UBC Faculty Research and Publications

How do they do that? : what can we learn from Everett, Washington about improving quality and outcomes… McGrail, Kimberlyn Aug 31, 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-McGrail_Kimberlyn_Everett.pdf [ 2.1MB ]
JSON: 52383-1.0048255.json
JSON-LD: 52383-1.0048255-ld.json
RDF/XML (Pretty): 52383-1.0048255-rdf.xml
RDF/JSON: 52383-1.0048255-rdf.json
Turtle: 52383-1.0048255-turtle.txt
N-Triples: 52383-1.0048255-rdf-ntriples.txt
Original Record: 52383-1.0048255-source.json
Full Text

Full Text

How Do They Do That?What can we learn from Everett, Washington about improving quality and outcomes without increasing costs?August 2011Kimberlyn McGrail, PhDHow Do They Do That: What can we learn from Everett, Washington about improving quality and outcomes without increasing costs? was produced by:Centre for Health Services and Policy ResearchUniversity of British Columbia201–2206 East MallVancouver, BC V6T 1Z3Phone: 604-822-4969Email: enquire@chspr.ubc.ca or kmcgrail@chspr.ubc.caYou can download this publication from www.chspr.ubc.caUBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch1About CHSPRThe Centre for Health Services and Policy Research (CHSPR) is an independent research centre based at the University of British Columbia. CHSPR’s mission is to advance scientific enquiry into issues of health in population groups, and ways in which health services can best be organized, funded and delivered. Our researchers carry out a diverse program of applied health services and population health research under this agenda. The Centre’s work is:• Independent• Population-based• Policy relevant• Interdisciplinary• Privacy sensitiveCHSPR aims to contribute to the improvement of population health by ensuring our research is relevant to contemporary health policy concerns and by working closely with decision makers to actively trans-late research findings into policy options. Our researchers are active participants in many policy-making forums and provide advice and assistance to both government and non-government organizations in British Columbia (BC), Canada and abroad. For more information about CHSPR, please visit www.chspr.ubc.ca.Contentsh o w  d o  t h e y  d o  t h a t ?21 About CHSPR3 Introduction5 Methods6 Results6	 Analysis	of	Quantitative	Data6	 Characteristics	of	the	population6	 Characteristics	of	the	health	care	system6	 Employers	7	 Synthesis	of	Interviews7	 Active	management8	 Collaboration9	 Service	area	boundaries	and	the	threat	of	competition11 Discussion11	 The	things	Everett	is	given12	 The	things	Everett	creates14 ReferencesU B C  C E N T R E  F O R  H E A LT H  S E R V I C E S  A N D  P O L I C Y  R E S E A R C H3IntroductionThe Everett Hospital Referral Region (HRR) stretches from just north of Seattle, Washington to just south of Bellingham, 30 miles south of the Canadian border. The region includes about 650,000 people, and its larg-est city, Everett, has a population of just over 100,000. The region comprises smaller cities as well as rural areas and extends east to a large national parks area. There are seven hospitals in the Everett HRR. One of these, the Providence Regional Medical Center Ever-ett, accounts for nearly half of the region’s acute care beds. All of the hospitals are non-profit, and there is a significant influence of large group medical prac-tices in this region. For example, The Everett Clinic has about 300 practicing physicians, and the Skagit Regional Clinics employs over 100 physicians. Since 1992, the first year of data available through the Dartmouth Atlas of Health Care,* the Everett HRR has in all but four years ranked in the lowest-spending half of the 306 regions, based on overall Medicare Part A and B costs per capita. Their spending is always below the national average. In the last two years of data, their rank (where “1” is lowest cost) was number 65 (2006) and number 31 (2007). Quality of care as measured in 2004 through Medicare Compare was good, and has been improving over time. SKAGIT COUNTYAnacortesMount VernonBurlingtonSedro WoolleyOak HarborArlingtonCoupevilleEverettSeattleMonroeNorth MarysvilleStanwoodWHATCOM COUNTYISLANDCOUNTYSAN JUANCOUNTYSNOHOMISH COUNTYEVERETT HOSPITALREFERRAL REGIONCHELANCOUNTYKITSAPCOUNTYJEFFERSONCOUNTYKING COUNTYMarysvilleHospitalEverett HospitalReferral RegionBuilt-up areaSource: DartmouthAtlas of Health Care,US Census Bureau* The data attributed to the Dartmouth Atlas of Health Care can be found in various data sources and publications found at www.dartmouthatlas.org.h o w  d o  t h e y  d o  t h a t ?4Based on this profile, Everett was one of 10 regions asked to participate in the 2009 “How Do They Do That?” conference, convened by the Institute for Healthcare Improvement and others.