Fostering Rapid Advances in Health Care: Learning from System Demonstrations (2002)lictpliwww.nup.cauJoivnhonkill30901471174/hrinl/R I .hcml, copyr4 Iv 2(l2. 21X)I Thr National Ac:Ocrny nt Nelms>, ull riprN reserved

Fostering Rapid Advances inHealth Care

Learning from System Demonstrations

Committee on Rapid Advance Demonstration Projects:Health Care Finance and Delivery Systems

Janet M. Corrigan, Ann Greiner, Shari M. Erickson, Editors

Board on Health Care Services

INSTITUTE OF MEDICINEOF THE NATIONAL ACADEMIES

THE NATIONAL ACADEMIES PRESSWctsllinoIcl, D.C.www.nap.edu

Fostering Rapid Advanc,3 in HciIth C, r;: ffon Sy51Ln p TD,:mon,ii .ai;uns (2002)http://www.ntm etlutopenhookAtIMM87074/him1i I 7.html, copyripht 2002. 2001 Mc National Academy or St:looms, all aphis rencrvcd

1

The Time for Change Has Come

In March 2001, the Institute of Medicine (IOM) released the report Crossing the Quality Chasm:A New Health System for the 21 s' Century, calling for fundamental change in the health care system(Institute of Medicine, 2001b). Responding to widespread and persistent, systemic shortcomings inhealth care quality, that report challenges the nation to undertake a major redesign of both the healthcare delivery system and the policy environment that shapes it. The recommendations in the QualityChasm report did not come altogether as a surprise. The safety and quality of health care in theUnited States had been brought to the forefront with a renewed sense of urgency starting in 1998through the release of three major reports on the quality of care. The IOM's National Roundtable onHealth Care Quality had concluded that "the burden of harm conveyed by the collective impact of allof our health care quality problems is staggering" (Chassin and Galvin, 1998, p. 1004). The AdvisoryCommission on Consumer Protection and Quality in the Health Care Industry (1998, Chapter 1)4:4111 luir a iittimmil ty improve quality after concluding that 'today in America, there isno guarantee that any individual will receive high-quality care for any particular health problem."And the conclusions of both of these national panels had been supported by the results of an exten-sive literature review conducted by researchers at the RAND Corporation, which encompassed publi-cations in peer-reviewed journals between 1993 and mid-1997 and revealed evidence of systemicquality problems throughout the health care sector (Schuster et al., 1998). Moreover, these findingshad been corroborated by studies that looked in more detail at the treatment of specific diseases (e.g.,cancer) or focused on particular types of quality problems (e.g., errors) (Institute of Medicine, 2000;Institute of Medicine and National Research Council, 1999; Leatherman and McCarthy, 2002).

In an effort to chart a direction for health system improvement, the Quality Chasm report identi-fied six national quality aims: h‘ttlth care should be safe, effc ‘tive, patient-centered, timely, efficient,and equitable (see Box 1-1). These aims address not only the serious quality challenges noted above,but also the need to use resources more wisely.

17

hepli/www.nap aft e:.Fostering Rap' vitncirrit Health Cittc: 1.entlifront Secin nonstrtns (20071

as-AUG/116

BOX 1-1 Quality Aims for the 21st-Century Health Care System

• Safe—avoiding injuries to patients from the care that is intended to help them.

• Effective—providing services based on scientific knowledge to all who couldbenefit and refraining from providing services to those not likely to benefit(avoiding underuse and overuse).

• Patient-centered—providing care that is respectful of and responsive to individualpatient preferences, needs, and values and ensuring that patient values guide allclinical decisions_

• Timely—reducing waits and sometimes harmful delays for both those who receiveand those who give care.

Efficient—avoiding waste, in particular waste of equipment, supplies, ideas. andenergy.

• Equitable—providing care to all who could benefit that does not vary in qualitybecause of personal characteristics such as gender, ethnicity, geographic location,and socioeconomic status.

SOURCE: institute of Medicine (2001b, p. 39-40).

In the 2 years since the release of the Qual-ity Chasm report, the challenges confronting thehealth care system have probably worsened.Overall;' national health spending has increasedas a portion of gross domestic product and isexpected to continue to do so for the remainderof the decade—from 13.2 percent in 2000 toapproximately 17 percent in 2011 (Heffler et al.,2002). Employers are expected to see a 13 to 15percent increase in their health care premiumsin 2002, which will be the sixth straight year ofrising premiums (Alliance for Health Care

fr,r Stutiyir3

System Change, 2001). Medicaid is also experi-encing cost increases—an average of 25 percentover the 2 years between 2000 and 2002(Alliance for Health Care Reform, 2002).

These rising costs, in combination with therecent economic downturn, are expected to havea number of consequences. Increases in employ-ers' health care premiums are likely to result inemployers narrowing benefits and/or shifting alarger portion of costs to workers in the form ofpremiums or copayments. More employees maychoose not to participate in employer-sponsored

plans, and more employers, especially smallbusinesses, may choose not to offer healthinsurance altogether.

Overall the number of uninsured people inthe United States has been increasing for morethan a decade—about one in six Americans iswithout coverage today (Institute of Medicine,2001a). The uninsured do not receive the heathservices they need, and this gap has serioushealth, financial, and other consequences forboth the uninsured individuals and their families(Institute of Medicine, 2002a, 2002b). More-over, the growing numbers of uninsured placeincreased demand on public hospitals, academichealth centers, community health centers, andother safety net providers that offer a sizableproportion of services to those who lack healthinsurance and cannot afford to pay.

There are also serious inequities in healthcare. A significant body of research revealsdisturbing disparities in health care access andquality, especially for racial and ethnic minori-ties (Institute of Medicine, 2002c). Minoritiesreceive a lower quality of health care than non-

18

Fostering Rapid Advances in Health Care: Learning from System DeTirtiolt 2002) ,shrm://www.u.5%,‘IMpotonha1090117074/html/19.html. CrTrighl 20o2, 2001 Mc Narmil A.:4144114y e..„ htnix for Chair-Ye Has Come

minorities, even after controling for such factorsas insurance status and income.

The Quality Chasm report calls for changesat four levels—patient experiences, small-practice settings or microsystems that delivercare (e.g., provider groups, multidisciplinaryteams), health care organizations that house themicrosystems (e.g., hospitals), and the healthcare environment (e.g., payment policies, legalliability, regulatory processes) (Berwick, 2002).There is little doubt that change of this magni-tude will be difficult to accomplish, but it isimperative that the process begin. This reportsets forth a strategy for health system reform inwhich states are used as laboratories for thedesign, implementation, and testing of alterna-tive redesign strategies. The set of demonstra-tions called for by this strategy addresses criti-cal leverage points at each of the above fourlevels.

ORIGINS OF THIS REPORT

The disturbing trends in health care summa-rized above have not gone unnoticed by healthcare leaders. In June 2002, the Secretary ofHealth find Human Services met with represen-tatives. of The National Academies andexpressed his concerns about the need to reversethese trends. It was agreed that workable solu-tions must be found quickly. Almost immedi-ately, the TOM initiated a fast-track study withthe objective of identifying interventions andapproaches that showed promise for solving keyproblems, and recommending a set of demon-stration projects in test the-.^ solution:. m:Secretary expressed a strong interest in demon-stration projects that might be conducted incollaboration with states starting in 2003.

