» Back to Table of Contents
Mr. LW, an 83-year-old WW II Veteran is about to be discharged from the VA hospital four days after presenting with a deep vein thrombosis (DVT), for which he has been treated with warfarin, an anticoagulant that requires close post-discharge monitoring. Past medical history includes a distant traumatic brain injury (TBI) that resulted in mild cognitive impairment. His discharge includes: a visit by a pharmacist who tells him to take his medication as prescribed in the hospital (10mg), and reminds him to have his blood levels checked next Thursday at the warfarin clinic. Ten minutes later, a nurse tells him that she will follow up with his daughter
(although he lives with his wife who is his health care proxy). Finally, a second-year medical
resident gives him a written discharge summary, and instructs him to break in half the 10mg tablets he will receive. He reminds him to follow up at the warfarin clinic next Wednesday.
Mr. LW is not seen again until four weeks later when he is again admitted, this time for a DVT in his left leg.
It would be easy to conclude from this actual case that Mr. LW's care is an isolated example of individual break-downs in communication. However, it is estimated that up to 20 percent of hospitalized patients like Mr. LW are re-admitted for problems which could have been prevented had communication and coordination been more effective. Health services researchers,
and others interested in quality and safety, have pinpointed the increasingly complex web of human/machine interactions that comprise health care delivery today. This web has expanded
to the point where it makes less sense to blame single individuals and weed out "bad apples," than it does to study the behavior of clinical "microsystems," the smallest functional work unit in a hospital or clinic.1 Incorporating a systems approach extends our understanding of high reliability health care as a complex constellation
of interrelated actions and activities rather than the addition or subtraction of a single individual's
Recognizing the critical need for new care delivery models, the Institute of Medicine (IOM) in 2001 published a highly influential monograph entitled "Crossing the Quality
Chasm: A New Health System for the 21st Century."2 In that document, the IOM asserted that patient-centered care—i.e., care that establishes a continuous partnership
among practitioners, patients, and their families to ensure high quality accessible services—was one of six indicators of quality.
One response to the IOM report has been the rapid growth of the patient-centered medical home (PCMH), also known in the VA as the Patient Aligned Care Team (PACT). Rather than viewing a patient's care as episodic and the responsibility of several to many unconnected individuals operating across time and space, the PACT views the care process as a continuum, from primary to specialty care, from hospital to home, and from the clinic to the community. The potential payoff from the PACT model is to provide patients with a seamless care experience whether it is in the hospital, clinic, or at home. Realizing this potential is a huge technological, logistical, and cultural challenge.
One promising research approach to transforming
performance within and across settings is "relational coordination," a term coined by Jody Gittell, a professor of management at Brandeis University. Relational coordination refers to, "a mutually reinforcing process of interaction between
communication and relationships carried out for the purpose of task integration." Gittell first applied the concept to studying Southwest Airlines and found that the company's success lay primarily in its ability to encourage and support
high levels of communication across multiple
job classifications and management. Her latest work has been in health care where she has shown that organizations with high levels of relational
coordination have better care outcomes and lower overall costs.3
Using the concept of relational coordination, we can return to Mr. LW's discharge and ask whether improved relational coordination could have led to a different outcome. It is clear that the health professionals caring for Mr. LW had different information and ideas about the post-discharge plan. One improvement might have been to connect the health care professionals to one another sequentially, each briefly noting in VA's Computerized Patient Record System what was discussed. Another approach might have been to start each conversation by asking Mr. LW who else had already talked with him about discharge and what the content of the discussion
was. A third approach might have been to ask which family member he would prefer to be contacted with follow-up information and appointment
reminders. Finally, asking Mr. LW to repeat back his understanding of the discharge plans (also known as a teach-back or a talk-back) might have uncovered discrepancies in the information
he had been given and perhaps have prevented
some of the confusion he experienced.
Relational coordination is not about blaming individuals for poor performance, but rather encouraging us to recognize the immense complexity
of creating coordinated experiences for the billion patient visits in the United States each year. Doing so requires everyone's best collective
efforts and creativity; to do less may put the future of patients like Mr. LW, and our current medical culture, at risk.
- Nelson, E.C. et al., "Microsystems in Health Care: Part 1. Learning from High-Performing Front-line Clinical Units," Joint Commission Journal on Quality Improvement 2002; 28(9): p. 472-93.
- Institute of Medicine 2001, Crossing the Quality Chasm: A New Health System for the 21st Century. Washington D.C.: National Academies Press.
- Gittell, J.H. 2009, High Performance Healthcare: Using the Power of Relationships to Achieve Quality, Efficiency and Resilience,
New York: McGraw-Hill.