Skip to main content
Journal of General Internal Medicine logoLink to Journal of General Internal Medicine
. 2011 Aug 12;26(11):1371–1373. doi: 10.1007/s11606-011-1817-3

Interval Examination: Regional Transformation of Care Delivery in the Hudson Valley

Susan Sherman Stuard 1,, A John Blair 1
PMCID: PMC3208472  PMID: 21837373

Current efforts to reform health care have put pressure on provider organizations to become more coordinated and accountable; to improve quality while controlling costs; and to begin sharing financial risk. Success in taking on these types of activities requires a relatively sophisticated infrastructure and many policy experts believe that integrated delivery networks (IDNs), such as Geisinger Health System and Intermountain Healthcare, are the type of organizations best equipped to handle the job. However, the vast majority of Americans do not receive their health care within an IDN. In 2006, more than 60% of the nation’s 956 million annual office visits were made to primary care physicians, and 78% of these practices had five or fewer providers.1These statistics indicate that current efforts to overhaul the system, if they are only viable within IDNs, may fall far short of transforming health care for all Americans. The Hudson Valley region in New York, however, provides one example of how many of the tasks seen as essential for health care reform can be accomplished in an open community comprised of smaller, unaffiliated practices, without a formal mechanism for integration.

The Hudson Valley in New York State is a seven county region formed by the valley of the Hudson River, extending from the suburbs of New York City in the south to the city of Albany, New York State’s capital, to the north. The Valley’s health care system is comprised of approximately 4,096 providers, of which 1,765 are primary care. Of these primary care providers, 23% practice solo, 23% in groups of 2–5 physicians, 26% in groups of 6–25 physicians, 13% in groups of 26–100 physicians, and 15% in groups of 100 physicians or more. There are eight federally qualified health centers (FQHCs) in the region. The Hudson Valley is also home to 35 hospitals, of which two are Critical Access Hospitals.

In the late 1980s, private medical practices loosely banded together into the Taconic Independent Practice Association (IPA). By the beginning of the 21st century, the Taconic IPA included 4000 physicians. For 15 years, the IPA assumed partial risk from the Valley’s major health plan, receiving per capita payments from the health plan and paying for primary, specialty, and ancillary care.

In 2000, the IPA came to a major decision: it would no longer assume risk for ambulatory care, but would instead focus on quality improvement and practice transformation of its member practices. The governing board shrunk from over 20 physicians to a nine member board of which five were primary care physicians. The IPA prioritized computerization as its initial focus, with the goal of linking all health facilities—physicians, pharmacies, hospitals, home care agencies and others—through electronic health records (EHRs) at the level of each facility and a health information exchange at the level of the entire Valley that would allow each facility, no matter the brand of the EHR, to talk to every other facility.

In order to accomplish these goals, the IPA formed two separate organizations. MedAllies was created as the technical component to assist with EHR adoption within physician practices. Taconic Health Information Network and Community (THINC) became an independent, not-for-profit, coalition-building organization, bringing together the health plans, hospitals, health centers, primary and specialty care practices, and others to negotiate the implementation of the health information exchange and community-based quality initiatives. The IPA’s chief function became the transformation of its practices into patient-centered medical homes. Separating these functions into three organizations allowed each organization to focus on its unique mission. The combined work of these three organizations has come to be known as the Hudson Valley Initiative, www.hudsonvalleyinitiative.com.

Since 2007, MedAllies catalyzed the adoption of over 670 EHRs for an overall EHR adoption rate of 37% of Taconic IPA physicians (43% among primary care doctors). The campaign emphasized EHR adoption in small practices and rural settings and was supported by THINC with a HEAL 1 grant from the New York State Department of Health. The initial premise and common theme of this regional effort has been use of a dedicated, highly skilled local workforce of practice coaches with deep knowledge of the deployed applications and a thorough understanding of ambulatory practice settings. The MedAllies implementation team includes experts in medical informatics, clinical care, health information management, training methodologies, physician practice management and billing, technical interfaces, and network infrastructure. Many features now making up the federal meaningful use requirements have already been implemented as part of the EHR campaign, including e-prescribing and registries for population-based care.

THINC can be considered the Switzerland of the Hudson Valley, a neutral place where competing physician practices, hospitals, health plans, employers and other stakeholders can leave their organizational hats at the door and figure out ways to work together for mutual benefit. THINC governs the Valley’s health information exchange network to ensure that patient privacy and confidentiality are protected. The health information exchange is first supporting two-way public health reporting between hospitals, physicians and state and local health departments. A community viewer, designed to enable look-up of a clinical summary about a patient, with patient consent, will be a second key function of the exchange and is slated to go live in the winter of 2011. In addition to the community record accessed via the THINC HIE, direct provider to provider communication for care setting transitions and care coordination will be handled by the MedAllies direct network. THINC also sponsors research to advance practice transformation and negotiates with health plans to provide pay-for-performance bonuses and extra payments to practices that qualify as NCQA-recognized patient-centered medical homes and implement targeted care coordination interventions.