To conduct this study, the IOM establishedthe Committee on Rapid Advance Demonstra-tions in June 2002. The committee began bydeveloping a set of criteria for use in selectingpotential demonstration projects. Workinggroups for each of the five categories of demon-strations (enumerated below) were thenconvened to delineate the specifics of the poten-tial demonstration projects. The full committee

then met to finalize the set of proposed demon-strations.

CRITERIA FOR SELECTION OFDEMONSTRATIONS

The committee went through a multi-stepprocess to identify potential demonstrationprojects. Each committee member was asked toidentify potential demonstration categories.These categories were then discussed with over-lapping or related areas being combined, result-ing in a list of seven categories. Small workinggroups were formed to develop detailed descrip-tions of these seven categories. The fullcommittee then discussed the seven categoriesfurther and narrowed the list to five. Categoriesthat were considered, but not selected, are dis-cussed later in this chapter.

The committee concluded that the demon-stration projects as a set, and individually ifpossible, must be bold and transformational.Recognizing the gravity of the problemsconfronting the health care sector, as well as theneed for a major redesign of health care proc-esses, the committee focused on projects thatwould address the fundamental building blocksof the health care system.

To guide its work, the committee generateda list of criteria encompassing factors thatwould lead to a successful demonstration initia-tive (see Box 1-2). These criteria fall into twocategories: those related to the intended resultsof demonstrations and those related to the likeli-hood of successful implementation.

The demonstration projects are intended toproduce four results:

• Improved health status for patients andpopulations—The health care system of the21' century should the hc..41th andfunctioning of both individual patients andcommunities. To accomplish this goal, thesystem should balance and integrate needsfor personal health care with broadercommunity-wide initiatives that target the

n

Box 1-2 Criteria for SelectingBold and Transformational

Demonstrations

Criteria related to Intended resultsof demonstrations

• improved health status of patientsand populations

• System improvements

• Reduced waste

• Stimulus for continued innovation

Criteria related to likelihood ofsuccessful implementation

• Resonates with public and policymakers

• Broad base of support

• Recognizes and addresses barri-ers

• Builds on existing competencies

n

Fostering Rai,'hap //,,,k,o,, flap call n

vancowilp Health Care: 1.eawire from SvctcinWnonstris (2002)4,4frienitorange,tirasom

entire population (e.g., prevention initiativesto address obesity). The health care systemmust have well-defined processes formaking the best use of limited resources.

• System improvements—In the 20 th century,"bricks and mortar" constituted the basicinfrastructure of the health care deliverysystem. To deliver care in the 21' century,the system must have a health informationand conuiamicatio.ss technology (IC.-.T)infrastructure that is accessible to allpatients and providers. Over the past severaldecades, the health care needs of the popu-lation have been shifting from acute tochronic care (The Robert Wood JohnsonFoundation, 1996). Although infectiousdiseases and acute care are still important,the vast majority of health care resourcesare now devoted to the ongoing manage-ment of chronic conditions. The processesused by the health system must be redes-igned to emphasize the prevention and

ongoing management of such conditions,and this redesign will require integrationacross sites of care and more sophisticatedinterfaces between the health care andsocial service sectors. Ready access to elec-tronic medical records will be essential aswell.

• Reduced waste—The 20th-century healthcare system is extremely wasteful, charac-terized both by clinical waste (e.g., unneces-sary procedures, redundant laboratory tests)and administrative waste (e.g., compliancewith the requirements of multiple insuranceprograms, which have not been standard-ized). Waste in the system must be reducedso resources can be rechanneled to meet theneeds of patients and populations.

• Stimulus for continued innovation—The21 st-century health care system must havethe built-in capacity to continuously changeand accommodate innovations in knowl-edge and technology.

The change process will not be easy, andthe demonstrations must be able to withstandmany challenges. In identifying promisingdemonstration projects, then, attention must bepaid to implementation issues, including theneed to:

• Resonate with the public and policymakers—The demonstration projects mustbe understandable to the lay public and pol-icy makers and mllst address their immedi-

ate concerns. The demonstrations should bestructured to produce some tangible resultsin the short run.

• Develop a broad base of support—While astart-up investment may be necessary toassist in initiating change, most demonstra-tions should be budget neutral to the federalgovernment over the long term or at leastbudget conscious. Careful thought shouldbe given to the benefits and costs of thedemonstrations to each of the major stake-holders, including patients, payers, and

20

Fostering Rapid Advances in Health Care: Learning from System DeT Ttroi.2.002)imp://www.nay.hilopenhook/1130087074/him1/21.1xml. eopyripM 7002. 2001 The National Academy ( lewlitne for Change Has Come

providers. Financial and other incentivesshould be offered to key stakeholders,recognizing that major change is difficult toinitiate and to sustain over long periods oftime. Both the public sector (i.e., federaland state governments) and the private sec-tor (e.g., philanthropic foundations) shouldprovide up-front support for the conduct ofthe demonstrations.

• Recognize and address barriers—Therewill be many barriers to change—political,cultural, organizational, regulatory, and oth-ers. To be successful, demonstrations mustidentify and eliminate (or at least mitigate)these barriers.

• Build on existing competencies—There isno time to lose. The set of demonstrationprojects initiated in 2003 should producethe building blocks of a model 21st-centurycommunity health care system by 2006. TheDepartment of Health and Human Servicesshould select demonstration sites that have ahigh likelihood of making rapid progress.

The committee identified five major catego-ries of: demonstrations--chronic care, primarycare, ICT infrastructure, state health insurance,and liability. These demonstration categories arediscussed in turn in Chapters 2 through 6. Foreach category, multiple demonstration projectsor sites are proposed for two reasons. First,within any given category, there would likely bea good deal of variability in design characteris-tics, which in turn will influence the likelihoodof ICTdemonstration project in a predominantly ruralstate would likely have different characteristicsthan one in a large metropolitan area. Much canbe learned from assessing the variability indesign characteristics across different types ofdemonstration sites, and the effects of differentdesigns on impact. Second, a sizable number ofsites will be needed for this strategy to begin tohave a measurable impact on the health systemoverall.

SUPPORTING AND EVALUATINGTHE DEMONSTRATIONS

As the demonstrations are launched, theremust be comprehensive parallel efforts tosupport exchange among organizations under-taking the projects within a given demonstra-tion, to evaluate the effectiveness of theapproaches and interventions being practiced,and to broadly disseminate best practices thusidentified. Such efforts are critical so that thedemonstrations can achieve their full potential,and those that show the most promise can berapidly replicated across the country.

