« What’s Happening at MCOL – Private Health Insurance Exchanges | Main | What’s Happening at MCOL – New lists posted on HealthSprocket »

Clinical Integration: Déjà Vu All Over Again?

by Nathan S. Kaufman, June 1, 2012

Last year, the Centers for Medicare & Medicaid Services (CMS) published proposed regulations for the formation of accountable care organizations (ACOs). Release of the regulations represents the latest of many efforts by the federal government to encourage healthcare providers to become “clinically integrated” and work together to coordinate care, thus reducing the cost and improving the quality of care provided to patients.  

The Federal Trade Commission (FTC) considers physicians who work in different practices to be competitors; thus, under most circumstances, joint negotiation by independent physician practices is considered to be illegal price fixing (Casalino 2006). In a 1996 statement, the US Department of Justice and the FTC provided a new antitrust safety zone to enable independent providers who work in separate practices to jointly negotiate with payers if they are sufficiently “clinically integrated”—that is, if they have formed a network in which there is an organized process to control costs and improve the quality of care resulting from a significant investment of monetary and human capital (Casalino 2006).

According to Gosfield and Reinertsen (2010), the FTC has been “fairly unwilling to define the boundaries of clinical integration, because they wish neither to stifle innovation, nor to encourage anticompetitive behavior.” In her remarks to the American Hospital Association in April 2009, FTC Commissioner Pamela Jones Harbour (2009) stated:

The essence of clinical integration is the interdependency among health care providers. Put simply, each provider must have a vested interest in the performance of the other providers such that their financial and other incentives are closely aligned to meet common objectives.

Since the publication of the guidelines in 1996, several provider networks have been deemed to be clinically integrated by the FTC. These networks have several key common factors:

  • Clinical practice guidelines or protocols to measure performance
  • Information technology to monitor care
  • The ability to evaluate provider performance and to act on the findings
  • The willingness to share data with payers

Medicare’s Physician Group Practice (PGP) demonstration (Iglehart 2011) and mature clinically integrated networks such as Advocate Physician Partners (Shields et al. 2011) have demonstrated that clinically integrated networks have been able to improve performance for select quality indicators. Even so, there is little empirical evidence that these networks have affected the overall cost of care. Even Advocate Physician Partners, one of the most mature clinically integrated networks, notes that a number of its inferred medical cost savings “are based on the achievement of key clinical outcomes that have been demonstrated in the literature to reduce costs” (Shields et al. 2011). One must question whether networks that were developed in part  to use their market clout to negotiate premium rates with payers are able to achieve significant measurable cost savings.

Compliance with FTC legal guidelines provides no assurance that a clinically integrated network possesses the competencies to improve quality, reduce cost, or remain a viable business model over the long term. As was the case in the mid-1990s, many “first-generation” clinically integrated provider networks can be expected to delude themselves into voluntarily taking economic risk for the care they provide. Because few possess the necessary competencies and culture, many will experience financial distress or worse.

The Law of Reciprocal Economics states that one person’s cost is another person’s revenue. In order to reduce the cost of healthcare, someone has to get paid less. With hospitals and physicians accounting for over 80 percent of health insurance benefits spending (PwC Health Research Institute 2011), one must assume that in order to reduce the overall cost of care, revenues for hospitals and high-end specialists, in aggregate, will have to decline.  

Implications for Hospital Leaders

To truly impact costs, provider networks must be committed to reducing utilization of high-end services and must possess the “second-generation” clinical integration competencies that are essential to create a financially successful, sustainable provider network. These competencies include:

A common electronic health record (EHR) with point-of care protocols. In order to monitor cost and quality and coordinate care, it is essential for providers to share a common electronic record. Embedding evidence-based care plans in the EHR can encourage their use by an entire provider network. Geisinger Health System has demonstrated that compliance with care plans can be enhanced by “hardwiring” the protocols into the EHR (Paulus, Davis, and Steele 2008).

As an interim step, networks have created data warehouses to which providers submit their encounter data for analysis and reporting. Transparent reporting of accurate data in near–real time is essential if a clinically integrated network is going to reduce cost and improve quality.

Sufficient primary care capacity. The primary care office serves as the “medical home” for the patient, ensuring that the patient receives appropriate preventive care and monitoring. Given the anticipated shortage of primary care physicians, a model of care involving physician extenders—for example, physician assistants and nurse practitioners—will be essential.

Engaged physician champions. By definition, a clinically integrated network requires independent physicians to work as a team to coordinate care. The evidence from first-generation clinically integrated networks is that physicians will volunteer to surrender their autonomy only to a physician-led enterprise.

Evidence-based inpatient and outpatient care plans. The high variability in the cost and quality of care for specific diagnoses is well documented in the Dartmouth Atlas (Brownlee et al. 2011). Looking at length of stay, compliance with core measures, and readmission rates, my personal experience indicates that this high degree of variability exists within local health systems as well. To have a predictable impact on the cost and quality of care for a population of patients, it is essential that all providers use a common set of evidence-based care plans.