* The intent of that conference, as the title implies, was to try and identify common features of successful hospital refer-ral regions, to try and ascertain what contributes to their success. The 10 invited regions were not the only ones identified as successful, based on the best data available on both Medicare costs and quality of care. They were selected for representation of different areas of the country, different region sizes, and for (appar-ent) differences in level and amount of integration. What was particularly interesting about these regions is that there was no a priori reason to believe that they should be providing high quality, low cost care. There may be dominant insurers or providers (e.g. the domi-nance of Kaiser in Sacramento), but they were not fully integrated regions. The tenor of the conference was if these regions can do it, perhaps others in the United States could too. If they could, then the current cost and quality problems would be solved. That conference identified some common denomina-tors across these regions, including physician leader-ship, the use of data to monitor and improve quality, and engagement of local communities in shaping and understanding health care services. There was also an identification of culture as an important attribute, but there was no common understanding of how to define culture much less to understand how successful cultures might be developed in other regions. *  For information on this and the follow-up meeting in 2010, see www.ihi.org/offerings/Initiatives/PastStrategicInitiatives/HowWillWeDoThat/Pages/Materials.aspxThe purpose of this paper is to describe an in-depth case study of the Everett HRR. The intent of this case study was to gather more information about how this region operates, and more particularly, what has con-tributed to its status as a lower cost and higher quality region in the United States. UBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch5Methods* See www.hospitalcompare.hhs.gov/?AspxAutoDetectCookieSupport=1This was a single site case study.1 It used mixed methods, including analysis of quantitative informa-tion available through the Dartmouth Atlas of Health Care project, information from the Medicare Hospital Compare quality assessments,* and interviews with key players in the health care system in the Everett HRR. A total of nineteen interviews were conducted between June and September 2010. The interviewees included hospital CEOs, CFOs, a chief nursing officer, managers, direct care providers, insurers, employers and a state health association. Some individuals had more than one role.Most interviews were conducted in person and lasted between 60 and 90 minutes. All but three interviews were recorded and transcribed. Where there were no audio recordings, detailed notes were taken and typed and augmented immediately after the inter-views. Transcripts were reviewed first to identify significant events, such as the start of a new program or service. Further review identified emergent themes both within and across interviews, for example use of data or collaboration. Several themes emerged, and prominence in the synthesis was given to themes that were expressed by more than one person representing more than one organization. The study protocol was reviewed and approved by the University of British Columbia Behavioural Research Ethics Board. h o w  d o  t h e y  d o  t h a t ?6Analysis of Quantitative DataCharacteristics of the populationCounty health profiles produced by the University of Wisconsin suggest that Snohomish County is reason-ably average by both Washington State and national benchmarks.* The Everett HRR does not include all of Snohomish County and is broader than this county, but a large percentage of the HRR’s population is captured in these statistics. The potential years of life lost per thousand population in Snohomish County is slightly lower than the national 90th percentile for counties. Thirteen percent of Snohomish County residents report being in fair or poor health, which is the same as the Washington state average, and slightly higher than the 10% national benchmark. Smoking and obesity, at 18% and 28% respectively, are slightly higher than the national benchmarks of 15% and 25%. The County has a lower percentage of people who re-port completing high school (70% vs. 92% nationally), and the unemployment rate, at 9.5%, is much higher than the 5.3% national benchmark. Characteristics of the health care systemIn 2006 there were 200 physicians per 100,000 popula-tion in the Everett HRR. This puts them in the highest third of all HRRs for overall physician supply.† The region is even closer to the top (greater supply) for primary care physicians, because 40% of their physi-cians are primary care providers; this is in comparison to 36% for the United States as a whole.As noted at the outset, there is consolidation of physi-cian supply in group practices. One of these practices, with 300 physicians, represents nearly a quarter of the physician supply in the region. No statistics exist for overall consolidation, but given the known groups in the region, more than half of all physicians work in a practice of significant size, i.e. more than 50 prac-titioners. Some of group practices are for-profit and physician owned, some are hospital-owned and one is operated by Group Health Cooperative.  