by Thomas H Lee
However, by visiting, you cross the bridge between clusters. You may find that in some ways the other institution is different but that the commonalities with your institution are actually strong. Their heart-failure patients also show up short of breath; their elderly patients also have trouble keeping track of 18 medications; their pediatric patients also have worried parents. The patients’ problems are the same, and many of the solutions from Geisinger or Mayo or Cleveland Clinic or Intermountain could easily be adopted elsewhere.
The truth is that there are practices to emulate almost everywhere you look—if you are looking for them. Here is one of my favorites. At Novant Health in North Carolina, clinicians have developed a Code Comfort protocol in which the needs of dying patients who are short of breath or in pain are met with the same intensity, discipline, and urgency as a cardiac arrest, or Code Blue. We are talking about patients who are not candidates for a transfer to the intensive care unit or for cardiopulmonary resuscitation.
At most hospitals, when such patients get short of breath or are in pain, the nurses page the doctors and the doctors phone in an order to increase narcotics and antianxiety medicines. Sometimes those measures are enough; sometimes they are not. The physician is often tied up with patients who are perceived as having “more active issues” because they are more likely to survive. The nurses are on their own with the dying patients in distress.
Novant’s leadership didn’t want patients, their families, or the nurses to feel helpless in these situations. They laid out clear steps for relieving pain and shortness of breath by increasing medications and enabled nurses to increase medications to relieve the suffering of patients without a physician’s order, just as they can use a defibrillator if a patient has a cardiac arrest.2
If these steps do not ease the patient’s symptoms, the nurses can call a Code Comfort in the same way they might call a Code Blue. In this case, the clinicians who come running are pain specialists and other personnel who are expert in relieving the suffering of dying patients. The development of the clear protocols has made the need for calling Code Comforts rare. The nurses and the families of dying patients get peace of mind just from knowing that the Code Comfort option exists.
Other organizations are imitating the Code Comfort program, and the whole initiative has brought Novant national attention and pride. These other organizations do not ask to see randomized controlled trial data demonstrating that the Code Comfort initiative improves some outcome metric. They learned about it by word of mouth and through a description on the Harvard Business Review website posted on December 9, 2014. The story did not take long to find its audience. Fiercehealthcare.com ran an article on December 11; theadvisory-board.com, on December 12; and forbes.com, on December 13.
Once the information was out there, why did some organizations absorb the idea and start to imitate it whereas others did not? As Burt writes, organizations can strengthen themselves by creating “holes” through which information can enter and building bridges by which information can move from one group to another. If these holes and bridges are planned and important features of organizational life, there is a better chance that a good new idea will be recognized and adopted. In contrast, if bridges and holes to the outside world are not prominent in an organization and its leaders are preoccupied with listening to one another, the variation that is necessary for improvement happens more slowly, if at all.
For this reason, healthcare organizations seeking to drive an epidemic of empathy should do what they can to be part of the larger healthcare community that shares the same goal. When colleagues from different organizations spend time with one another even in informal settings, unintentional learning can occur. For example, I learned about Mayo’s cultural norm of promptly answering beeper pages during a taxi ride with a Mayo Clinic radiologist at a medical meeting. The only reason we were sharing a taxi was that it was raining and the line of people waiting for taxis was so long. A doorman suggested that we might share a cab since we were going to the same place; otherwise, we both would have ridden alone, and my learning would not have occurred.
Burt describes four levels of brokerage through which value can be created:
1. The simplest is to make people in two different organizations aware of the interests and challenges in each other’s settings. In healthcare, common ground can virtually always be identified quickly, since the goal every organization shares is meeting the needs of its patients. Sick people everywhere are in pain and worried; healthy people everywhere are hoping to stay healthy.
2. Transferring best practices is a higher level of brokerage. When an organization adapts something that is creating value in another organization, that effort has a better chance of success than does a completely original idea because the “stolen” concept has already been tested. For this reason, a colleague and I have (half jokingly, but only half) recommended the formation of a healthcare Institute for Imitation as opposed to innovation.3
3. Drawing analogies between two groups that seem to be irrelevant to each other can lead to creative solutions to patients’ problems. For example, awareness that banking and retailers are meeting consumers’ needs 24/7 suggests that forward-looking healthcare providers should consider doing the same thing. Similarly, understanding that the military is able to entrust 21-year-old youths with nuclear submarines and other complex devices suggests that lessons might be learned from those services that could make healthcare delivery safer. In healthcare, social capital can be created simply by recognizing that perceived cultural differences between provider organizations are not as vast as imagined and that best practices at one can be adapted at another.
4. Synthesis is the highest form of brokerage: creating something new that doesn’t exist on either side of the bridge, combining strengths and concepts seen in both groups.
In sum, brokerage leads to learning and fosters improvement through both imitation and innovation. Brokerage can happen by accident—like my taxi ride with the Mayo Clinic physician—or it can be promoted systematically. Organizations that are inwardly focused are obviously at higher risk of falling behind and losing out in a competitive environment because they are slower to learn and change.