Proactive programmatic approaches to chronic disease. There is some empirical evidence that patients with diabetes and congestive heart failure who participate in disease management programs have better outcomes at lower costs than patients who do not participate in these formal programs (Stock et al. 2010; Russell and Chambers 1999). Home visits, telephone coaching, and web-based monitoring are components of disease management programs that can improve health status and prevent expensive hospital admissions.

Dedicated, sophisticated, mature infrastructure. In addition to the aforementioned information technology solutions, infrastructure will be needed to develop care plans, enroll and train physicians and their office staffs, design disease management programs, and report results. Most health plans possess this infrastructure. Facing a make-or-buy decision, many clinically integrated networks are exploring the prudent approach of joint venturing with one or more payers.

Performance-based rewards and consequences. A fundamental requirement for being clinically integrated is a process for evaluating individual provider performance and acting on the findings. Positive incentives usually involve financial rewards for compliance with key metrics. It is also necessary to sanction providers for poor performance and ultimately eliminate noncompliant providers from the network.

Pilot-testing network performance with health system employees and their beneficiaries. Most health systems are self-insured, giving them significant latitude in structuring benefits and provider incentives. Reducing the use of high-end  services  by employees and their beneficiaries has an immediate positive impact on the health system’s financial performance. Only after a clinically integrated network successfully manages the care of the system’s employees should it consider taking risk for other patient populations.

Proceed with Caution

Compliance with the legal requirements for clinical integration does not guarantee that all participating providers will be better off in the long term. Many providers are forming clinically integrated networks so they can negotiate better rates. But rate pressure on health insurance premiums and the formation of narrow, low-cost networks will limit the ability of clinically integrated networks to negotiate premium rates in the future. 

To succeed over the long term, clinically integrated networks must possess the second-generation clinical integration competencies listed above. They must be flexible enough to adjust to a rapidly changing healthcare landscape, and they must truly be committed to improving measurable quality and lowering the cost of care regardless of the short-term impact on provider revenues.

Getting Started

Clinical integration requires a significant investment in both human capital and dollars. It requires:

  • Strong physician leadership committed to reducing cost and improving quality
  • Sophisticated, proven infrastructure, including personnel and IT similar to that existing in most health plans
  •  A strong, geographically distributed base of primary care physicians
  • Access to a comprehensive set of acute and postacute services.

Healthcare provider s learned in the early 1990s that if a health system does not possess the attributes and competencies necessary to succeed in a new delivery model (which at that time was capitation), it is better off seeking a partner that does, or saving its money and doing nothing.


Brownlee, S., J.E. Wennberg, M.J. Barry, E.S. Fisher, D.C. Goodman, and J.P.W. Bynum. 2011. “Improving Patient Decision-Making in Health Care: A 2011 Dartmouth Atlas Report Highlighting Minnesota.” [Online information; retrieved 8/24/11.] www.dartmouthatlas.org/downloads/reports/Decision_making_report_022411.pdf

Casalino, L.P. 2006. “The Federal Trade Commission, Clinical Integration, and the Organization of Physician Practice.” J. Health Politics, Policy and Law 31 (3): 569–85.

Gosfield, A.G., and J.L. Reinertsen. 2010. “Achieving Clinical Integration with Highly Engaged Physicians.” [Online information; retrieved 8/23/11.] www.reinertsengroup.com/publications/documents/True%20Clinical%20Integration%20Gosfield%20Reinertsen%202010.pdf

Harbour, P.J. 2009. “Clinical Integration: The Changing Policy Climate and What It Means for Care Coordination.” [Online information; retrieved 8/23/11.] www.ftc.gov/speeches/harbour/090427ahaclinicalintegration.pdf

Iglehart, J.K. 2011. “Assessing an ACO Prototype—Medicare’s Physician Group Practice Demonstration.” New Engl. J. Med. 364 (3): 198–200.

Paulus, R.A., K. Davis, and G. D. Steele. 2008. “Continuous Innovation in Health Care: Implications of the Geisinger Experience.” Health Affairs 27 (5): 1235–45.

PwC Health Research Institute. 2011. “Behind the Numbers: Medical Cost Trends for 2012.” [Online information; retrieved 8/24/11.] pwchealth.com/cgi-local/hregister.cgi?link=reg/behind-the-numbers-medical-cost-trends-2012.pdf

Russell, B., and D. Chambers. 1999. “Congestive Heart Failure Disease Management Program.” [Online article; retrieved 8/23/11.] www.dcmsonline.org/jax-medicine/1999journals/sept99/chf.pdf

Shields, M.C., P.H. Patel, M. Manning, and L. Sacks. 2011. “A Model for Integrating Independent Physicians into Accountable Care Organizations.” Health Affairs 30 (1): 161–72.

Stock, S., A. Drabik, G. Buscher, C. Graf, W. Ullrich, A. Gerber, K.W. Lauterbach, and M. Lungen. 2010. “German Diabetes Management Programs Improve Quality of Care and Curb Costs.” Health Affairs 29 (12): 21972205.

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>