In contrast to relative abundance of physicians, the Everett HRR has the lowest supply of acute care beds of all the HRRs, at 1.4 beds per thousand population. This constrained bed supply has been consistent over time, and is likely to continue. More hospital beds  are opening in the region, but population growth is high and is likely at least to keep pace with that increasing supply. Employers There is some consolidation of employers in Everett as well. The Port of Everett and a naval base are both large employers. Neither, however, compares to Boe-ing, which employs about 60,000 people overall in the broader region, including an estimated 30,000 in the Everett HRR. Considering family members and pensioners, this represents 100,000 or more of the  650,000 total HRR lives covered through insurance offered by Boeing. Boeing is not only a large employer, but has used that leverage to effect change in the health care system. They were a founding member of the Leapfrog Group, and have instigated experimentation in health care delivery. One example of this experimentation is the Results*  See www.countyhealthrankings.org/washington/snohomish. Accessed 3 June 2011.†  All reported data are available through the Dartmouth Atlas of Health Care at www.dartmouthatlas.org. Accessed 3 June 2011. UBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch7Intensive Outpatient Care Program undertaken at three different medical groups (one of which was in the Everett HRR) that ran from 2007 to 2009.2 The experiment was run a bit differently at each of the three sites, but in broad terms it was meant to identify the top 15% or so of health care service users, who represented about 80% of costs (these members were part of their non-HMO plan). An intense set of case management and behavioural health services were provided to this group, starting with a lengthy visit with a primary care physician who provided a full assessment of their current condition and care needs. By the end of the program, this group had 20% lower health care costs than a propensity-matched comp-arison group. They reported feeling better physical and mental functioning, and decreased their missed work days by more than half; this was a health care experiment but its effects were felt beyond the health care system. Synthesis of InterviewsThree over-arching themes emerged from the analy-sis of interviews. There is some overlap between and among themes, so the headings should be considered broad categorizations rather than distinct and mutu-ally exclusive designations. Active managementA major sentiment cutting across individuals and organizations is a commitment to continual improve-ment. Some of this appears to be a carry-over from  the experience of managed care in the early to mid-1990s. Some providers described themselves as being “good at” care management, which is conveyed in an active as opposed to reactive approach to providing health care. For example, in 1996 a large multi-specialty physician practice approached a major insurer and negotiated a risk contract around pharmaceutical use. If the clinic exceeded their commitments, they would share in the savings to the insurer. If they fell short, they would have to reimburse the insurer. In the first year of this contract they did, in fact, have to write a check to the insurer. After that, however, they were able to change internal patterns of practice, and as one indication, move their generic use rate from about 38% in 1996 to 82% in 2010. As part of this effort, they no longer saw pharmaceutical sales representatives, and instead hired two clinical pharmacists to work with them in-house. This same clinic was part of the Boeing experiment on an Intensive Outpatient Care Program described above, was one of ten Medicare Physician Group Practice demonstration projects (also a suc-cess),3 and instituted internal protocols for ordering advanced imaging. As one interviewee said, “We  all believe no matter how good we do it, we can do  it better.”One reason for this active management approach seems to be in the future orientation and community commitment expressed by nearly all interviewees. These two features were in fact intertwined; the senti-ment was that there needs to be an affordable and effective health care system for the community in the long term, there is a commitment to being part of that health care system, and therefore there is a desire to make plans and decisions accordingly, even when the resulting decisions might take several years to pay off. One insurer, for example, has created a separate and sequestered pool of money that can be used for experiments in health care