However, brokerage is just one part of the creation of social capital. For the ideas that flow from brokerage to achieve their full effect, organizations have to take on the challenging work of closure.
Closure
Almost every organization has people who are eager to innovate. They like to visit other places, and they come back bubbling with new ideas. They are the brokers. Almost every organization also has respected and trusted individuals who are watched closely and emulated by others. Frequently, they are the leaders who can create closure, in which controversy and variation around a practice fade away and that practice becomes the norm within the group.
To enable the dissemination and adoption of a powerful idea, such as an epidemic of empathy, organizations need to integrate their abilities to learn and to trust. Learning allows brokerage, and trust enables closure. A new idea (e.g., we should take the time with each patient to ensure that we understand his or her needs) is introduced, and trust enables the group to function like a team and commit to standardizing the practice and thus reach closure.
Powerful ideas have a way of spreading, but they spread much faster and more effectively in organizations in which people trust one another. It takes trust to adopt ideas from elsewhere at the suggestion of a colleague who has seen them in action and says that they lead to better care. It takes trust to work like a real team, one in which you can depend on others to do their jobs so that you don’t have to check and recheck their work. It takes trust to know that if you make the extra effort to deliver compassionate care and coordinate your work with that of others, your colleagues will do the same thing.
As Burt writes, “Trust is a relationship with someone (or something if the object of trust is a group, organization, or social category) in which contractual terms ar
e incompletely specified. The more unspecified, taken-for-granted, the terms, the more that trust is involved. You anticipate cooperation from the other person, but you commit to the relationship before you know how the other person will behave.”4
Leaders who are effective in driving closure in an organization tend to be those who have earned trust from their colleagues over time. Without such trusted closure leaders, the potential value created by information brokers who have brought ideas across the bridges cannot be translated into social capital. Here is a simple example. Many healthcare organizations started considering full public transparency of their patient experience data after University of Utah Health Care introduced the concept in 2013. However, the organizations that have moved ahead more quickly and implemented this approach tend to be those with a strong physician leader who endorsed it and helped create a review system that clinicians trusted would be fair and would be aimed at improvement rather than humiliation.
Trust is more easily cultivated and closure more easily achieved in organizations that are “tight,” that is, organizations in which personnel know one another and have regular formal and informal interactions and deviations from the norm are easily noticed and acted on. For example, in a small group practice, a clinician who does not deliver care with empathy—or any other dimension of quality—is easily detected. The question is whether the group will exercise its advantage and use its social capital to enforce the social norm of expected excellence.
The value of closure is realized only when such variation is reduced. Even though detection of variation (e.g., poor performance on some dimension of quality, including patient experience) is easier in a small group, that detection is possible with systematic efforts in large organizations, and effective programs to reduce variation can be implemented even without the use of financial incentives. However, this requires effective systems of accountability, which are discussed more directly in Chapter 6.
Let’s close this section with an audacious suggestion with implications for boards and senior management: cultivation of social capital may be more important to the success of healthcare organizations today than amassing financial capital is. I’m not naive about the importance of money, but you can borrow money, whereas you cannot borrow culture or the loyalty of patients drawn to care that is coordinated and empathic. To provide the care required to be competitive in a value-drive marketplace, organizations must be able to innovate, learn, and work in teams. The ability to consider ideas from outside the group, decide which ones to adopt, and standardize care around the new ideas will be a competitive differentiator.
Let us turn to the question of how social capital can be used with advances in social network science to drive an epidemic of empathy.
Social Network Science
Social capital provides broad insights into what organizations need to do to improve performance in this era in which medicine is so complex, so much a team sport, and so expensive. These organizations need to encourage variation in their communities through brokerage—that is, trying new ways of working and working together—but they also have to be ready to decrease variation through closure and promote standardization when that will improve the quality and/or efficiency of care.
Social network science provides insights into how the spread of the values and social norms occurs. Researchers such as Nicholas Christakis and James H. Fowler, the authors of Connected: The Surprising Power of Our Social Networks and How They Shape Our Lives,5 have done much more than describe the structure of communities defined by membership on Facebook or LinkedIn. They have shown how networks function almost like organisms and have characterized the “physiology” through which values spread within them. Their work provides a game plan for healthcare organizations that want to make empathetic, coordinated care a basic value among all their personnel not just some of the time but for every single patient.
Contagious Behavior: An Example
Do behaviors, emotions, and values really spread in the same patterns as infectious disease, from person to person to person? Sometimes the effects of social networks look exactly like those of an infectious condition. For example, in August 2011, Katie Krautwurst developed facial tics. At that time, she was a high school cheerleader in Le Roy, New York, a small town in the western part of that state. Within weeks, her best friend, Thera Sanchez, was displaying similar tics as well as stuttering and involuntary arm movements. Soon a third cheerleader was exhibiting symptoms like Thera’s. By midwinter, 18 residents of Le Roy were in the thrall of the mysterious disorder, including one male high school student and a 36-year-old nurse.