The committee believes that learningcollaboratives are the best mechanism forproviding support for the demonstrations, andthat such collaboratives should be formed foreach of the five areas enumerated above. Thelearning collaboratives would be modeled aftersimilar efforts at both the national and statelevels, in which provider organizations havedefined common goals and related performancemeasures and collaborated successfully—exchanging ideas and information—to improveclinical care for patients with diabetes, heartdisease, and other conditions (Institute forHealthcare Improvement, 2002; Oswald, 2002).In the process, these organizations havesuccessfully reengineered delivery systems tomeet their quality improvement targets. Thesedemonstration-specific collaborative s—whichwould exist virtually but would need some staffsupport—would be created by various organiza-tions, depending upon interest and existingcapacity. For example, the Health Resourcesand Services Administration might take respon-sibility for establishing the primary carecollaborative, and the Centers for Medicare andMedicaid Services the chronic care collabora-tive. Of course, either or both agencies mightchoose to conduct the collaborative directly orto contract with a private-sector organization.

In addition to the learning collaboratives,the committee believes there needs to be anational evaluation and dissemination effort thatwould span all five demonstration categoriesand would include an advisory council withrepresentatives from each of the areas. Given

L1

Fostering Rap.http //wwa, nap c.lulop

vancevidt Health

rG̀CIx

'lnc

'CCl7 "iJ 1ome i 2002 1

the previous, related work of the Agency forHealthcare Quality and Research (AHRQ), itwould be logical for this agency to take the leadin creating and nurturing such an effort. Therewould need to be adequate support to carry outthis critical activity. Planning for the evaluationshould begin at the same time as planning forthe demonstrations. The criteria, performancemeasures, and data to be used in assessing pro-gress must be defined in advance. Thoseinvolved in the effort would, over time, rigor-ously review quantitative and qualitativeperformance data from all of the demonstrationsto assess effectiveness, and then extensivelydisseminate the best practices identified. Theyalso would be able to discern how the fivedemonstration categories—potential buildingblocks for a reformed health care system—might fit together in the future. In addition, theywould be well poised to identify the specificenvironmental obstacles that need to beaddressed if demonstrations that prove success-ful are to be replicated on a larger scale.

Learning Collaboratives

As the demonstrations were being designedand initiated, the learning collaboratives wouldplay an important supporting role in enablingthe sharing of information about strategies,

tools, and techniques (see Box 1-3). Such ar-rangements allow implementing organizationsto benefit from the creativity and experiences ofothers, help guard against reinventing the wheel,and foster continuous learning. Learningcollaboratives rely on regular contact, mainlyelectronic, and regular reporting of agreed-uponperformance measures and qualitative progressreports. The collaboratives for these demonstra-tions would also provide informal and, to alesser degree, formal technical assistance to theprojects.

Once performance could be assessed, thecollaboratives would provide a venue fordiscussions about what does and does not work,generating information necessary for midcoursecorrections. This kind of transparency andaccountability across the demonstration organi-zations could help foster a culture of change in ahealth care system that has firmly entrenchedinterests and has over the decades stubbornlyresisted reform.

Evaluation and Dissemination

A critical step, whether carried out byAHRQ or another organization, is to identify upfront what would constitute success in each ofthe five demonstration categories and to trans-

Box 1-3 Components of a Learning Collaborative

• Multiple organizations that make a commitment, signed by the top leader, toachieve measurable improvements on a given set of metrics and to support staff intheir efforts to do so

• Shared goats and related performance measures

• Education and training on how a collaborative functions

• A coordinated, supportive network through which demonstration project staff canactively learn from each other

• Informal coaching for project staff

• Reporting of performance measures back to the collaborative and to the largerevaluation effort

SOURCE: Adapted from Oswald (2002).

22

Fostering Rapid Advances in Health Care: Learning from System Delorhatioh002)http:llsvww.nap.curopcntiook/031190N7074/html/23.hrml. copyright 201)2. 2001 'ITCNational & grimly c • c , eilTitne for Change Has Come

late these ideas into quantifiable measures andassociated data requirements. This effort isimportant because limited documentation existson approaches that represent alternatives to thetraditional ways in which care is delivered andfinanced. With such measures, a rigorousevaluation can be performed, including, wherepossible, a business case and economic analysis.This business case would help determinewhether the demonstration benefits—as meas-ured by clinical quality indicators and othermeasures—outweigh the costs, after accountingfor up-front investment, particularly in the caseof the ICT infrastructure demonstrations. It isessential to identify the interventions that areand are not successful and to understand whatfactors contributed to their success or failure.Such an evaluation can go a long way towardconvincing powerful stakeholders about whyand how they need to change.

The evaluative measures should helpprovide a strategic focus for the participatingorganizations that emphasizes the objectives ofenhancing quality of care and reducing waste.To the extent possible, these clinical measuresshould be aligned with the process and outcomemeasures included in the National Health CareQuality Report, which is to be published byAHRQ in September 2003. As a conceptualframework, the National Health Care QualityReport will use the six quality aims enumeratedearlier (i.e., safety, effectiveness, patient-centeredness, timeliness, efficiency, and equity).Specific measures falling into one or more ofthese domains have been selected—for exam-ple, the percentage of diabetics with hemoglo-bin Ale under control and the percentage ofheart failure patients prescribed an angiotensin-converting enzyme (ACE) inhibitor at dischargeare measures of effectiveness (Agency forHealthcare Research and Quality, 2002). AHRQis also developing the National DisparitiesReport, and measures from this report might behighly useful in ihs:.“—Jing efforts to addressracial, ethnic, and geographic disparities. Ifdemonstration sites apply some or all of thesame measures, it will be possible to gauge theirprogress in comparison with that of the nationas a whole.

At the close of the demonstrations, when itis clear which approaches and interventionshave yielded best practices and on what specificdimensions, it will be time to get the word outto the broader community. The information dis-seminated should include all the documentationand analysis generated over the course of aproject, including costs incurred, gains realized(particularly in the clinical realm), and opera-tional issues confronted and overcome.

This would also be an appropriate time toidentify environmental obstacles that must beconfronted for best practices to take hold,including those that cut across a number ofdifferent demonstrations and therefore necessi-tate priority action. It is clear that future wide-scale implementation of the best practicesresulting from the demonstrations will requiremore than Medicaid waivers, Medicare demon-stration authority, or communities and statesthat are uniquely supportive of a given demon-stration.

Those individuals involved in evaluatingand disseminating demonstrated best practiceswill have an important vantage point. They willunderstand not only which of the demonstrationbuilding blocks are effective, but also how tocombine them into a more comprehensive,synergistic reform model. They will understandwhere gaps exist and how to fill them. Finally,they will have detailed knowledge about envi-ronmental obstacles that need to be overcomeand areas in which new ground rules need to bearticulated for the seeds of the successfuldemonstrations to be sown and to take hold

pro th,.; LJunt.1,, tikeof the health care system in the process.