delivery. The insurer recog-nized the importance of future focus, but also realized that in the absence of sequestered funds, short-term emergencies or desires for profitability likely would always intervene. h o w  d o  t h e y  d o  t h a t ?8Sometimes this commitment goes even further, as in decisions that have a direct and negative impact on the bottom line. One medical clinic instituted an internal system to control ordering for advanced imaging. This decision cost them $1.5 million in revenue in the first 18 months alone because the group owns imaging equipment. They were willing to control use because they believed such scrutiny was inevitable, and their preference was to put the systems in place themselves rather than waiting for them to be imposed by others. A desire for active management also arises from a sense of stewardship. Several interviewees expressed the sentiment that “with great resources comes great responsibility”. They see health care resources as per-haps not strictly finite, but constrained in some way. They view the efficient use of those resources as part of their responsibility. Altruism was not always the motivation. Some expressed that health care spend-ing increases are not sustainable, and that increasing numbers of people who are without insurance is a mark of failure of the industry overall. Control of cost growth is imperative for the insurance industry as much as for the financial viability of providers and provider organizations. At the same time, eyes were as firmly fixed on improving quality as on controlling costs. As one interviewee said: “Managed care. The motivation was wrong. The motivation was dollar savings, rather than organizing and integrating the delivery of care to serve the community.”CollaborationAll interviewees acknowledged that there was no pos-sibility that they or their organization independently could create significant or long-lasting change in the health care system. Almost all related some incident of failure of policy that resulted from trying to do things without the necessary partnerships, engagement or collaboration. The collaborations discussed took many forms. The largest hospital in the Everett HRR is part of a large, Catholic chain. They discussed the impor-tance of the resources that exist within that chain for many of the things they were able to do. Clinical  quality improvements are seeded often in one or a small number of hospitals, and then resulting best practices are spread to the other hospitals. Even more simple things like monthly teleconferences with coun-terparts in other hospitals were viewed as important and helpful.Even more important were collaborations discussed across providers within the Everett HRR. One of the smaller hospitals in the region, for example, described a collaborative clinical program with the largest hospital. This was viewed as a success because it helped keep some patients in the community, but provided an easy link to more advanced services if they were needed. It also helps with clinical compe-tence, through things like shared case conferences and shared training. Another important connection and collaboration was described as that between physicians and hospitals, and physician engagement generally. In some cases, this results as much in consolidation as collaboration; in 2010, one of the HRR’s hospitals amalgamated a 100-physician multi-specialty medical group. The physicians viewed this as beneficial as it gave them access to “deeper pockets” for the capital acquisi-tion they viewed as necessary to provide care (this included renewal of their existing electronic health record system). The hospital viewed the amalgama-tion as important because it sees a future of bundled payments for hospital and physician services; this sort of integration is viewed as an essential step in being prepared for that change. The integration of physi-UBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch9cians and hospitals was seen to carry some potential risk (e.g. the hospital getting into the management of physicians, with which they have no experience), but as nevertheless necessary.A different form of collaboration and physician en-gagement is the hospitalist program at the hospital in the city of Everett. This program started in 1996 and is given some credit for the region’s control over cost growth. There are two unique aspects to this program, its breadth and its management structure. The hos-pitalist program started as a 24/7 team of internists, and that still exists, in 2010 with 34 FTEs in a 370 hospital. It expanded from there to include intensiv-ists, then started one of the first programs for general surgery hospitalists, and now has weekday coverage from orthopaedics and neurology as well. One benefit of hospitalists is seen as an ability to standardize care pathways, since the number of physicians working within the hospital is smaller. A corollary is that these