In late 1994, the health columnist for the Boston Globe, Betsy Lehman, died as a result of an overdose of the chemotherapeutic agent she was receiving for breast cancer at the renowned Dana-Farber Cancer Institute in Boston. The error involved mistaking the cumulative dose of cisplatin to be given over a total of four days for the daily dose to be given on each of four days, causing Lehman's heart to fail.
Michael L. Millenson, author of Demanding Medical Excellence and a principal at William M. Mercer, says the death of a high-profile patient in such a highly respected institution in a city that's synonymous with high-quality health care triggered a flood of government hearings and meetings and reports, culminating in the Institute of Medicine's estimate in 1999 that medical errors of all kinds result in up to 98,000 deaths annually.
Beyond the human suffering, employers turned a sharp eye to some salient economic factors tied to such mishaps — and that's where the Leapfrog Group comes in.
Leapfrog, sponsored by the Business Roundtable, which consists of the CEOs of many of the nation's largest companies, wants to use meaningful marketplace incentives — more patients, more money, public accolade — to encourage the health care sector to adopt the systemic quality improvement processes widely used in industry. Leapfrog's initial efforts are focused on inpatient safety, an area in which the group believes that evidence-based change can be rapid, meaningful, measurable, and readily appreciated.
As of mid-May, about 80 big-league players representing 25 million health care beneficiaries (and $50 billion in annual health care expenditures) had agreed to play Leapfrog. They include Delta, Ford, GE, IBM, 3M, Merck, PepsiCo, Schering-Plough, SmithKline Beecham, UPS, and Xerox, to name to few. Notable among the newest members are Aetna U.S. Healthcare and Humana.
The Leapfroggers initially are focusing on just a few measures intended to improve the safety of patients in hospitals — using computers to reduce medication prescribing errors, referring patients undergoing certain high-risk procedures to high-volume hospitals (because high volume has been correlated with improved mortality rates), and staffing ICUs with physicians certified in critical care medicine (also correlated with improved mortality rates).
Although hospitals are the focal point of these efforts, HMOs most certainly will be involved. That's because the health care buyers that subscribe to Leapfrog's set of purchasing principles (see table) may delegate the responsibility of dealing with hospitals to the health plans. Here's what's involved in each of the three safety standards:
|Leapfrog Group's Purchasing Principles|
|A variety of techniques are endorsed for employers to get the best quality care and to encourage providers to improve their care.|
|Use comparative ratings||Using NCQA, JCAHO, and other nationally recognized performance data in addition to the Leapfrog safety standards, purchasers compile comparative ratings of major health care providers for employees' use.||Comparative ratings strengthen link between performance and customer's perception of value.|
|Inform/educate employees||Purchasers educate employees about importance of comparative ratings and how to use them to make health care decisions.||For effecting improvement in quality, employees' behavior toward providers is more important than purchasers' selection of providers.|
|Use substantial incentives||Purchasers reward highly rated major providers by using at least two of the following methods:
||Robust market rewards are needed as motivation.|
|Focus on discrete leaps forward in patient safety||
||Improvements in patient safety are likely to win broad support.|
|Hold health plans accountable for Leapfrog implementation||Purchasers may delegate the responsibility of implementing Leapfrog principles to the health plans that serve as their intermediaries.||Having health plans apply Leapfrog principles to their providers can leverage the efforts of purchasers.|
|Encourage support of consultants/brokers||Consultants and brokers are encouraged to apply Leapfrog principles in advising clients and assessing health plans and delivery systems.|
Computerized physician-order entry (CPOE). To fulfill this standard, a hospital must require physicians to enter medication orders using a computer equipped with software to prevent prescribing errors. Using a testing protocol being developed by the Institute for Safe Medication Practices, the software must be able to catch at least 50 percent of common serious prescribing errors. Results of the test must be posted on a web site designated by Leapfrog. In addition, before a prescribing physician can override an error intercepted by the system, the interception must be acknowledged by the physician and documented.