OTHER POSSIBLEDEMONSTRATION AREAS

The committee believes that the fivedemonstration categories enumerated above rep-resent a reasonable starting point from which tostimulate fundamental change in the healthsystem, but they are not the only promisingareas. The following are summaries of the twoother areas that were seriously considered but

n

Fostering Rap' vaneolto Health Cr: I A;tiir on' Svstcinhrtp://www.nap.cilu3o n or, . ange„ aogrnos (2l&

not selected, which may also represent goodcandidates:

• Making America's hospitals safe and effec-tive and a decent place to work—Many ifnot most of the country's hospitals werebuilt decades ago. Since that time, a greatdeal has been learned about how best todesign work environments to promotepatient and worker safety and improve effi-ciency. There have been many advances ininformation technology and medical devicesthat have specific space and other physicalrequirements. There have also been innova-tions in architectural design that result inenvironments more conducive to the provi-sion of patient-centered care and workforcesatisfaction. It should be noted that althoughthe committee did not ultimately choose thiscategory, some of the categories selected—particularly ICT infrastructure—could welllead to improvements in hospital care andenvironments.

• Evidence-based, paitem-centered pharmacynt.:.Gagemeni—iliodication., both prescrip-tion: and over-the-counter, represent one ofthe. fastest-growing components of heathcare services. Safety is a serious concern,with many suffering preventable adversedrug events that could have been avoidedthrough the use of computerized medicationorder entry systems (Bates et al., 1999).Cost is a major issue, given that Medicareand some other insurance plans providelittle or no insurance coverage for prescrip-tion drugs. Numerous options exist forpromoting evidence-based prescribing ofmedications and improving efficiency.Although pharmacy management was notselected as one of the five categories, thecommittee believes that projects in some ofthe selected categories—including chroniccare, primary care, and ICT infrastructure—will have a highly positive impact in thisarea.

REFERENCES

Advisory Commission on Consumer Protection andQuality in the Health Care Industry. 1998."Quality First: Better Health Care for AllAmericans." Online. Available at http://www.hcqualitycommission.gov/finn1/ [accessed Aug."S, 2002].

Agency for Healthcare Research and Quality. 2002.National Healthcare Quality Report: PreliminaryMeasure Set.

Alliance for Health Care Reform. 2002. "Health CareCosts and Health Coverage." Online. Availableat http:r7www.allhealt1t.urgipublpdfrosts-Coverage_Aug2002.pdf [accessed Sept. 3,2002].

Bates, D. W., J. M. Teich, J. Lee, D. Seger, G. J. Ku-penman, N. Ma'Luf, D. Boyle, and L. Leape.1999. The Impact of Computerized PhysicianOrder Entry on Medication Error Prevention. JAm Med Inform Assoc 6 (4):313-21.

Berwick, D. M. 2002. A User's Manual for the IOM's'Quality Chasm' Report. Health Aff (Millwood)21 (3):80-90.

Center for Studying Health System Change. 2001."Tracking Health Care Costs: Hospital Care KeyCost Driver in 2000 (Data Bulletin Number 21,September 2001 - revised)." Online. Available athttp://www.hschange.org/CONTENT/380/[accessed Sept. 3, 2002].

Chassin, M. R., and R. W. Galvin. 1998. The UrgentNeed to Improve Health Care Quality. Instituteof Medicine National Roundtable on HealthCare Quality. JAMA 280 (11 ):1000-5.

Hefner, S., S. Smith, G. Won, M. K. Clemens, S.Keeban, and M. Zezza. 2002. Health SpendingProje -ions for ?nO1J.N11: Latcst Outlook.Faster Health Spending Growth and a SlowingEconomy Drive the Health Spending Projectionfor 2001 Up Sharply. Health Aff (Millwood) 21(2):207-18.

Institute for Healthcare Improvement. 2002."Institute for Healthcare Improvement Home-page." Online. Available at http://www.ihi.org/[accessed Oct 8, 2002].

Institute of Medicine. 2000. To Err Is Human: Build-ing a Safer Health System. eds. L. T. Kohn, J.M. Corrigan, and M. S. Donaldson. Washington,D.C.: National Academy Press.

24

Fostering Rapid Advances in Health Care: Learning from System DeT rhop://www.nap.cau/openhook/V309087074/huol/25.1=1, copyright 2002. 2001 The National Acudcmy o cht 'i„ime for Change Has Come

. 2001a. Coverage Matters: Insurance andHealth Care. Washington, D.C.: National Acad-emy Press.

. 2001b. Crossing the Quality Chasm: A NewHealth System for the 21st Century. WashingtonDC: National Academy Press.

. 2002a. Care Without Coverage: Too Little,Too Late. Washington, D.C.: National AcademyPress.

2002b. Health Insurance Is a Family Mat-ter. Washington, D.C.: National Academy Press.

2002c. Unequal Treatment: ConfrontingRacial and Ethnic Disparities in Health Care,eds. B. D. Smedley, A. Y. Stith, and A. R. Nel-son. Washington, D.C.: National AcademyPress.

Institute of Medicine and National Research Council.1999. Ensuring Quality Cancer Care. eds. M.Hewitt and J. V. Simone. Washington, D.C.:National Academy Press.

Leatherman, S. and D. McCarthy. 2002. Quality ofHealth Care in the United States: A Chapbook.New York, NY: The Commonwealth Fund.

Oswald, N. 2002. Structured Collaboratives: Accel-erating Quality Improvement in California.Oakland, CA: California Healthcare Foundation.

Schuster, M. A., E. A. McGlynn, and R. H. Brook.1998. How Good Is the Quality of Health Carein the United States? Milbank Q 76 (4):517-63,509.

The Robert Wood Johnson Foundation. 1996.Chronic Care in America: A 21" Century Chal-lenge. Princeton, NJ: The Robert Wood JohnsonFoundation.

For— ring Rapid Advanccs in Hzalth Care: 1.earning from System De now..trui*Jun (2002)h ttp://aww. nap. c.infor ,o17030')1T47074/Itt nn177.1, htntl, copyr10121)M. II NbIlonal iu...xl.cnty tic lenccs, all r 1p hitt v-."..c.r,

Fostering Rapid Advances in Health Care: Learning from System Demonstrations (2002)hop,t/www.nap.cu/openhook/0309087074/htmlitil.honl, copyright 2002, 2001 'lac National Academy of Scienx.s. all nano: reserved

6Liability: Patient-Centered andSafety-Focused, Nonj udicialCompensation

SUMMARY DESCRIPTION

Demonstrations in this category would create injury compensation systems outside of the court-room that are patient-centered and focused on safety, while also addressing provider concerns aboutrapidly rising liability insurance premiums. Specific characteristics of this non-judicial approach tomalpractice reform are replacing tort liability with alternative systems for ensuring that patients whohave experienced avoidable injuries receive timely, fair compensation from responsible parties; limit-ing financial exposure for health care providers; promoting apology and non-adversarial discussionswith patients; encouraging provider organizations to report and analyze medical errors; rewardingproviders that put in place effective programs for reducing medical injury; and involving patients insai cty impfovement arts.