physicians build relationships with hospital staff, and so become specialists in working within that organi-zational milieu. There is also a belief that hospitalist care improves quality both because of standardization of practice and also because of volumes; a physician working in a hospital might take care of 100 patients with congestive heart failure in a year, while a GP might have 10 such patients. The GP may provide very high quality community-based care, but would be less well-equipped in an acute care setting. The other distinguishing feature of Everett’s hospitalist program is that it is co-managed by the hospital and the collaborating physician group (there is more than one). The hospitalist physicians are, in fact, employ-ees of the physician groups, which are completely independent of the hospital. This was a deliberate and strategic decision on the part of the hospital. “Most hospitals employ their “ists”. We’ve made the choice in most of these ist programs to contract with a major medical group... Part of that is it’s one more opportu-nity to create glue between us and the medical group.” For the medical group, part of the appeal of this ar-rangement is that it ensures that in-hospital care will reflect that culture of the organization that is respon-sible for community-based care.Service area boundaries and the threat of competitionWhile things appear to be working well in Everett at the moment, there remain many concerns about the future. Part of this threat is endemic; it is the result of the Everett HRR not actually being formed as an integrated delivery system. The boundaries are reso-nant but not orienting. The result is a continual need for role definition, for negotiation around collabora-tions and partnerships and the potential for shifting loyalties. As might be expected given the fluid nature of health care, these conversations are not limited to providers and organizations located within the Everett HRR. One smaller hospital, for example, is currently building clinical partnerships with providers in Seattle, which is the on the other side of Everett—that is, patients would have to drive past Everett to get to these services. The outlying hospital views this as a sound decision because they are partnering with a provider they view as capable of delivering the highest quality and most advanced care.The hospital in Everett, however, views this as a direct assault on their “natural” market—why should patients go right by Everett when good care could be had there? Their concern is driven by a simple calculation that they need catchment areas of a certain size (different for different services) in order to justify providing certain technologies like an advanced h o w  d o  t h e y  d o  t h a t ?1 0surgical technique and/or certain clinical specialties. This is both about economics, because there needs to be enough business for any new provider or piece of equipment, and quality, since there is a known volume and outcomes relationship. Their response, if they cannot negotiate a clinical partnership, may be to try to capture primary care in the outlying area. The control of primary care is viewed by all providers as critical, since it is from there that referrals for more intense services are made. In other words, partnership decisions outside the HRR can lead to competition within the HRR, which may then draw some of the catchment away from the smaller hospitals. Things get further complicated when providers from outside the HRR begin to make similar “threats” on “territory”. A Seattle-based hospital is opening up a clinic just south of Everett, and will clearly direct all referred care from that clinic south to Seattle rather than north Everett. This and other similar expansions are viewed with some trepidation. Part of the difficulty here is that none of the interview-ees really viewed the Everett HRR as a “region” in a true sense of the word. There are some natural referral patterns (which is how the region was defined), but there is no intrinsic loyalty or commitment to that region. So on the one hand, a larger entity may make its expansion decisions or strategic plans based on a particular view of its catchment area or market, but that by no means obliges all the providers in that catchment to work with the same assumptions. There is inherent instability. Some of that instability may be what leads to the other themes of active management and collaboration. But another outcome is the con-stant negotiations around and fight for market share. UBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch1 1The main lessons from the Everett HRR are two-fold. One, which is seen as well in other places like Ce-dar Rapids, Iowa and Grand Junction, Colorado, is that innovation and success are possible despite the current system with all its funding intricacies and perverse incentives. There are places that provide high quality and lower cost care in spite of