Stanford’s Big Health Care Idea

Physician impossible: Arnold Milstein had decades of experience coming up with innovative approaches to controlling health care costs by the time he got to Stanford.

When Arnold Milstein arrived at Stanford University in 2010 to create the Clinical Excellence Research Center, he already had several careers’ worth of experience in medical innovation. He had been in private practice as a psychiatrist; founded a health care consulting company; examined the organizational structure of hospitals and private practices, poring over the data on the quality of health care; and applied what he learned to improve care for Boeing employees in Seattle and hotel workers in Atlantic City. The biggest lesson he took from all those experiences was that American health care was ill-serving the very people who needed it the most. He had come to Stanford to study ways to make health care work better.

Tall and slim, with a kind face and short hair cropped straight across his forehead, the sixty-seven-year-old Milstein explains the problem succinctly: “It’s a 5/50, 10/70 world.” That is, 5 percent of patients account for 50 percent of health care spending, and 10 percent account for 70 percent, whether they’re insured privately or by the government. These high spenders are the sickest and frailest, patients Milstein calls the “medically fragile.”

At Stanford, in sunny, health-conscious California, Milstein saw the same thing. As a member of Stanford’s employee benefits committee, which oversees the university’s self-funded health insurance plan, he knew that medically fragile Stanford employees were sucking up the vast majority of health care spending and straining Stanford’s system, without many signs of improved health. He had a theory for why this was happening. The patients weren’t the problem; the problem was that the health care system was treating them the way it treats everybody else.

Milstein also had a theory for how to solve the problem. What if you took the concept of an intensive care unit—a single location that pulls together all the personnel and technology needed to care for the sickest patients in a hospital—and applied it to patients who were well enough not to be in the hospital but a lot sicker than the average patient in a primary care doctor’s practice? Some of these people are old and frail, but many are young, hold down jobs at Stanford, raise families, and coach Little League, even though they have one or more chronic illness, like diabetes, depression, or cancer. They are among the most expensive to care for not just because they are sick, but also because the health care system is inefficient and disorganized when it comes to taking care of their multiple conditions. Why not organize the care around them the way a hospital organizes all the nurses, doctors, and technology needed for patients in the ICU?

Through his experience with Boeing and in Atlantic City, Milstein knew that the hardest part of setting up this new model of care would be not the patients, but the limitations of the current system, which organizes individual physicians around individual appointments, not overall patient health. Years ago, he said, he had realized that for this kind of project he would need what he jokingly called a “mission impossible” team, one willing to organize a radically new kind of primary care practice. He immediately thought of Alan Glaseroff, a “diabetes guru” who ran a private practice and led a group of 240 physicians in California’s Humboldt County. Glaseroff’s group had figured out how to control diabetes far more effectively than most primary care practices without raising costs. So Milstein invited Glaseroff to meet; Glaseroff, in turn, invited Ann Lindsay, Humboldt County’s public health officer and the co-owner of Glaseroff’s own private primary care practice—and his wife. Milstein asked the couple if they would consider bringing their innovative model of care down from the forests of Humboldt County to the rolling hills and shopping malls of Palo Alto. It would be part of a new clinic, Glaseroff remembers Milstein telling him—Stanford Coordinated Care, or SCC—that would be “designed specifically for the 5 percent of people who cost 50 percent of the money.” The pair decided to give it a try.

If you exit the eight-lane freeway that bisects the hills just south of San Francisco, pass the turn for the local community college, and travel up a winding hill dotted with eucalyptus trees, you’ll find a wood-shingled bungalow with a sign next to the door: “The Emerald Hillbillies.” The name is a play on the grandiose name of the town, Emerald Hills, and the doctors who have lived there, with their poodle, Lola, since 2011.

Glaseroff and Lindsay met in college and attended medical school at Case Western Reserve University in Cleveland. For both of them, medical training was essential, but their personal experiences were perhaps more instructive.

Lindsay’s epiphany came first. As a college student in the 1960s, involved in the women’s health movement, she did something that women a generation previous never would have considered: a breast self-exam. She found a lump. The next logical step was to visit her doctor—who asked, condescendingly, “What are you doing feeling your breast?”

