You just clicked a link to go to another website. If you continue, you will leave this site and go to a site run by someone else.
It is possible that some of the products on the other site not be licensed for sale in Canada.
Your browser is out of date
With an updated browser, you will have a better Medtronic website experience. Update my browser now.
By choosing to accept, you acknowledge that you are a Certified Healthcare Professional.
The highest-need patients are the least likely to be heard.
By Prabhjot Singh, MD and Niyum Gandhi
November 6, 2017
William Osler, often called the father of modern medicine, famously advised his students: “Just listen to your patient; he is telling you the diagnosis.” A century later, clinicians and health system leaders started tuning out the patient’s voice, turning instead to electronic health records and the latest care protocols to manage their most complicated and high-need patients. We believe it’s time for an urgent and strategic reset. The factors that lead people to become our nation’s costliest are complex. But they call for, at the start, the simplest intervention: listening.
According to the National Academy of Medicine (NAM), “High-need individuals are disproportionately older, female, white, and less educated. They are also more likely to be publicly insured, have fair-to-poor self-reported health, and be susceptible to lack of coordination within the health care system.” Overall, these patients make up just 5% of the patient population, but account for nearly half the spending on health care in the United States. Over the past several years at Mount Sinai Health System, we’ve focused on developing a new generation of clinical services for high-need patients by drawing heavily on strategies pioneered by others across the nation, guided by the recommendations in the newly released NAM report, “Effective Care for High-Need Patients“ (one of us, Dr. Singh, helped develop the report).
Here are the three lessons we’ve learned:
Tune into your patients. Sometimes, the same shortcuts that help physicians save lives can also lead to grave errors. In his groundbreaking book, How Doctors Think, Jerome Groopman provides harrowing examples of misdiagnoses and other negative consequences of tuning out the patient’s voice. In one case, an emergency room doctor failed to recognize a patient was having an acute heart attack because he appeared fit and healthy and did not have any typical risk factors — even though the patient told the doctors he was having sharp chest pains.
Groopman’s book is widely recognized as a clarion call for doctors to listen more deeply to their patients and to make them active participants in their care. This same concept must be applied to designing health systems and care models.
When we began planning for a new high-need-patient clinical service at Mount Sinai called Peak Health, one of the care team’s first steps was to set up an ongoing advisory group made up of patients. Over and over, we heard a clear need expressed for familiar faces on the care team so that patients could build lasting relationships, rather than seeing “all kinds of different doctors giving you all different medications.” We took this feedback to heart and shifted our staffing model from one that depended on part-time clinicians and staff, who might see patients only in “snapshots,” to a model based on full-time, dedicated clinicians and staff committed to knowing all the clinic’s patients. One of our current patients recently said, “Once they get to know you, they know you like a book and always make sure we’re on the same page.”
This relationship building has translated into net promoter scores consistently above 85, well above the norm for health care, along with early indications of improved adherence to treatment plans and reductions in hospital utilization. By listening, the team prioritized important elements of the clinical service design that they may have otherwise overlooked. Based on our early experience, Stella Safo, a physician leader, and Bruno Silva, an interaction designer, collaborated to build a real-time patient engagement tool that we are replicating in other settings.
Listen to other organizations. As the NAM report points out, caring for high-need patients extends beyond their physical ailments, and into behavioral and social services they need in the neighborhoods where they live. While this makes sense in theory, how do you address these broad needs in practice? In designing our program, we made a point of visiting exemplary organizations including Oak Street Health, CareMore Health System, and Iora Health to find out how they handled this challenge.
Each one asked their care teams to, in essence, shift their mindset from “the person in front of you” to “the people reliant on you.” This resulted in operating models built to manage patients longitudinally rather than from visit to visit. The teams redesigned their workflows to support this model and developed simple tools to focus the team’s time on the patients who need it the most.
For example, Iora Health uses an intuitive and powerful “worry score” to identify patients that anyone on the team believes need more attention. The system scores patients on a 1-to-10 scale based on risk-adjustment data, risk factors such as recent hospitalizations, and any relevant changes in patients’ lives that a member of the care team learns about through conversation.
We’ve adapted what we heard at these institutions to develop and implement new models for high-need patients. We now use a similar worry-score system in our Peak Health clinical service so that the entire team spends more time focusing on the people who require the most attention on any given day, rather than just those who came to the clinic that day. We have early data showing that applying what we have learned from our top-performing peer organizations is translating into better care, lower costs, and higher patient and provider engagement.
Listen to colleagues. To tap expertise throughout Mount Sinai Health System, over the past year Natalie Privett, a colleague with an industrial engineering and learning background, has created a learning network called the Ambulatory Care Transformation (ACT) Network. She identifies evangelist “doers” across the system who recognize that the traditional fee-for-service system is hindering their ability to develop new care models; that we need workflows that support proactive, rather than reactive, care; and that care teams need to include significantly more nonclinical workers.
As the ACT network began to form, both online and offline, members found that familiar themes — such as focusing on culture in the hiring and training process, using nontraditional care-team members to integrate behavioral health, and deploying home-based care services creatively — emerged and continue to be incorporated as they transform clinical programs. It would have been much harder to go the other way around, trying to implement textbook care models from the top down, rather than allowing them to be developed from the bottom up through a learning network.
In a fast-paced health care system, it is clear that patients will benefit from the work of researchers and technologists focused on data-driven technologies to improve care. However, the clinical insights and strategies these technologies can give rise to are most useful when they’re incorporated into clinical care by providers who listen carefully to their patients, their colleagues, and the exemplary organizations around them.
Prabhjot Singh, MD
Prabhjot Singh, MD, is Director of The Arnhold Institute for Global Health and Chair of the Department of Health System Design and Global Health at the Icahn School of Medicine at Mount Sinai. He is co-author of the forthcoming book, “Investing in Better Healthcare,” published by Columbia Business School Publishing.
Niyum Gandhi is Executive Vice President and Chief Population Health Officer for Mount Sinai Health System. He is co-author of the forthcoming book, “Investing in Better Healthcare,” published by Columbia Business School Publishing.
Referenced path does not exist