Reflections on the Five Laws of Integrating Medical and Social Services—21 Years Later

Milbank Quarterly Classics

In 1999, Walter Leutz wrote a seminal article in The Milbank Quarterly on the need for and ways to better integrate medical and social services.1 He formulated five laws and offered three sets of lessons and recommendations. How relevant are the laws, lessons, and recommendations developed over two decades ago to the current challenges of better integrating health and social services? Do they still offer guidance? What has been learned? Are new “laws” needed? I assess these questions for each of the five laws and the three sets of lessons and recommendations relevant to the US experience.

Law Number One—You Can Integrate All of the Services for Some of the People, Some of the Services for All of the People, But You Can’t Integrate All of the Services For All of the People

Simply stated, this law says there are tradeoffs between the breadth and depth of the services that can be provided—and integrated—depending on patients’ needs. The On Lok/PACE program is offered as an example of providing all services (medical, social, housing, etc.) to a targeted group of people. However, which services should everyone receive? In addressing this question, Leutz develops three levels of integration—linkage, coordination, and full integration. Linkage is designed for people with relatively mild or moderate needs that can be addressed in systems that serve the whole population. It is based on referrals and follow-up, such as transitions from the hospital to skilled nursing facility care or to home care or a referral from a primary care physician to a mental health specialist. Coordination involves the use of formal structures and designated positions that ensure that people receive the services they need across two or more systems of care. Case or care managers for patients with multiple chronic illnesses requiring care from both the health care system and the social services sector is an example. Full integration involves the pooling of resources from multiple units and systems of care. The goal is to provide care in all relevant settings using a common record and shared care teams. It is most needed by patients with severe needs that cut across sectors of care. As previously noted, the PACE program, with its use of interdisciplinary teams, is an example.

Read The Milbank Quarterly Classic Article:

Five Laws for Integrating Medical and Social Services: Lessons from the United States and the United Kingdom

Walter M. Leutz

The key to using the three levels of integration is to stratify patients based on their needs. Many patients will need only linkage services, some will require more formal coordination, and relatively few will require full integration. The first law reminds us of the need to set priorities. It causes us to consider the following provocative questions. Is integration universally “good”? Are some systems of care “overly integrated”? Where does integration, in various forms, add value? These questions grow in importance given the need for effective partnerships between health care organizations and those in the social determinants of health sectors (education, housing, transportation, and others). One example is the development of Accountable Communities for Health (ACH) that are designed to improve population health by bringing together the diverse organizations that impact population health through developing shared goals, resources, data exchange/ information, and accountability.2 Restating Law #1 in network theory terms: Figuring out the strength of ties or connections among the organizations involved from loose linkage ties to intermediate coordination ties to full integration ties is likely to be one of the keys to their success.3

Law Number Two—Integration Costs Before It Pays

Integrating services can be expensive. It may require adding staff, retraining existing staff, purchasing data systems, and incurring related infrastructure costs. Further, the costs are incurred upfront and often are based on the hope that there will be a financial return in the future or, at least, the ability to cover the costs. For example, a major deterrent to physician organizations joining the various Centers for Medicare and Medicaid Services (CMS) Accountable Care Organization (ACO) payment models has been the startup costs involved. This is particularly true for smaller practices and those located in rural areas. Recognizing this, CMS has developed the Alternative Investment Model (AIM) program in which advance payments are made for practices to invest in staff, data systems, and related infrastructure. At the community-wide level, a major challenge for ACHs will be maintaining financial viability after the start-up funds go away. Each organization involved in the ACH has its own financial needs and priorities, and each will need to assess the extent to which some of its financial needs can be met through ongoing participation in the ACH. One potential solution to this challenge would be for the individual financial supporters of the organizations involved in the ACH to pool some of their funds to create a common budget for the ACH as a whole.4,5