“At that time, the medical care for women in particular was that things were done to them rather than done with them,” she said. In part as a result of that encounter, she decided to become a doctor herself, committed to the idea that physicians should work with patients, male and female, not dictate to them.

Glaseroff’s moment of enlightenment came in 1983, when he was diagnosed with Type 1 diabetes. By that time, the couple was already in Humboldt, sharing a primary care practice. When he received the diagnosis, he pictured his life unspooling like that of a family friend with the disease: amputations, dialysis, and, finally, death.

What he learned was that diabetes only kills if it’s not well managed. Chronic illness requires patients to reorient their whole lives around daily vigilance. And, Glaseroff said, having a supportive medical team was essential. He incorporated those lessons into his own practice. “What I did was stop yelling at people,” he said. When doctors scold patients about the consequences of their weight, diet, and exercise habits, he explained, all it does is scare them.

Soon, he was seeing most of the diabetes patients in Humboldt County. While he emphasized self-management, he also saw that the way primary care practices typically work—medical assistant measures blood pressure and weight; doctor spends seven minutes with patient, writes prescription, tells patient what to do—wasn’t doing the trick. Many of Glaseroff’s patients would do well for a while and then, for lack of a better word, relapse. A prescription for insulin does no good for a patient who doesn’t have a refrigerator to store it in. If the reason someone can’t control her eating is that she is coping with the untreated after-effects of trauma, nutrition counseling is useless.

Suddenly Glaseroff and Lindsay were pulling in multiple disciplines to meet their patients’ mental and physical needs. Their medical assistants, who in traditional primary care offices may do little more than take vitals and escort patients to exam rooms, began following up with the clinic’s patients on the phone—monitoring their progress, providing test results, and answering questions. This new strategy improved the care they were able to provide to their patients, but it was tough under the standard fee-for-service model, which pays physicians only for seeing patients in the office. Glaseroff and Lindsay managed to pay their assistants by making do on half of a typical primary care doctor’s income each.

While Glaseroff was learning how to manage diabetes, Lindsay became the public health officer for Humboldt County. She’d always understood the concept of “social determinants of health”—that is, that things like education levels, socioeconomic status, and the physical environment can make it easier or harder for patients to get and stay healthy. But until she worked in public health, she didn’t have deep experience with it.

Then came an outbreak of shigella—infectious diarrhea—at a homeless encampment on a beach jetty in Humboldt. When Lindsay sent public health outreach workers out to the beach, they found that what most of the jetty residents needed wasn’t health care; it was tires on their cars, spaying or neutering for their pets, official identification, addiction counseling. And, inevitably, housing. So the public health workers got to work hooking the people on the jetty up with social services that were already available. As more people were housed, sanitation improved—and the shigella outbreak evaporated.

When Lindsay and Glaseroff arrived at Stanford to create the coordinated care clinic, they had three guiding principles: one, they wanted to emulate the community outreach model Lindsay had used to deal with the shigella outbreak; two, they knew that engaged patients fared better; and three, fee-for-service payment would have to go.

The fee-for-service model is straightforward and profitable for specialists like orthopedic surgeons, gastroenterologists, dermatologists, and cardiologists. They get a handsome fee each time they perform a knee surgery, do a colonoscopy, or spend thirty seconds burning off a suspicious mole. For primary care doctors, however, who don’t do procedures, and whose fees are far lower, the only way to make more money is to see as many patients as possible a day, often in fifteen-minute slots. Short visits work for healthy patients who have a sudden, minor ailment that can be cleared up with a prescription or a referral. But for patients who are medically fragile, fifteen minutes isn’t enough.

Lindsay and Glaseroff began interviewing potential SCC patients—Stanford employees who, on average, had nine separate medical conditions and cost the health plan $43,000 per year, according to data from a study funded by the National Institutes of Health. The couple heard over and over again that these patients were dissatisfied with their primary care. They were treated like a list of conditions, they said, and were constantly being shuttled between specialists who didn’t communicate with one another. If one specialist glossed over something in the chart, the patient had to catch it. And their primary care providers—whom many of the patients valued—just didn’t have time to help them decipher the complicated and often conflicting medical advice they were getting from their various doctors.