the forces that work against them. The second lesson is that this success is fragile. Health care systems are always works in progress, but there is an additional complexity presented by the lack of a formal responsibility for integration and provision of care for a defined population. The Everett HRR can work collaboratively to provide high quality and cost-effective care for its population (however loosely de-fined that might be), but there is no preventing parts of that alliance from making separate alliances outside the HRR, or from others making incursions in. This leaves us with the task of identifying what fea-tures of the Everett HRR are unique and less poten-tially replicable in other HRRs, and what features might be more broadly generalizable. Our concern is less with what Everett can teach us about Everett, and more what the experiences in this HRR might tell us about the ability of other HRRs to achieve the Triple Aim of higher quality, better population health and controlled cost growth.4 The things Everett is given  In the course of interviews, several individuals identi-fied that both the population of the Everett HRR and the providers working there may be different from other places. Some of this is reflected in the charac-teristics outlined from the county health rankings. This is a small city, town and rural area that is 25 miles north of a major metropolitan area. The area has Discussionstrong working-class roots and is still home to a major manufacturer (Boeing) as well as a port and a large naval base. Perhaps more importantly, people talked about the attraction of the physical environment of the northwest. People who choose to live and work in this area often do so because of the opportunities for skiing, hiking, biking and the proximity to water. Pro-viders who had moved to Everett from other areas of the country talked about a less status-driven attitude and a greater desire for work-life balance. If providers are different in their personal character-istics, they also work in a health care milieu that is somewhat distinct. The Everett HRR has the lowest rate of acute capacity of the 306 HRRs. Interviewees were not aware of their ranking on this measure. At the same time, they were quick to note both that their hospitals tend to work with high occupancy, and that constrained supply in the hospital sector can force col-laboration to ensure appropriate use of services. The other feature that is unique is the existence of long-standing (in some cases decades) and large multi-specialty group practices. There is some evi-dence that larger group practices may provide higher quality care, though the evidence is somewhat mixed and the mechanisms are not well understood.5,6 At the very least, these group practices provide a platform for experimentation, and a sort of countervailing force to the general power of hospitals in the health sector. It is possible to imagine being able to export some but not necessarily all of these attributes. A constrained hospital bed supply is likely as much an accident of history as the result of a planned and deliberate strategy. Similarly the natural beauty and desirabil-ity of a region brings an ability to attract and retain people that some regions do not and likely will not h o w  d o  t h e y  d o  t h a t ?1 2possess. The region likely benefits from being close to the University of Washington, whose medical school is known for its training of high quality primary care physicians. At the same time, the high proportion of family medicine clearly must be at least partially delib-erate in a region so heavily dominated by large group practices. In general, provider consolidation is some-thing that can be encouraged elsewhere. Accountable Care Organizations are one mechanism for doing so. The things Everett createsDon Berwick has often said that “all medical care is local”. The extensive literature on variations in health care services use across regions and the presence of re-gional “signatures”7 suggest the importance of culture in influencing the provision of health care services.8 The culture of medical care in the Everett Hospital Referral Region is characterized by active manage-ment, a willingness to experiment with different forms of care, collaboration within and across providers and provider groups, a future orientation and recognition of potential threats to long-term viability. If we believe that these themes are in fact part of the Everett HRR`s success, then it is important to figure out how to encourage these same sorts of orientations elsewhere. The Institute for Healthcare Improvement is certainly trying to promote an active management approach, and the Medicare and Medicaid Center  for Innovation will help with promotion of experi- mentation in organization and system delivery. The  mechanisms for encouraging multi-stakeholder  Attribute Description Exportability?Population  characteristicsBlue