Evidence-based hospital referral. A hospital must meet the volume characteristics listed in "Favorable Hospital Volume Characteristics," (below) to fulfill this standard. However, in regions where publicly reported comparative outcomes data are available, the volume characteristics will be replaced by risk-adjusted outcomes, which are preferable (but seldom available).
|Favorable Hospital-Volume Characteristics|
|Studies made public in recent years indicate that higher volume of any specific treatment is associated with better outcomes.|
|Treatments||Favorable hospital-volume characteristic||Comment|
|Coronary artery bypass graft||>=500/year||Could result in up to 164,261 referrals per year from low-volume hospitals in urban areas, saving 1,486 lives through a 27 percent relative reduction in in-hospital mortality at high-volume hospitals (2.4% vs. 3.3%).|
|Coronary angioplasty||>=400/year||Could result in up to 121,292 referrals per year from low-volume hospitals in urban areas, saving 345 lives through a 25 percent relative reduction in in-hospital mortality at high-volume hospitals (0.9% vs. 1.2%).|
|Abdominal aortic aneurysm repair||>=30/year||Could result in up to 18,534 referrals per year from low-volume hospitals in urban areas, saving 464 lives through a 37 percent relative reduction in in-hospital mortality at high-volume hospitals (4.2% vs. 6.7%).|
|Carotid endarterectomy||>=100/year||Could result in up to 82,544 referrals per year from low-volume hospitals in urban areas, saving 118 lives through a 22 percent relative reduction in in-hospital mortality at high-volume hospitals (0.5% vs. 0.7%).|
|Esophageal cancer surgery||>=7/year||Could result in up to 1,696 referrals per year from low-volume hospitals in urban areas, saving 168 lives through a 63 percent relative reduction in in-hospital mortality at high-volume hospitals (5.9% vs. 15.8%).|
|Delivery with expected birth weight <1,500 grams or gestational age <32 weeks||Regional neonatal ICU with average daily census >=15||Applies in states in which hospital licensing agency designates regional NICUs. Could result in moving up to 26,477 deliveries per year, saving 1,369 lives through a 28 percent relative reduction in mortality at NICUs (13.0% vs. 18.1%).|
|Delivery with prenatal diagnosis of major congenital anomalies||Regional neonatal ICU with average daily census >=15||Applies in states in which hospital licensing agency designates regional NICUs. Could result in moving up to 16,559 deliveries per year, saving 494 lives through a 31 percent relative reduction in mortality at NICUs (6.8% vs. 9.8%).|
|SOURCE: Birkmeyer, John D. Leapfrog Patient Safety Standards: The Potential Benefits of Universal Adoption. November 2000. Available at http://www.leapfroggroup.org/PressEvent/birkmeyer.pdf|
ICU physician staffing. To fulfill this standard, a hospital must have an adult intensive care unit that is managed by a physician who is board-certified or -eligible in critical care medicine and who is present during daytime hours and provides clinical care exclusively in the ICU. In regions where publicly reported comparative outcomes data are available, the staffing standard will be replaced by risk-adjusted outcomes.
All three of these safety standards are supported by studies demonstrating that they can save lives, reduce morbidity, or reduce costs.
"The IOM has flagged some extremely severe safety flaws in the American health care system," says Arnold Milstein, M.D., M.P.H., one of the founders of the Leapfrog Group. Milstein also is the chief physician at William M. Mercer, and the medical director of the Pacific Business Group on Health, the largest regional employer health care coalition. "Addressing these flaws substantially and with due haste is every shareholder's responsibility."
On behalf of the Business Roundtable, Milstein delivered a statement before the Senate Committee on Health, Education, Labor, and Pensions last year to explain why the Business Roundtable was about to launch Leapfrog: At the error rates calculated by the IOM, hospital errors cause one of the Roundtable's member company enrollees to suffer an avoidable death and five to suffer an avoidable disability every hour.
A study conducted at Boston's prestigious Brigham and Women's Hospital found that there were 10.7 nonintercepted serious medication errors per 1,000 patient-days. Some would argue that Brigham and Women's would be expected to have a higher rate of medication errors than the average hospital because its challenging case load requires drug regimens of greater complexity.