“In many ways, they were just overwhelmed,” Lindsay said. “Many had lost all hope. . . . No one [on the medical team] put their health needs in the context of their overall goals.” And how could they, in fifteen minutes?

To figure out how to improve care for the most fragile patients, Lindsay and Glaseroff analyzed the records of the biggest regular health care users among Stanford employees and their dependents, looking for conditions shared by multiple patients. The goal, said Glaseroff, was to build a team that could meet the most common needs. That way, they wouldn’t have to refer to specialists outside of the clinic. Fewer referrals meant fewer additional charges to Stanford Health Care-—the university’s self-funded health insurance system-—and fewer copays for patients. The intensive care would also, theoretically, help patients avoid health crises that often sent them to the emergency room.

The final team was small but powerful: two physicians (Glaseroff and Lindsay each served part-time and together made up less than a single full-time equivalent physician), a registered nurse, a physical therapist, a social worker, a pharmacist, and four medical assistants who would be trained to act as “care coordinators.”

Glaseroff and Lindsay divided the total cost of those providers’ salaries by twelve to arrive at a monthly cost for the clinic. They calculated that they could provide care to about 500 patients. They landed on a reasonable figure: the clinic needed $286 per patient per month from the Stanford Health Care plan. The monthly amount would allow the practice to do all the work that was “unbillable” under fee-for-service, like phone calls, emails, and video calls, to keep patients connected without having to come into the office.

Under fee-for-service, most primary care practices want to keep their waiting rooms filled, whether the patient really needs to come in or not. “If a patient thought they had a urinary tract infection, they would have to come in,” said Glaseroff. Under the per-patient, per-month payment plan, “We handled them on the phone and got them treated. We handled them in a way that under fee-for-service would have been financial suicide.”

But the even more important benefit of up-front payments, Glaseroff said, was that if patients didn’t have to come into the office, the clinic could see fewer patients in a day. That let first-time appointments expand from fifteen minutes to two hours. The physician would spend an hour with a patient, going through their medical history thoroughly and addressing complaints. That session would be bookended by two thirty-minute meetings with a care coordinator, who would help the patient set reasonable goals. Follow-up appointments would allow thirty minutes with a doctor and another thirty with a care coordinator. In between, patients could call or email their care coordinators and doctor at any time, and get a thorough response.

They had determined the model; now they had to sell it to Stanford. It wasn’t easy, said Milstein. For one thing, while SCC stood to save Stanford Health Care substantial money, it would hurt Stanford’s hospital in the short term. Assuming they succeeded in keeping patients out of the hospital, fewer admissions would mean less revenue, because the hospital is paid on a fee-for-service model. The insurance side stood to win, the hospital to lose.

They also had to win over their Stanford Medical Center colleagues. Unlike in Humboldt County, where Glaseroff had made a name for himself leading the local physicians’ association and everyone knew Lindsay from her public health work, none of the doctors at Stanford had ever heard of the couple. And bureaucracy meant that what would have taken three months to set up in Humboldt took eighteen months at Stanford.

Theresa Munoz Wynn was a typical patient for SCC. When she began coming to the clinic, five years ago, she had diabetes, high blood pressure, high cholesterol, sleep apnea, and depression. Even with insurance, she was spending more than $500 a month out of pocket for four insulin shots a day, another diabetes drug, cholesterol and blood pressure drugs, and an antidepressant.

For all the money she (and Stanford Health Care) were spending, her blood sugar was still wildly out of control, she smoked a pack of cigarettes a day, and she woke up just about every morning feeling achy and heavy, with a headache and, sometimes, flu-like symptoms. Her first thought would be, “Ugh. I wish I hadn’t woken up.” She worked up to seventy hours a week at Stanford’s contracts and procurement department, and all she wanted when she got home was to get back into bed.

“I was on a path of self-suicide,” she said. “I didn’t have a gun or a handful of pills or a rope, but I was on a collision course. I didn’t care. How I felt was that everything was futile. I was going to be sick forever.”

Most primary care practices struggle to protect patients like Munoz Wynn from health crises in the midst of what often seems like intractable pathology. Insurers struggle to predict the cost of their care, and the sticker price of medical fragility under a fee-for-service model can be astronomical for payers, whether they are the employer, a private insurer, or the government.