Prompt federal action to provide incentives for adopting this approach—coupled with appropriatestate legislation—has the potential to produce immediately measurable benefits in terms of provideraccess to affordable, high-dollar liability coverage; gains in administrative efficiency once criteria forcompensation are fully developed, allowing quicker payments to be made to many more injuredpatients; and longer-term improvements in patient safety and stabilization of insurance markets. TheDepartment of HeTilth and Human Service (DHHS) would issue a Request for Proposals (RFP) tothe states and select four to five of those that apply for demonstration projects in this category. Thedemonstrations would be designed to ascertain a reform's effect on the number and nature of claimsfiled and associated total costs, as well as to permit comparison of claim and cost information acrossall the demonstrations. All of these components would be part of the overall evaluation of the demon-strations. If successful, the reforms could continue indefinitely.

Fostering Rapt Advatia,siiii fit.ymith Care: Learning from Syst:rn Demonstrations (2002)hopii/www.liapt.11.0 , 1 intml.copyriOt 20112, 21E1 TI c National Acialcmy nl N,iicnix.s. all right, ocsorml

The committee suggests that participatingstates implement one of two options on ademonstration basis:

• Option 1: Provider-Based Early Payment—Under this approach, the federal govern-ment would provide reinsurance on ashared-cost basis to self-insured or experi-ence-rated provider groups that voluntarilyagreed to identify and promptly compensatepatients for avoidable injuries. States wouldprospectively set limits on non-economicdamages, including pain and suffering, foridentifiable classes of avoidable injuries.

• Option 2: Statewide Administrative Resolu-tion—Under this approach, states wouldgrant all health care professionals and facili-ties, however organized, immunity from tortliability (under most circumstances) inexchange for mandatory participation in astate-sponsored, administrative systemestablished to provide compensation topatients who have suffered avoidable inju-ries.

Both options are compatible with the Admini-stration's recent proposal related to liability re-form, which caps noneconomic damages andsupports the concept of "early offers" of com-pensation (U.S. Department of Health and Hu-man Services, 2002). However, both optionsalso differ in certain respects from the Admini-stration's proposal and do not depend on itsenactment.

BACKGROUND

For the first time in nearly 20 years, theUnited States is facing a broad-based crisis inthe availability and affordability of malpracticeliability insurance for physicians, hospitals, andother health care providers. The Americanhealth care system has undergone dramaticchanges since the last malpractice crisis twodecades ago (Abraham and Weiler, 1994; Sage,1997). Reforms to address the current situation

should therefore take into account a number ofnew concerns and constraints: (1) increasedsensitivity among providers and the public tosubstantial rates of medical error and the need toimprove patient safety at a system level;(2) lower margins and reduced provider capac-ity resulting from private- and public-sector costcontainment, which increases vulnerability to"liability shocks"; and (3) organizational andtechnical innovations in health care financingand delivery, including provider integration andconsolidation, that have affected the dynamicsof litigation as well as expanding the range ofpublic policy responses to a liability insurancecrisis.

There is widespread agreement that thecurrent system of tort liability is a poor way toprevent and redress injury resulting from medi-cal error (Bovbjerg et al., 2001). Most instancesof negligence do not give rise to lawsuits, andmost legal claims do not relate to negligent care(Localio et al., 1991). Many injured patients donot know they have suffered an injury resultingfrom error, and those who go through the legalprocess often do not even recover the cost oftheir continued health care (Sloan et al., 1991).A few plaintiffs and their attorneys, however,win large sums that may be disproportionate totheir injuries or unrelated to the defendant'sconduct. Prolonged, adversarial haggling overclaims by plaintiffs' attorneys and liability in-surers alienates both providers and patients, andgenerates legal fees and administrative expensesthat consume more than half the cost of liabilityinsurance premiums (Kakalik and Pace, 1986).

The apparent randomness and delay associ-ated with this pattern of accountability not onlyprevent severely injured patients from receivingprompt, fair compensation, but destabilizeliability insurance markets and attenuate thesignal that liability is supposed to send healthcare providers regarding the need for qualityimprovement. Fear and distrust breed inefficient"defensive medicine," and lead to missedopportunities for information exchange andapology that might avoid lawsuits in the firstplace. Unfavorable economic conditions andcatastrophic events external to the health care

Fostering Rapid Advances in Health Care: I.earning from System Demonstrations (2002)hoivr/www.napedu/openhook/03001417074/himinSli.html, copyright 2002, 2001 National Acadc my nt Scicnets, all rights reserved Liability

system add to the effects of legal uncertainty onliability insurance premiums, particularly forhigh-dollar coverage that depends on globalreinsurance markets.

The shortcomings of the current malpracticesystem therefore come from three directions, allof which have contributed to the present crisis:inefficient and inequitable legal processes forresolving disputes, problematic responses byclinicians to the threat and cost of liability, andvolatile markets for liability insurance.Although some states face greater insurance in-stability than others as the result of differentlegal standards, public expectations, and profes-sional cultures, no state is immune to the threatof service interruptions affecting physicians,hospitals, and other health care providers.

The current liability insurance crisisprovides a compelling case for reform. How-ever, approaches that focus narrowly on reduc-ing the number and value of legal claims (e.g.,limiting damage awards) may lower liabilityinsurance premiums but do nothing to improvepatient safety or produce prompter and fairercompensation for patients who are injured. Thesystemic problems noted by the Institute ofMedicine (IOM) in To Err Is Human and Cross-ing the Quality Chasm (Institute of Medicine,2000, 2001) strongly suggests the need to createa legal environment that both fosters high-quality patient care and relieves financial strainand administrative burden for health care pro-viders. The committee believes that replacingtort liability with a system of patient-centeredand safety-focused non-judicial compensation—

claim - resolution to orpinizistion-lia:;ederror disclosure and safety improvement proc-esses—can best accomplish these goals. Suchsystems would cap providers' financial expo-sure at reasonable levels, both directly by limit-ing damages and indirectly by providing afford-able umbrella coverage. They would alsoencourage and oversee health care organiza-tions' efforts to identify, compensate, andreduce errors in cooperation with patients. Incombination, these improvements shouldenhance patient safety and enable a greaternumber of patients with valid claims to receive

compensation, while simultaneously stabilizingliability insurance markets by decreasing theunpredictability associated with high-dollar,outlier cases (Studdert and Brennan, 2001a;2001b).

Approaches intended to compensate moreinjured patients by using a standard of"avoidability" rather than the narrower tort stan-dard of "negligence" raise appropriate concernsabout increased cost. Rigorously testing suchsystems on a demonstration basis would allowpolicy makers to determine the total cost ofcompensating medical injuries outside thecourtroom. Further, by gathering, analyzing,and comparing claims and cost data across par-ticipating states, policy makers would gaininsight into how definitions of avoidable injuryand the generosity of the compensation pack-ages selected influence total cost. Finally, policymakers would obtain important informationabout the possible dollar benefits of reducingthe incidence of avoidable injuries.