The counterargument is that the skill level of the personnel is such that Brigham and Women's would be expected to have a lower rate of errors. If this error rate is more or less accurate, a hospital with 500 occupied beds could expect to have five such errors every day. Not every serious medication error results in injury, but even so, that kind of an error rate is of no small consequence, given that the medical cost per adverse drug event is estimated to exceed $2,000.
However, researchers at Brigham and Women's found that CPOE could reduce serious medication errors by at least 55 percent, resulting in cost savings at that hospital of between $5 million and $10 million annually. Although 32 percent of U.S. hospitals have a CPOE system wholly or partially in place, only 2 percent require physicians to use it.
In other words, 98 percent of U.S. patients are being treated at hospitals that fail to meet Leapfrog's standard, leaving an awful lot of room for improvement. That also calls for a lot of capital investment, because it's estimated to cost upward of $1 million to start a CPOE system and $500,000 annually to maintain it.
"I have little sympathy for hospitals that have paid little attention to the literature on error prevention and now complain that its going to cost them money," protests Millenson. "Will it cost money? Sure. But if they don't spend that money, how can they look themselves in the mirror in the morning?"
Not an easy undertaking
Milstein says hospital executives agree with him about the value of the Leapfrog standards, but they point out that nothing on the list is easily accomplished because the tasks require significant investments, cultural changes, or belt-tightening due to reduced revenues.
He says his greatest challenge has been in trying to garner support for the Leapfrog initiative from managed health care plans, some of which have reached a point where their business strategy points toward less involvement with providers instead of more. "They feel they've been demonized by the media, and so they're pressing now for a more contained role."
However, Leapfrog seeks health plans' collaborative involvement in pursuing the safety standards. After all, the role of intermediary is one that HMOs have rather deliberately carved out for themselves.
In his conversations with physicians, Milstein says, the dominant reaction has been private acknowledgement that the Leapfrog standards make sense, to the extent that these doctors would very much like to have the standards in place if they or their family members happened to be hospital patients.
Even if hospital administrators have their concerns about implementing the Leapfrog standards, Milstein feels certain that the standards enjoy the support of hospitals' quality-improvement personnel.
"My most gratifying experience with Leapfrog so far has been to hear the response from the people who have been laboring in the vineyards of quality improvement in hospitals for many years," Milstein says. "They say thanks for putting a public spotlight on opportunities to improve patient safety. They've been working for these all along, but haven't succeeded because of the lack of money or conflicting priorities. Now Leapfrog is providing the extra push, and the quality improvement leaders within hospitals welcome it."
So, as health plans approach hospitals on behalf of the Leapfrog Group, they're likely to find people sympathetic to Leapfrog's safety standards even if they're dubious about their implementation. Pockets of resistance on philosophical grounds — hospital administrators who think these patient safety issues are none of Leapfrog's business — also will be encountered, says Suzanne Delbanco, Ph.D., Leapfrog's executive director.
"It didn't use to be our business, but now it is," she says. But she's looking forward to the day when, through collaborative efforts and positive reinforcement, the quality of U.S. health care will have improved to the point that Leapfrog's members can concentrate on their core businesses, content in the knowledge that quality improvement has gained the full focus of hospitals nationwide.
In the meantime, get ready to jump.
Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA. 1998;280:1311–1316.
Bates DW. Error in medicine: what have we learned? Ann Intern Med. 2000;132:763–767.
Birkmeyer, John D. Leapfrog Patient Safety Standards: The Potential Benefits of Universal Adoption. November 2000. Available at www.leapfroggroup.org/PressEvent/birkmeyer.pdf.
Millenson, Michael L. Demanding Medical Excellence: Doctors and Accountability in the Information Age. Chicago: University of Chicago Press, 1997.
Milstein A, et al. Improving the safety of health care: the Leapfrog initiative. Eff Clin Pract. 2000;3:313–316.