Helping these patients begins at SCC with a team-wide meeting every Friday. On a typical Friday, Samantha Valcourt, a registered nurse, takes the train to Palo Alto from her home in San Francisco. As the train ticks past the stops, she scans through her work email, checking for messages from patients with critical needs or questions. When she gets to the office, she takes the elevator to the fourth floor and enters what is, essentially, SCC’s command center.

It’s a large room, set up for collaboration: twelve work-stations line the four walls, with small pony walls dividing each team member slightly from the others. Just before 9 a.m., Valcourt is at her desk and, like everyone else, juggling calls and emails from patients. One patient might report that his prescriptions haven’t been filled. Another might be anxious and need thirty minutes of reassurance to stave off a trip to the ER. In just about all of the calls, the care coordinators ask their patients an important question: What’s your goal?

For one diabetic patient with toe pain, the answer is to dance again. For another with heart failure, it’s the dream to fly to Norway for a family reunion. For many patients, said care coordinator Delila Coleman, it’s simply to get off most of their medications.

As they go through the calls and emails, Valcourt, Coleman, and the rest of the team constantly remind patients that they are in charge of their own health and can make their lives better. The theory is that framing care in terms of helping patients achieve their own goals makes them more hopeful, more motivated to care for themselves, and more likely to call SCC before heading to the ER when they may not need to.

At 9 a.m., team members turn from their desks and assemble around a modest table. After five minutes of mindfulness exercises, they begin reviewing new patients and existing patients who are experiencing complications or are in need of additional support. After discussing a patient’s medical issues, staff members divvy up duties for what the patient needs—a care coordinator will volunteer to call the patient back to talk about his or her goals, or the physical therapist will schedule an appointment to address pain issues, or the pharmacist will do a review to make sure the patient is on the right combination of medications. The approach sounds simple, but it’s completely different from standard American health care, in which the various practitioners often only know what they see in the chart—if they even have time to thoroughly review it.

Over time, that approach saved Theresa Munoz Wynn’s life. First, she started getting mental health and nutrition counseling. With Coleman, she began to devise an exercise plan. Every once in a while, when her steps toward self-care faltered, she would be the subject of that 9 a.m. meeting, and, Munoz Wynn said, the team would come back to “regroup and redirect my care until we found a self-management approach that worked.”

Today, Munoz Wynn doesn’t smoke. She has lost 100 pounds, is no longer diabetic, and doesn’t have high blood pressure. The only medication she still takes is an antidepressant. With more energy, she has picked up hobbies. She crochets now. She has joined some clubs. She is back in school, working toward a bachelor’s degree in business.

In March 2016, Lindsay and Glaseroff presented the initial results of their approach, based on the clinic’s first 253 patients. After six months at SCC, the overall cost of their health care dropped by 13 percent compared to the previous six months, largely due to a 59 percent drop in ER visits and a 29 percent drop in inpatient admissions. (Those numbers exclude one astronomically expensive heart transplant patient.) Using the $43,000 average health care cost figure, that would amount to a savings of about $1.4 million in one year for that cohort.

Today, Lindsay and Glaseroff are in the midst of moving back to their beloved Humboldt, which Lindsay missed every day of the five years they spent at Stanford. Glaseroff will commute to Milstein’s Clinical Excellence Research Center, where he will help spread SCC’s model and others the center has developed. Lindsay, meanwhile, will be working remotely with the Pacific Business Group on Health, a large group of self-insured employers who are looking for ways to improve the quality of care and reduce costs, and might be willing to test out SCC’s model.

Spreading that model will depend in part on how the Trump administration and Congress approach their promise to “repeal and replace” Obamacare. A key provision in the Affordable Care Act was the creation of the Center for Medicare & Medicaid Innovation, which supports the development and implementation of new models of care delivery. Recently, the Department of Health and Human Services, which runs the CMMI, released a new rule for care for medically fragile patients based in part on Stanford’s model. But Tom Price, the new health and human services secretary, has made no secret of his disdain for the CMMI and any effort to reform fee-for-service payment.

The stakes are not merely financial. The care Theresa Munoz Wynn gets from SCC has restored not just her health, but also her dignity and self-worth. “I’m a person, not a statistic,” she said. “I don’t want to sit and die anymore. I want to live.”