This can be interpreted as evidence that competition will be introduced. As usual, however, the devil is in the details. One of the sticking points is that, as noted above, reductions in surgical complications and other improvements in clinical quality result in empty beds, thus imposing a financial burden on hospitals. As a result, hospitals that deliver the highest value – outcomes divided by price – would appear to suffer most. However, as we showed in our Health Affairs article, hospitals have a positive financial incentive to reduce complications if they can fill the empty beds resulting from the improvement program. Value-based purchasing (VBP), a policy of rewarding hospitals that achieve a target level of performance with bonuses of up to two percent of revenue would provide some incentive. But it’s not clear that it will suffice.
A New Approach: Facilitated Quality Competition
We propose what we believe to be a more effective policy: Facilitated Quality Competition (FQC). It has two requirements: (1) Transparency: Only hospitals that publish their risk-adjusted outcomes and prices for one or more services are eligible to compete for patients seeking such care. (2) Mandate: All payers are mandated to steer patients to services – not hospitals – that deliver average or better-than-average value.
Our policy recommendation is designed to ensure that patients receive the highest quality of care for a particular episode at a competitive price. Since it is unlikely that the services delivering the highest value will all be delivered at a single hospital in a given region, FQC requires hospitals to publish risk-adjusted outcomes and associated bundled prices for procedures or families of procedures. This provision would introduce real competition. We acknowledge that it would not be feasible to cover all procedures; thus we propose to concentrate on the most common ones for which rating methods could be devised, with the expectation that this would cover the majority of cases. Further, because merely publishing information on quality and price is unlikely to foster competition in the face of other pressures, a mandate is required to ensure that patients are directed to services delivering above average or average quality of care at a competitive price.
Distinguishing FQC from value-based purchasing. Our proposed policy differs from the VBP policy in three distinct dimensions: (1) it grades performance at the service level, rather than at the level of the hospital; (2) performance is assessed relative to the absolute best risk-adjusted outcomes rather than a somewhat arbitrary level; and (3) it rewards top performing services with more patients rather than rewarding the hospital with an across-the-board bonus.
Our recommendation thus avoids the contentious issue of selecting hospital-wide performance metrics and focuses on the question uppermost in patients’ minds: where do I go to obtain my treatment? By measuring clinical performance relative to the best, it stimulates ongoing competition based on the quality of risk-adjusted outcomes and on prices. We also believe that hospital executives will prefer an incentive that increases patient volume for a particular service to a hospital-wide bonus that would have to be apportioned among services.
Because the policy requires outcome-rating systems that are accepted by specialists, the ratings must not be devised by government agencies or by payers, and they cannot vary by region. Success will thus rest on the cooperation and leadership of medical societies. So far, only the Society of Thoracic Surgeons has risen to this challenge with its three-star rating system for CABG. In addition, the policy requires the development of robust, up-to-date risk-adjustment methodologies, data collection processes that are protected from being gamed, and the development of credible, market-based prices for a procedure or treatment.
Where Would A Facilitated Quality Competition Approach Lead?
Facilitated Quality Competition would lead inevitably to centers of clinical excellence by service, not at the hospital level, and would likely lead to the closing of some facilities. However, we believe – provided that rules are observed – the institution of this Darwinian policy will lead to more rapid improvement in quality than VBP, which, in effect, leaves hospitals to struggle with excess capacity. Incidentally, under FQC a rural hospital would not be disadvantaged: It could pick one or more specialties and become a center of excellence.
The challenges associated with the implementation of this policy are many. Physicians will resist a system with the potential of generating individual ratings; the resistance will add to the challenge faced by medical societies as they undertake the development of quality rating systems. Large hospitals and large hospital systems with significant market power could use that power to challenge the implementation of any policy that might diminish their influence. An even greater barrier may be the establishment of contracts by service line rather that at the hospital level. By comparison, the problems associated with the development of risk-adjustment methodologies and methods to prevent gaming the process of data collection appear to be manageable. Finally, there is the issue of agreeing on a numerical definition of “value.” Should a service awarded a “three-star” quality rating carry a higher price than one with a “two-star” rating? We recommend no premium because the financial reward is implicit in the additional volume.
But, with the exception of service-based contracts, these potential problems arise for any policy, based on bundled payments and transparency and designed to improve clinical outcomes while holding hospital costs in check. Our policy is unique in acknowledging the importance, from a financial perspective, of keeping hospitals full, while recognizing that this benefit should accrue only to institutions that deliver the highest value within an episode of care at the service level. The focus on the delivery of value at the service level introduces the additional benefit of simplifying the allocation of resources of a hospital, and, as noted earlier, of finessing the need to develop hospital-wide performance metrics.
A final benefit: The policy is ready to be tested. We suggest a pilot for CABG, based on the three-star rating system of the Society of Thoracic Surgeons, in a state with a sufficient distribution of ratings, and under the auspices of Medicare. The volume of surgeries is high, the rating system has been publicly available for three years, and there are enough states that could be considered for the pilot study.