An early skeptical voice was that of Laszlo Mechtler, a neurologist who treated 15 of the girls at the Dent Neurologic Institute, about an hour away from Le Roy. He diagnosed the mysterious epidemic as conversion disorder: the development of neurologic symptoms, including involuntary motion, facial and vocal tics, paralysis, and seizures, without an identifiable physical cause. Neurologists and psychiatrists believe that conversion disorders are generated by psychological stress that is expressed in physical symptoms.
Mechtler’s diagnosis was disputed by many of the patients, their parents, and community leaders. They called in the environmental activist Erin Brockovich to investigate toxic-waste emissions as a possible cause. Brokovich’s investigation failed to find any connection between her prime environmental suspect—a 1970 chemical spill—and the teens’ symptoms. Investigators from the Environmental Protection Agency and the New York State Environmental Commission also came up empty-handed.6
By January 2012, the national media had caught wind of the mysterious epidemic and some of the girls had been interviewed on Today and Dr. Drew. They captured the attention of a pediatric neurologist, Rosario Trifiletti, who specializes in a rare condition called pediatric autoimmune neuropsychiatric disorder associated with streptococcus (PANDAS). Trifiletti saw a possible PANDAS outbreak and volunteered to consult with the families of the affected teens. Several signed on as his patients. Although their blood tests for strep antigens were inconclusive, he prescribed antibiotics and anti-inflammatories. In contrast, those who stuck with Mechtler were referred to talk therapy.
During their treatments, the girls were posting updates about their condition on Facebook. Some also uploaded videos showing their tics and spasms to YouTube. The local and national media persisted in their coverage throughout the school year. Some teens began to show improvement from their treatments, with the greatest improvement apparent in those at the periphery of the epidemic: the youths who were not in Krautwurst’s and Sanchez’s inner circle but who may have shared a class with one of them. Those youths had also escaped the media spotlight.
By late winter, all the affected youths were doing better. Their tics had abated, and some were able to resume cheerleading and other activities. Whether the antibiotics or talk therapy had helped was still a subject of debate. Neither treatment seemed to be as strongly related to improvement as much as another factor: the cessation of media coverage. Once attention was diverted from the epidemic, most of the affected teens started to recover.
As Mechtler explained in a British documentary The Town That Caught Tourette’s, “One of the ways we know to treat this problem is to restore normality—not just to the individual but to the community. But every time a patient would come to me doing better, all it would take is for them to turn on the TV or be approached by the media to aggravate their symptoms.”7
Although Mechtler and his colleague Jennifer McTige had seen many cases of conversion disorder, this was the first time they had witnessed an epidemic of it, a situation known as mass psychogenic illness. They posited that Katie Krautwurst, the index case, had been under stress from her mother’s repeated brain surgeries. Her close friends had responded to her tics and twitches by unconsciously mirroring them. Susceptible students who shared classes with the affected girls were similarly stricken. Even the afflicted nurse was a close friend of a parent of one of the teens.
Bringing Science to the Study of Social Networks
At the Yale Institute for Network Science, Christakis and his colleagues have been studying how social networks work; their insights help explain episodes such as the Le Roy incident and also provide guidance on how to drive an epidemic of empathy. These are certainly not the only social network scientists making important contributions today, but their work is often directly relevant to healthcare delivery, reflecting Christakis’s background as a palliative care physician.
Christakis stumbled into the field of network science 20 years ago, when he was a hospice physician studying the “widower” effect: the increased risk of dying “of a broken heart” after the passing of a spouse. As one of his patients was dying of dementia, Christakis noted that as expected, there was a terrible toll on her husband, her daughter, and the daughter’s husband. The younger generation was worried both for the patient and for the soon-to-be widower. Christakis was taken aback when he got a phone call from one of the son-in-law’s friends, who described how the son-in-law’s distress was casting a pall over the caller’s life. The caller had never met the woman with dementia—in fact, he was three degrees of separation from her—but her illness was making him sick, too.
At that time, national alarm was intensifying over the increase in obesity, which had doubled over the previous two decades. Public health officials began to talk of an obesity epidemic sweeping the country. If there was indeed an epidemic, Christakis reasoned, he should be able to track its spread.
Christakis and his colleague, the political scientist James Fowler of the University of California, San Diego, found the ideal population in which to track obesity: the Framingham Offspring Study, a database of 5,124 people—all of them offspring of participants in the Framingham Heart Study—who had been examined seven times between 1971 and 2003. At each examination, the participants were weighed and asked to list their parents, spouses, siblings, and children. Here is the remarkable part of the Framingham database: the research subjects were also asked to list the name of at least one close friend. In addition, the participants’ addresses were updated frequently, and so the researchers knew how close they lived to one another.