Law Number Three—Your Integration Is My Fragmentation

Providing health care and social services requires a high degree of interdependence in the work involved. Any effort to achieve better integration of the tasks involved will be disruptive to current workflows. Leutz provides examples of this for primary care physicians in particular. Recent experience in implementing patient-centered medical homes underscores the importance of addressing the changes involved. These include developing expanded roles for nurses and medical assistants; training and motivating primary care physicians to delegate care responsibilities to other qualified members of the health care team; training providers in motivational interviewing; implementing shared decision-making and related patient engagement strategies; integrating behavioral health specialists into the primary care team; adding pharmacists, medical social workers, and nutritionists to the team as needed; improving the functionality of electronic health records; and learning how to effectively use telehealth (phone and video) visits and related digital technologies.6 The end goal is more coordinated, integrated care for the patient, but providers experience all of these changes as disruptive to their established routines, “fragmenting” what they have customarily been trained to do. Thus, not only does integration have a financial cost (see Law #2) but it also has a psychological cost to the health care professionals involved. Financial incentives in the form of new payment models that reward more integrated care, and technical assistance to primary care practices are both needed to deal with the challenge. If the goal is better-integrated care, then we need to pay for it and train for it.

Law Number Four—You Can’t Integrate a Square Peg and a Round Hole

In developing Law #4, Leutz primarily used the differences in payment and types of providers involved in the acute care (square peg) and long-term care (round hole) sectors. Aside from this example, the law today takes on increased importance given the growing recognition of the importance of social determinants of health—housing, education, food insecurity, and related drivers. To the extent that health care organizations and systems are increasingly held accountable for the health of the populations they serve, they benefit from developing relationships with organizations in these other sectors. The differences in goals, nature of the work, technologies used, payment, and the people involved among the sectors are marked. For the most part, the health care sector is about healing (“fixing people”) and often using a vast array of sophisticated technologies and highly trained professionals who earn more money the more care that they provide. The “social determinants of health” sectors differ greatly among themselves on the above dimensions but, most importantly, have distinct and limited goals and objectives related to their own strategic priorities—for example, having students meet educational standards or providing affordable housing. Their agenda is not a “health agenda.” “Health in all” policies and ACHs are efforts to bridge the divide and “square the circle.”7 However, existing experience suggests that these will most frequently involve linking and coordinating mechanisms rather than full integration.

Law Number Five—The One Who Integrates Calls the Tune

Leutz’s fifth law raises the question of why payers of care don’t call the “integration tune.” His response was that integration is largely a provider-professional initiative rather than a payer-public initiative. In brief, open-ended fee-for-service (FFS) reimbursement creates no incentive to better integrate care unless creating a “bundle of care” can be paid for as such. In the United States, this has resulted in the costliest health care system in the world, with highly variable quality of care and some of the poorest health outcomes. A major question is the extent to which public (CMS) and private (commercial insurance) payers are moving away from FFS payment to value-based payment models based on risk-adjusted prepayment. This creates incentives for providers to keep people well, and when people become ill, to see that their care is well-coordinated and integrated using patient navigators, care managers, and interdisciplinary care teams. The movement toward payers calling the “integration tune” needs to be expanded to cover those people that also require services from the other sectors that produce health.

Recommendations and Lessons

Leutz’s three recommendations based on lessons learned were to: 1) involve the users and providers early on in the process of planning and oversight; 2) develop systems to integrate, coordinate, and link services; and 3) clarify the borders between medical and other systems. Over the past 20 years, there has been a growing evidence base regarding the importance of involving all parties early in the coproduction of more integrated care and services. Some of this knowledge comes from experience gained doing community-based participatory research, although there are still lessons to be learned about managing expectations, respecting the partnership, and appreciating each party’s strengths and capabilities.8 A recent initiative to develop a new paradigm for doing health services research that re-allocates the power or influence between researchers and community-based organizations promises to add to this knowledge.9

Leutz asked what systems should be integrated at what levels, and who should oversee them? Over the past 20 years, the tools available to answer this question has increased substantially, including advances in information technologies and predictive analytics that stratify populations to determine their level of need and associated requirements for integrating services. The expansion of electronic health record functionality and associated digital technologies also is assisting in coordinating and integrating care and services, although there are ongoing challenges in exchanging data and maintaining data privacy. ACOs and associated value-based payment models are creating incentives for broader and more integrated systems of care. These are important developments for learning more about how best to integrate systems of care and services at various levels and across various settings.10

The essential question behind “clarifying borders” between medical and social services involves determining the comparative advantage of each in terms of meeting patient needs and preferences. This involves each organization getting to know one another well enough to determine their comparative competencies, and to develop trust and confidence in each other’s capabilities. Once this is determined, it becomes easier to clarify roles and responsibilities. At times, however, what may be needed to promote greater coordinated/integrated care is a “blurring” of roles and responsibilities in which the capabilities of the parties involved overlap with or substitute for each other’s. This is likely to be the case in coordinating patient behavioral/mental health services in which psychiatrists, clinical psychologists, psychiatric social workers, and related mental health professionals may be involved. The challenge of “clarifying borders” is also complicated by different state licensing laws governing what various health professionals can do. Expanding the opportunity for nurses, pharmacists, and others to practice at the “top of their license” to fully utilize their training and competencies will likely both blur and clarify the boundaries involved in coordinating/integrating patient needs.