For the practice transformation work of the Hudson Valley Initiative, the Initiative’s leaders formulated a step-wise plan analogous to Maslow’s hierarchy of needs. First, implement the EHR. Second, make sure all EHRs talk to each other. Third, assist practices to become NCQA-certified patient-centered medical homes. Next will be to improve care coordination, a goal which is now achievable given the functioning health information exchange and better-organized primary care building blocks. The entire plan works toward the larger goal of higher quality at lower cost.

The major achievement of 2009 to 2010 was the designation by NCQA of 15 primary care practice groups in 67 practice sites as patient-centered medical homes (PCMH). In total, these providers represent 18% of primary care providers in the region. The Hudson Valley Initiative invited early-adopter practices to seek PCMH designation, which requires improvement in a variety of functions such as access, quality, and population management. Because the project hand-selected only motivated practices that already displayed signs of readiness, none of the invited groups declined to participate in the project and all who participated were successful in earning Level 3 designation. The details of NCQA designation can be found at www.ncqa.org. Taconic IPA deployed practice coaches to assist practices to achieve NCQA recognition; the coaches were people with years of experience in primary care and very familiar with the Taconic IPA practices.

The early adopters include medium-large sized practices, small practices, and federally qualified health centers (FQHCs). The 15 practices created a medical council to oversee the process; initially the differences between small and larger practices came to the fore (small practices feeling that larger ones had resources they lacked, larger practices believing that it is easier to make changes when only 1 or 2 physicians need to be convinced). Similar differences were aired between private practices and FQHCs. Over time it became clear that FQHCs, who had participated in improvement collaboratives, could teach improvement methods to the private practices. As the council matured, the diverse practices buried their differences and learned from one another

Two health plans were initially willing to provide financial incentives for practices to become PCMHs. It is critical for as many payers as possible to enter into the PCMH process since additional payments are per capita; only if the majority of a practice’s patients are involved will the payment be sufficient to support practice improvement. Eventually, six health plans including a Medicaid HMO joined in the PCMH process. The plans will gain because the process is being evaluated by a Weill Cornell Medical College research team, which is gathering patient-specific and practice-specific data on costs, quality, patient experience and physician satisfaction. These data will allow health plans to know which practices are successful in keeping down costs by controlling unnecessary hospital and emergency department visits.

Practices achieving Level 3 PCMH designation are receiving reasonable financial incentives. New York Medicaid is paying $6 per patient per month and commercial health plans are paying between $2 to $10 per patient per month for their fully-insured commercial business. In an effort to address anti-trust concerns, each plan has the flexibility to apply its own payment methodology, a single approach was not required for participation. These payments supplement existing practice income from patient visits. Currently the PCMH payment is a capitated fee; a proposal is to move toward a hybrid payment including both the capitated fee to support care coordination activities plus a pay-for-performance bonus. Eventually, the Hudson Valley Initiative hopes to add a shared savings bonus for reducing inappropriate utilization.

Despite the great strides that have been made in EHR adoption, health information exchange, practice transformation and payment realignment, the Hudson Valley Initiative has encountered its share of challenges in pursuing this work. Maintaining the engagement and participation of the health plans in the project has been an area of ongoing focus. While the plans have consistently expressed their conceptual support for the work of the Hudson Valley Initiative, each plan has its own individual perspective on how exactly they will be involved. Each has its own internal mechanisms for processing and sharing data. Each has its own legal counsel weighing in on proposed activities. There is a constant process of working through the specific details of each health plan’s participation and striking a balanced approach that works for all parties involved.

In working on promoting EHR adoption and building the health information exchange, the Hudson Valley has tried to align the community in order increase market power when dealing with technology vendors. However, this effort notwithstanding, the community has seen that it sometimes has relatively little leverage with vendors around issues of interoperability, training resources, and configuration.

Perhaps the most difficult challenge for the Hudson Valley Initiative, however, has been convening the disparate health care stakeholders in the community and encouraging collaboration in the face of competitive forces and disparate incentives. To address this challenge, THINC tries to offer potential collaborators multiple opportunities to get involved. If one project does not appeal to a particular stakeholder, perhaps another will. By maintaining a diverse portfolio of concurrent projects, THINC offers health care organizations multiple on-ramps to participate in its work.