Through the demonstrations in this cate-gory, states would have the opportunity—andthe incentive—to select one of the two non-judicial claims resolution options outlinedabove. All participating states will refine thetechnical and scientific underpinnings of such asystem through an expert or participatory proc-ess, depending upon the state's preference.States would build on well-developed, butuntested proposals such as "ACEs" ("avoidableclasses of events," also called "accelerated com-pensation events"), early offers of settlement,and scheduled ranges of allowable damages forpain and sufferings. ACTS i:Rntify, in i-,ientifi-cally rigorous fashion, situations in which inju-ries that typically are preventable occur, such asgiving a patient two drugs that are known to in-teract (Tancredi and Bovbjerg, 1991). Earlyoffer systems protect defendants from additionalliability if they reliably and promptly acknowl-edge problems and offer fair compensation(O'Connell, 1982). Damage schedules ascertainreasonable levels of compensation for pain andsuffering on the basis of jury awards for injuriesof defined severity and cap damages at those

Qt1

Fostering Rapt AdvBolia ealth Care: Learning from System Demonstrations (2002)h„„:„,,,,,,,...,,,...,,, ... la 7 1,1 •4.html. o . , right 21X)2. 2(a) The National Academy of Selcncr4. all rights perceived

amounts rather than imposing a one-size-fits-alllimit (Bovbjerg et al., 1989).

States would need to create centralizedmechanisms to ensure the identification, disclo-sure, and analysis of avoidable injuries, as wellas voluntary, confidential reporting of "nearmisses." The way in which such mechanismswould be operationalized will be left up to thestates, necessitating resolution of importantpolicy issues. The federal government mightplay a role by helping to develop consistentdefinitions and data reporting standards, therebyreducing the need for each state to reinvent thewheel and allowing for comparisons acrossstates. States also would help health careproviders communicate more effectively, bothinternally and with patients, when errors occurby encouraging apology and the use of facili-tated discussion procedures such as mediation(Cohen, 2000; Sage, 2002). Finally, stateswould engage in sustained efforts to educate thepublic with respect to the trade-off involved inreplacing tort liability with administrative reme-dies for avoidable medical injury: faster, fairer,surer compensation but forgoing a jury trial.

Some states might choose to phase in non-judiciarapproaches to compensation, beginningwith selected provider organizations (e.g., hos-pitals, large medical groups, and closed-panelhealth maintenance organizations [HMOs]) thathave demonstrated their willingness and abilityto detect, disclose, and prevent medical errorsand have entered into voluntary contractualagreements with patients that establish the termsof compensation (Havighurst, 1995; O'Connell,1996; Sagc et al., 1994). Other state.: mightwant to move more quickly by establishingcomprehensive state-wide systems of adminis-trative claims resolution with mandatory partici-pation by all health care providers in the state(e.g., physicians, nurse practitioners, hospitals,nursing facilities).

The former approach has the advantage ofbuilding on the IOM's earlier recommendationsregarding the optimal structure and conduct ofhigh-quality health care organizations (Instituteof Medicine, 2000, 2001). Specifically, thisapproach creates incentives for physicians and

hospitals to join together to form well-managedclinical entities that bear primary financialresponsibility for avoidable errors and have themedical know-how to minimize patient injury.The strength of the latter approach is that itgives all health care providers equal, immediateaccess to relief from the current liability crisisand does not &v end upon particular organiza-tional forms (e.g., integrated group practice)that may not be well developed in many juris-dictions.

The workers' compensation system is themost familiar example of substituting adminis-trative claims rc.,ulutioli for tort liability on astate-wide basis (Fishback and Kantor, 2000;Kramer and Briffault, 1991). Experience withworkers' compensation demonstrates that no-fault systems can enhance predictability and im-prove access to compensation. It also confirmsthe importance of maintaining safety incentives(e.g., through self-insurance or experiencerating) and establishing reasonable injurythresholds and clear categories of compensableinjury that reduce waste and discourage fraud.

Non-judicial approaches to compensatingunexpected medical injuries are the norm inNew Zealand, Sweden, and elsewhere(Bovbjerg and Sloan, 1998; Danzon, 1985,2000). Similar programs were debated inten-sively in Colorado and Utah in the mid-1990s,but were not adopted (American College ofPhysicians, 1995; Petersen, 1995). The commit-tee believes the time is now ripe for successfulimplementation of such approaches in theUnited States because of two contributions bythe emerging science of p..:tient safety. Fix hu-man factors engineers have shown that non-punitive approaches encourage the detection ofavoidable injuries and foster systems for con-tinuous improvement, which suggests thatresolving malpractice cases without a determi-nation of fault will help rather than harm quality(Institute of Medicine, 2000). Second, as morehealth care providers accept their responsibilityto disclose errors to patients, capping liability atdefined amounts—an essential attribute of anyaffordable non judicial system—will likely

t

Fostering Rapid Advances in Health Care: Learning from System Demonstrations (2002)Imp://www.nap.edu/openhook/0:109087074/html/85.html, copyright 2002. 21511 The National Academy of Sciences. all :iglus reserved Liability

result in more rather than fewer patients receiv-ing compensation.

GOALS

The demonstration projects in this categorywould have the following goals:

1. Improve the malpractice system for patients

Make compensation for injury morepredictable, timely, and fair.

- Promote honesty, transparency, andtrust in clinician-patient relationships.

- Prevent liability concerns from compro-mising the availability of health careservices for patients.

Put patients and physicians, not lawyersand courtrooms, at the center of areformed system

2. Enhance patient safety

- Promote robust reporting of errors in asafe environment.

- Promote system-level responsibility for• errors through organization-based

financial incentives for improvement,such as self-insurance and experiencerating.

- Involve patients in safety improvementefforts.

3. Maintain access to liability insurance

- Improve predictability of liability costs

- Increase affordability of high-dollarliability coverage.

- Decrease the administrative costs ofresolving disputes.

4. Assess cost impact

- Generate definitive data regarding errorrates, claims rates, compensation costs,and administrative costs under variousstate systems for identifying avoidable

errors and related compensation formu-las.

- Analyze and compare data within andacross states.

DEMONSTRATION ATTRIBUTES

Both liability reform options outlined earlierrequire the following actions by participatingstates, with federal grants for up-front costs andtechnical assistance, as well as waiver authorityif necessary:

• Infrastructure—The state would developand maintain objective indicators of avoid-able errors (ACEs), relying on experts, abroader and more participatory process, or acombination of the two. The state alsowould develop and maintain fair, consistentmethods (schedules) for calculatingeconomic harm and reasonable compensa-tion for pain and suffering. Both the ACEsand the schedules would need to be updatedon a regular basis, with ACE categoriesexpanding over time to encompass the largemajority of avoidable injuries. There alsowould need to be centralized collection ofdata related to the state-level demonstra-tions. To help states in developing ACEsand damage schedules, DHHS should pro-vide support for related grants to the HealthResources and Services Administration(HRSA) or the Agency for Health CareResearch and Quality (AHRQ).