Moving Forward

Research on integrated care models over the past 20 years has generally supported as well as extended the original five laws and overarching recommendations and lessons. This is especially true for targeting patients most likely to benefit from the higher or full forms of integration, the need for effective interdisciplinary teamwork, joint cross-sector funding where possible, and the need for expanded data exchange across sectors.11-14

There are three relatively underdeveloped areas, however, that present opportunities for further advancing our understanding of how to promote more coordinated/integrated care. These include assessing how to best use the newly emerging virtual digital health technologies (e.g., telehealth) to promote more integrated care; addressing how artificial intelligence and machine learning can be used to better identify patients requiring different levels of care; and recognizing and identifying the emerging and expanded roles of patients to integrate their own care. Cutting across all of these areas is the urgent need to address the inequities in health and social services delivery caused by systemic racism and related biases built into our health care institutions over many decades. This will require, at a minimum, systems thinking leaders with an enduring moral compass, and payment models that reward integrated care not only within the health care sector but also across sectors targeted to populations most in need.


  1. Leutz WN. Five laws for integrating medical and social services: lessons from the United States and the United Kingdom. Milbank Q. 1999;77(1):77-110.
  2. Mongeon M, Levi J, Heinrich J. Elements of accountable communities for health: a review of the literature. NAM Perspectives. Discussion Paper, National Academy of Medicine, Washington, DC. November 6, 2017.
  3. Griffiths, F, Cave J, Boardman F, Reb J, Pawlikowska T, Ball R, Clarke A, Cohen A. Social networks—the future of health care delivery. Social Science & Medicine. 2012;75(12):2233-2241.
  4. Nuno-Solinis R, Shortell SM, Scheffler, RM, Kellogg MA. Implementing population health in the US: lessons from Spain. Health Affairs Blog. July 9, 2019.
  5. Butler SM, Higashi T, Sheriff N. How budget flexibility tools help address social determinants of health. Washington, DC: Brooking; 2020. Accessed November 18, 2020.
  6. Sinaiko AD, Landrum MB, Meyers DJ, et al. Synthesis of research on patient-centered medical homes brings systematic differences into relief. Health Aff. 2017;36(3):500-508.
  7. National Academies Report. Community-Based Organizations Are Important Partners for Health Care Systems, Washington DC, 2020.
  8. Williamson HJ, Carmenlita C, Dulce J, et al. Voices of community partners: perspectives gained from conversations of community-based participatory research experiences. Int J Environ Res Public Health. 2020;17(14):5245.
  9. The Blueprint: October 2020 Paradigm Project Updates. AcademyHealth. October 14, 2020. Accessed November 18, 2020.
  10. Kennedy G, Lewis VA, Kundu S, Mourques J, Colla CH. Accountable care organizations and post-acute care: a focus on preferred SNF networks. Med Care Res Rev. 2020;77(4):312-323.
  11. Goodwin N, Shapiro J. The road to integrated care working. Health Services Management Centre, University of Birmingham; 2001.
  12. McCarthy D, Ryan J, Klein S. Models of care for high-need, high-cost patients: an evidence synthesis. The Commonwealth Fund. October 29, 2015. Accessed November 18, 2020.
  13. Anderson GF, Ballreich MHS, Bleich S et al. Attributes common to programs that successfully treat high-need, high-cost individuals. Am J Manag Care. 2015;21(11):e597-e600.
  14. Swanson J, Weissert WG. Case managers for high-risk, high-cost patients as agents and street-level bureaucrats. Med Care Res Rev. 2018;75(5):527-561.


Read on Wiley Online Library

Shortell SM. Reflections on the Five Laws of Integrating Medical and Social Services—21 Years Later. Milbank Q. 2021;99(1):91-98.