The Taconic IPA, THINC and MedAllies have partnered with Weill Cornell Medical College since 2003 to provide independent evaluation of their initiatives. Weill Cornell has evaluated or is in the process of evaluating several interventions in the Hudson Valley, including electronic health records, electronic prescribing, health information exchange, and the PCMH. Outcomes of interest include quality, safety, utilization and cost, the patient experience, the provider experience, public health, and implications for health policy. Several studies have already been published and many are underway.2,3

Currently, the health plans have opted to stay with the 15 PCMH practices and not spread to the rest of the Hudson Valley; they are waiting to see whether the Weill Cornell evaluation finds that the PCMH practices achieve high quality and low costs compared with control practices.

The Hudson Valley Initiative recently decided to focus on care coordination. This decision was made by the medical council and reinforced by patient and physician satisfaction data, which found some negativity regarding care coordination. In its care coordination campaign, the Initiative has sent RNs to the Geisinger Health System in Pennsylvania to be trained in care management. In collaboration with Geisinger, the initiative plans to expand this work with roll-out of an on-site care manager program to address high-risk, high cost patients in these medical home practices. In addition, practices are receiving training in the Johns Hopkins Guided Care process of care coordination for patients with complex healthcare needs and high medical costs.

The community has also begun preparing itself for future Accountable Care Organization (ACO) activity. With funding from the New York State Health Foundation, THINC has offered a series of educational webinars featuring nationally recognized experts to the Hudson Valley. Each webinar focuses on a specific aspect of establishing or operating an ACO. THINC has also sponsored an ACO workgroup comprised of provider organizations, commercial payers, and legal counsel, who are discussing the key issues for consideration when developing contracts between private health plans and ACOs. Ultimately, the group will issue a white paper to describe their findings and make it available as a community resource.

The Hudson Valley story is unique in being a regional transformation of medical care delivery within a dispersed practice model. Because health information technology was implemented as the foundation for further improvement, primary care practices have many of the tools needed for success in transforming themselves. Many policy thinkers feel—based on considerable evidence—that only large integrated delivery systems can make the changes needed to move primary care into the modern era. Hudson Valley is demonstrating that success is possible with a regional strategy involving virtual integration among multiple independent entities electronically linked.

For communities with a similar provider landscape that include smaller, independent practices, the Hudson Valley offers the possibility that the infrastructure necessary for health care reform can be built outside an IDN. As was the case in Hudson Valley, the unique features of each community will have a significant impact on how the community can best self-organize and support infrastructure development. Community level transformation is an ongoing journey; the most important step is to find willing collaborators and embark with one well-defined project that leverages unique local strengths.

graphic file with name 11606_2011_1817_Figa_HTML.jpgHudson Valley.

graphic file with name 11606_2011_1817_Figb_HTML.jpgLeadership (left to right): Jill Quaresimo, JD, RN, Research, Taconic IPA; Susan Stuard, MBA, Executive Director, THINC; Paul Kaye, MD, Medical Director, Taconic IPA; A. John Blair III, MD, President, Taconic IPA and CEO, MedAllies; Dianne Koval, Senior Vice President of Operations, MedAllies.

Acknowledgements

The authors would like to acknowledge the contributions of Thomas S. Bodenheimer, MD, MPH, Adjunct Professor, University of California, San Francisco, School of Medicine.

Conflict of Interest A. John Blair, MD is the President and CEO of MedAllies, Inc, a for profit technology company based in the Hudson Valley.

Contributor Information

Susan Sherman Stuard, Phone: +1-845-8964726, Email: rlkravitz@ucdavis.edu.

A. John Blair, Phone: +1-845-8960191, Email: jblair@medallies.com.

References

  • 1.Cherry DK, Hing E, Woodwell DA, Rechtsteiner EA. National Ambulatory Medical Care Survey: 2006 summary. National health statistics reports; no 3. Hyattsville, MD: National Center for Health Statistics. 2008. [PubMed]
  • 2.Kaushal R, Kern LM, Barron Y, Quaresimo J, Abramson EL. Electronic prescribing improves medication safety in community-based office practices. J Gen Intern Med. 2010;25:530–536. doi: 10.1007/s11606-009-1238-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Kern LM, Barron Y, Blair AJ, 3rd, et al. Electronic result viewing and quality of care in small group practices. J Gen Intern Med. 2008;23:405–410. doi: 10.1007/s11606-007-0448-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of General Internal Medicine are provided here courtesy of Society of General Internal Medicine

RESOURCES