• Legal ci5i1v.amt....t.ti—States would need toauthorize statutory or contractual modifica-tions of tort liability to reflect the terms ofthe option they select, as well as to createclear, narrow exceptions to the malpracticereform (e.g., intentional harm). The statewould also need to protect from legal expo-sure individuals and organizations acting ingood faith to implement the demonstrationapproach (e.g., health plans and employersnegotiating group contracts on behalf ofenrollees). And to make the demonstrationaffordable, states would need to prevent

R5

Fostering Raticlvaglifioillth Care: Learning from System Demonstrations (2002)hniEllwww.nap.cAtuf n ti 17 t copyrialit 2002. 2001111c National Academy Scienecs. all Apin g rotund

health insurers, disability insurers, and otherparties who pay costs incurred by patientssuffering compensable injuries from suinghealth care providers to recover those pay-ments (i.e., barring subrogation claims).Finally, states would need to ensure thatapologies and other systematic communica-tions, such as discu.Auas L‘twcoiiproviders and patients following the occur-rence of an avoidable injury, do not increaseproviders' financial liability or legal expo-sure.

• Patient safety reporting systems—Stateswould establish oversight mechanisms toverify the detection of injuries and disclo-sure to patients. The specifics of thesemechanisms would depend on whether theclaims resolution system operates at thestate-wide or institutional level (i.e., uponthe demonstration option selected). Thesemechanisms would build upon existing statereporting requirements. States would alsoneed to establish mechanisms for collectingand analyzing patient safety data, includingvoluntary, confidential reporting of nearmisses. Federal legislation currently underconsideration by Congress would aid thisprocess if enacted. The collection andreporting of patient safety informationwould need to rely on computer-basedmonitoring systems within health care insti-tutions. With time and experience, thesesystems could be linked to decision supportand knowledge management systems thatwould help prevent errors from occurring inthe first place. Federal technical assistancewould be available for these activities.

• Education—The states would be chargedwith educating the public about the benefitsand costs of liability reform, which offersfaster, fairer, surer compensation on the onehand but requires waiving the right to acourt trial on the other. States would alsoneed to work with the principal stakeholdergroups (e.g., consumer organizations, healthcare providers) to build familiarity with andtrust in the public policy goals of the option

selected, and to allay concerns about theconstraints imposed by the demonstration.

Option 1: Provider-Based EarlyPayments

Eligibility

A variety of health care provider organiza-tions could participate in such a demonstration,including hospitals, academic health centers,large medical groups, closed-panel HMOs, andskilled nursing facilities. Independent physi-cians might decide to affiliate with provider or-ganizations (sometimes called "channeling") inorder to participate in a demonstration and re-ceive liability protection.

Participating Provider Responsibilities

Provider organizations would first need toself-insure their liability risk, or purchaseexperience-rated primary coverage so that theorganization benefits (or not) from how wellthey reduce the number of avoidable injuries.They would also need to inform patients abouttheir participation in the demonstration, provid-ing contractual notice of modified liability(perhaps through payers at the point of healthinsurance enrollment) (Moore and Hoff, 1986;O'Connell and Bryan, 2000-2001). It is likelythat mandatory patient participation as a condi-tion of treatment would best serve the goals ofthe demonstration for states adopting theprovider-based early payment option. However,a 11-ight polith....11, or otherreasons, to allow patients to opt out of thereformed system at the time of health insuranceenrollment or hospital admission (i.e., pre-injury), or might modify the early payment sys-tem in special situations (e.g., emergency care).

Should a specific avoidable event (ACE)cause injury, providers would need to notifypatients promptly; express regret; and tenderpayment for both net economic harm (medicalcare, lost wages, lost domestic production, withcollateral source offset) and capped, schedulednoneconomic harm (pain and suffering). Provid-

Fostering Rapid Advances in Health Care: Learning from System Demonstrations (2002)hopli/www.nap cdu/opcnhook/0309087074/hrm1/87.html. copyright 2002.21N11 ENc National Acodcmy of Scicncts. all riphtx rat:mut Liability

ers also would be permitted, but not required, totender similar payments in other cases of avoid-able medical injury that do not fit into an ACEcategory, thereby gaining immunity from suit.Because providers would be required to complywith external oversight and error reporting obli-gations, states would be able to monitor earlypayment practices to determine the circum-stances under which providers are voluntarilytendering or not tendering early payment. Interms of patient safety, providers would need tooffer patients meaningful opportunities to par-ticipate in efior-N cleitined to recitioe error(e.g., mediated discussions with patient safetycommittees regarding their experiences).

Government Responsibilities

The state would protect participatingprovider organizations from tort liability incases where payment is promptly tendered.States, either directly or through intermediaryorganizations, would ensure that organizationsreliably inform patients in all instances ofavoidable injury. . The federal governmentwould provide reinsurance or other umbrellacoverage as an incentive for provider organiza-tions till+ . self-insure and channel coverage foraffiliated physicians, with shared-risk corridorsor similar safeguards against moral hazard. Thefederal government should consider whetheradditional funds should be made available in theevent that the early payment system is moreexpensive than the current system.

In addition to the general support describedabove, DHHS mi ght provide technical assis-tance to states establishing oversight mecha-nisms for captive insurance companies or risk-retention groups formed by provider organiza-tions. Finally, DHHS would create "safe har-bor" protection under the antikickback statute sothat hospitals and other provider organizationscan offer shared, subsidized liability coverage toaffiliated physicians.

Option 2: Statewide AdministrativeResolution

Eligibility

All licensed health care providers(professional and institutional) within a state (orlarge geographic area within a state) would par-ticipate. States might explore including healthplans and other potentially liable parties as well.

Provider Responsibilities

Providers, along with the state, would notifypatients about the state's modified liabilitysystem and give them related information. Nopre-injury opt-out would be available under thisoption. Providers also would need to set upsystems to detect errors and disclose them topatients and to provide related apologies. Forinjuries to patients that fall within the scope ofthe demonstration, providers or their liabilitycarriers would be responsible for payingamounts determined by the publicly adminis-tered adjudication system in the manner author-ized by that system.

Government Responsibilities

States would need to create a publiclyadministered adjudication system, with eachstate having latitude to determine how it will doso. Key elements of such a system include thefollowing:

• Compensation criteria based on avoidability(e.g., expansion over time of establishedACEs categories)

• A definition of compensation that combinesnet economic harm (medical care, lostwages, lost domestic production, with col-lateral source offset) and capped, schedulednoneconomic harm (pain and suffering)

• Injury thresholds (days of hospitalization,days of disability, total economic loss)

Fostering Rai Advi ce.

calth Care: Learning from System Demonstrations (2002)hrtp://o.ww.nap.cilu/ 1 .hunl, copyright 21512. 2001 Elm National Anadcmy of Soicnixs, ill tiphot retterved

• An administrative system of adjudicationfor determining eligibility and compensa-tion in individual cases

• A consumer and provider appeals mecha-nism

• A multidisciplinary expert panel, includingconsumer representatives to oversee thesystem

DHHS would provide start-up funding forthe administrative adjudication system, with theunderstanding that there will eventually be atransition to a provider surcharge with federalmatching funds. This might be done usingHRSA or AHRQ grants to states in amountssufficient to cover the operating costs of calcu-lating payments and resolving disputes, as wellas initial expenses associated with definingcompensable events and developing damageschedules.