collar, independent, outdoor-oriented ??Physician  characteristicsLess status-oriented?; desire for work-life balance Recent medical graduates  generally express increased desire for work-life balanceSupply  characteristicsLow acute care beds; relatively high  proportion of primary careConstraint vs. restraintProvider  consolidationLarge group practices, increasing hospital employmentSeeing these trends elsewhere; ACOs may encourage moreActive manage-ment and future orientationDesire for improvement, commitment to the premise of the Triple Aim. Focus on longevity  for service to the communityIHI promoting this; Medicare demonstration projectsCollaboration Within and outside organizations Encourage multi-stakeholder collaborations / common pool managementBoundary issues Balance between competition and collabora-tion; need for “catchment” along with constant threats of incursionSame everywhere; needs  resolutionUBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch1 3collaboration and an approach to common pool resource management9 is perhaps less obvious right now, but there are experiments happening now that may help move in this direction.* The boundary  issues are perhaps the most intractable in the current delivery and payment environment. These need  more attention. Given all of this, it may be that the Everett HRR is slightly ahead of the curve. Work-life balance has been important there for some time, but that trend is seen elsewhere now, particularly among younger physi-cians.10 There are decades of experience in Everett with group practices, but those too are increasing elsewhere. In addition, there are signs of increasing collaboration of the sort described in the Everett HRR, at least at the level of employment relationships. More hospitals are reporting employing physicians, and especially speciality physicans.11 One motivation in employing physicians seems to be related to market expansion, precisely as described here. Another is a drive for quality, with the assumption that an em-ployee relationship may assist in getting physicians on board with changes in practice toward more evidence-based medicine.11At least some parts of the Everett HRR culture can be exported. Its orientation to long-term viability and the population’s health are consistent with the Institute for Healthcare Improvement’s Triple Aim of improving population health, improving the experience of care and controlling costs.4 Rather than being exceptional, the Everett HRR may be one of many regions in the United States that are helping to shape the future of health care. * See www.rippelfoundation.org/category/initiatives/more-initiatives/managing-the-health-commons-nobel-prize-winner-studies-health. Accessed 3 June 2011.h o w  d o  t h e y  d o  t h a t ?1 4References1. Yin RK. Case Study Research: Design and Methods (Applied Social Research Meth-ods). Sage Publications, Inc; 2008.2. Milstein A, Kothari PP. Are Higher-Value Care Models Replicable? Health Affairs Blog. 2009. Available at: http://healthaffairs.org/blog/2009/10/20/are-higher-value-care-models-replicable/.3. Iglehart JK. Assessing an ACO prototype—Medicare’s Physician Group Practice Demonstration. N Engl J Med. 2011;364(3):198–200.4. Berwick DM, Nolan TW, Whittington J. The Triple Aim: Care, health, and cost. Health Aff (Millwood). 2008;27(3):759-69.5. Casalino LP. Which type of medical group provides higher-quality care? Ann In-tern Med. 2006;145(11):860-1.6. Mehrotra A, Epstein AM, Rosenthal MB. Do integrated medical groups provide higher-quality medical care than individual practice associations? Ann Intern Med. 2006;145(11):826-33.7. Wennberg JE. Tracking medicine: A researcher’s quest to understand health care. Oxford: Oxford University Press; 2010.8. Fisher ES. Learning to Deliver Better Care. Issues in Science and Technology Online. 2008. Available at: http://www.issues.org/24.3/fisher.html. Accessed February 14, 2011.9. Ostrom E. Building Trust to Solve Commons Dilemmas: Taking Small Steps to Test an Evolving Theory of Collective Action. In: Levin SA, ed. Games, Groups, and the Global Good. Berlin, Heidelberg: Springer-Verlag; 2009: 207-228.10. Crossley TF, Hurley J, Jeon SH. Physician labour supply in Canada: a cohort analy-sis. Health Econ. 2009;18(4):437-56.11. Casalino LP, November EA, Berenson RA, Pham HH. Hospital-physician rela-tions: two tracks and the decline of the voluntary medical staff model. Health Aff (Millwood). 2008;27(5):1305-14.UBC  C e n t r e  f o r  h e a lt h  s e r v i C e s  a nd  p o l i C y  r e s e a r Ch1 5UBC Centre for Health Services and Policy Researchthe University of British Columbia201-2206 east Mallvancouver, B.C. Canada v6t 1Z3tel:  604.822.4969fax:  604.822.5690email: enquire@chspr.ubc.cawww.chspr.ubc.caAdvancing world-class health services and policy research, training and data resources on issues that matter to Canadians


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items