As states develop the infrastructure neededfor the demonstrations, they would need to workwith stakeholder groups, including consumeradvocacy groups, to anticipate and avoid stateconstitutional challenges and other implementa-tion delays. The states would also need to put inplace a system for funding compensation pay-ments that maintains financial incentives forsafety improvement within health care organiza-tions, possibly modeled on workers' compensa-tion systems that segment employers accordingto size and structure into tiers of class-ratedindividual risk, commercially insured(experience-rated) risk, and self-insured risk.Pooling of individual risks might best be han-dled in the long run by a state fund supported byphysician surcharges and administered byprivate entities under contract to the state. Forpracticality, however, demonstration sitesshould encourage liability carriers that currentlyinsure individual clinicians to accept essentiallyall applicants, with discounts for meaningfulpatient safety activities.

Finally, the federal government would guar-antee fiscal neutrality from the state's and itsproviders' perspectives to account for the possi-

bility that a comprehensive system that identi-fies and compensates avoidable injury may bemore expensive than the current patchworksystem of tort litigation. As part of this guaran-tee, appropriate maintenance-of-effort and otherdesign safeguards would need to be in place.

REFERENCES

Abraham, K S , and P. C. Weiler. 1994. EnterpriseMedical Liability and the Evolution of theAmerican Health Care System. Hary Law Rev108 (2):381-436.

American College of Physicians. 1995. Beyond MI-CRA: New Ideas for Liability Reform. Ann In-tern Med 122 (6):466-73.

Bovbjerg, R. R., R. H. Miller, and D. W. Shapiro.2001. Paths to Reducing Medical Injury: Profes-sional Liability and Discipline Vs. PatientSafety--and the Need for a Third Way. J LawMed Ethics 29 (3-4):369-80.

Bovbjerg, R. R., and F. A. Sloan. 1998. No-Fault forMedical Injury: Theory and Evidence. Univer-sity of Cincinnati Law Review 67 (1):53-123.

Bovbjerg, R. R., F. A. Sloan, and J. F. Blumstein.1989. Valuing Life and Limb in Tort: Schedul-ing "Pain and Suffering". Northwestern Univer-sity Law Review 83:908-76.

Cohen, J. R. 2000. Apology and Organizations: Ex-ploring an Example From Medical Practice.Fordham Urban Law Journal 27 (5):1447-82.

Damon, P. M. 1985. Medical Malpractice: Theory,Evidence and Public Policy. Cambridge, MA:Harvard University Press.

. 2000. Liability for Medical Malpractice. InHandbook of Health Economics, IB. Amster-dam: Elsevier.

Fishback, P. V. and S. E. Kantor. 2000. A Prelude tothe Welfare State: The Origins of Workers'Compensation. Chicago, IL: University of Chi-cago Press.

Havighurst, C. C. 1995. Health Care Choices: Pri-vate Contracts As Instruments of Health Reform,Washington, D.C.: AEI Press.

Institute of Medicine. 2000. To Err Is Human: Build-ing a Safer Health System. eds. L. T. Kohn, J.

RR

Fostering Rapid Advances in Health Care: Learning from System Demonstrations (2002)hap://www.nap.cilu/openhoold113090N71174/hrml/89.html. copyright 21102. 2001111c National Academy or Selcnixs. all riphia rocrval Liability

M. Corrigan, and M. S. Donaldson. Washington,D.C.: National Academy Press.

. 2001. Crossing the Quality Chasm: A NewHealth System for the 21st Century. WashingtonDC: National Academy Press.

Kakalik, J. S. and N. M. Pace. 1986. Costs and Com-pensation Paid in Tort Litigation. No. R-3391-ICJ. Santa Monica, CA: RAND Institute.

Kramer, 0. and R. Briffault. 1991. Workers Compen-sation: Strengthening the Social Compact. NewYork, NY: Insurance Information Institute Press.

Locale, A. R., A. G. Lawthers, T. A. Brennan, N.M. Laird, L. E. Hebert, L. M. Peterson, J. P.Newhouse, P. C. Weiler, and H. H. Hiatt. 1991.Relation Between Malpractice Claims and Ad-verse Events Due to Negligence. Results of theHarvard Medical Practice Study III. N Engl JMed 325 (4):245-51.

Moore, W. H., and J. S. Hoff. 1986. H.R. 3084: aMore Rational Compensation System for Medi-cal Malpractice. Law Contemp Probl 49(2):117-24.

O'Connell, J. 1982. Offers That Can't Be Refused:Foreclosure of Personal Injury Claims by Defen-dants' Prompt Tender of Claimants' Net Eco-nomic Losses. Northwestern University LawReview 77:589-632.

. 1986. Nee-No-Fault Remedies for MedicalInjuries: Coordinated Statutory and ContractualAlternatives. Law Contemp Probl 49 (2):125-41.

O'Connell, J., and P. B. Bryan. 2000-2001. MoreHippocrates, Less Hypocrisy: "Early Offers" Asa Means of Implementing the Institute of Medi-cine's Recommendations on Malpractice Law.Journe Health is (1):?3-51.

Petersen, S. K. 1995. No-Fault and Enterprise Liabil-ity: the View From Utah. Ann Intern Med 122(6):462-3.

Sage, W. M. 1997. Enterprise Liability and theEmerging Managed Health Care System. LawContemp Probl 60 (2):159-210.

. 2002. Putting the Patient in Patient Safety:Linking Patient Complaints and MalpracticeRisk. JAMA 287 (22):3003-5.

Sage, W. M., K. E. Hastings, and R. A. Berenson.194. Enterprise Liability for Medical

Malpractice and Health Care Quality Improve-ment. Am J Law Med 20 (1-2):1-28.

Sloan, F. A., R. R. Bovberg, and P. B. Githens. 1991.Insuring Medical Malpractice. New York, NY:Oxford University Press.

Studded, D. M., and T. A. Brennan. 2001a. Towarda Workable Model of "No-Fault" Compensationfor Medical Injury in the United States. Am JLaw Med 27 (2-3):225-52.

. 2001b. No-Fault Compensation for MedicalInjuries: the Prospect for Error Prevention.JAMA 286 (2):217-23.

Tancredi, L. R., and R. R. Bovbjerg. 1991. Rethink-ing Responsibility for Patient Injury: Acceler-ated-Compensation Events, a Malpractice andQuality Reform Ripe for a Test. Law ContempProbl 54 (1-2):Spring 147-77.

U.S. Department of Health and Human Services.2002. "Confronting the New Health Care Crisis:Improving Health Care Quality and LoweringCosts By Fixing Our Medical Liability System."Online. Available at http://aspe.hhs.gov/daltcp/reports/litrefm.pdf [accessed Oct. 1, 20021

ru..41,7ri F.:ipid Advances in R.:alth Care' Learning From Sv.;Ict), Dcnionstr:o.',ons (2oen)hap://witu.nap c,),',44,9vAL1.01)91/010744um1,010.hunl. COOTIVO 2.11111. Acacka4 Scicno.... —I rip hr. nacaul

.1‘