As a neurologist just beginning my career and specializing in Parkinson’s disease (PD), most of my training has focused on diagnosis and pharmacological treatment, but I am increasingly preoccupied with the system in which care is delivered. Despite a robust arsenal of symptomatic treatments for this multi-organ disease, fragmented systems of care often require the patient to navigate across many subspecialties alone. However, several Parkinson’s providers around the world have shown that a more integrated system is possible. Several of my patients’ experiences have highlighted the urgent need to restructure this system of care.
Mr. J’s Hip Fracture: The Beginning of the End
Mr. J’s story is particularly telling. He was a talkative man in his 70s, whose stories alluded to a life of accomplishment and adventure that stretched far beyond Boston. I saw him only twice in clinic for medication adjustments for what we suspected was idiopathic PD. Then, just a few months after his second appointment, he fractured his hip while stepping down two steps from his house into his garage. He later told me that he lay on the floor, shouting for help, for nearly 24 hours before being rescued.
Mr. J’s course thereafter was rocky. He developed severe orthostasis postoperatively which — without proper treatment — prevented him from tolerating the hours of physical therapy that he needed. As a result, he never returned to fully independent living. Every time he was readmitted, his Parkinson’s medications were inadvertently held or timed incorrectly. He also suffered from severe depression.
I connected him with a neuropsychiatrist and tried to keep tabs on him, but it was not until his health care proxy hired a private care manager (at considerable personal expense) that the neuropsychiatrist and I were fully in the loop. In this fashion, I continued to manage Mr. J’s parkinsonism from afar with occasional clinic appointments until his recent death.
Although Mr. J’s care coordination was imperfect, the shocking reality is that he was probably getting much better care than most individuals living with PD in the United States. According to an observational cohort of Medicare patients with a diagnosis of PD in 2002, an estimated 42% did not receive any neurological care during the 4 years for which they were followed.
Currently, if a patient with PD in the U.S. even has access to neurological care, most of it is provided through monodisciplinary specialists working in isolation. This model leaves the patient alone to navigate the often emotionally overwhelming diagnosis and then to coordinate care between the neurologist, speech pathologist, physical therapist, psychiatrist, and other providers.
Ms. M’s Depression: More Than Grief?
In a 2016 qualitative study of 30 individuals with PD, many patients described feeling abandoned after their diagnosis and wished that there had been more support. Early in my fellowship, I became the neurologist for Ms. M, whose primary care physician suspected that she had PD. The symptoms were classic, and the diagnosis was straightforward. Less straightforward was the patient’s profound depression.
Her husband had recently passed away, but I suspected that her untreated PD was also a major cause of her current mood. In addition to starting her on dopamine replacement therapy, and in consultation with her primary care physician, I started her on an antidepressant. I also scheduled her to come to see me monthly for the next few months. Her rigidity melted away immediately, and her face grew animated. It took another 6 months, but eventually the depression lifted. At our most recent visit, she was nearly unrecognizable compared with the debilitated and tearful woman who had first walked into my office.
I wondered, though, had I been a busy attending neurologist booking appointments 6 months out, would I have had the time to support her through her initial diagnosis and treatment? Was it the best use of resources for her to see me monthly, especially if doing so limited access to other patients? Perhaps a social worker or psychiatrist would have been better equipped to help her — yet I didn’t have a social worker to offer, and psychiatry or psychology referrals take months.
Multidisciplinary Care vs. Interdisciplinary Care
Some centers have moved toward multidisciplinary care, a model in which various health care professionals in a predefined network refer patients to one another. In such a model, I might have been able to expedite a referral for my patient, but coordinating her care still may not have been easy. Multidisciplinary models involve providers working in parallel, with each being responsible for a different patient care need. Participants in the 2016 study (mentioned above) voiced a desire for true interdisciplinary care. One 57-year-old woman commented:
“You know what I would love? Because there are so many components to this illness, I wish there was one-stop shopping. I wish that I really could have a care team under one roof and that it wouldn’t be me who was always responsible for making sure that everybody worked together.”
An Integrated Care Model for Parkinson’s: One-Stop Shopping
What would this “one-stop shop” look like? The Integrated Practice Unit (IPU) model of care can serve as a roadmap. IPUs are a specialized structure for delivering true interdisciplinary care. They utilize a colocated team of health care professionals who organize around a specific medical condition and jointly evaluate and manage the patient over the entire course of the disease. A successful IPU eliminates the barriers between siloed departments and different settings of care.
Two critical components of an IPU are (1) the utilization of outcomes that matter to patient as the North Star for designing care and (2) joint accountability for the outcomes and costs of care. In an IPU model, Mr. J would have remained seamlessly connected to his care team despite his physical transitions between different settings of care. Similarly, Ms. M would have had immediate support from a team after her diagnosis and assistance navigating care going forward.
Effective Parkinson’s treatments include drugs, deep brain stimulation, physiotherapy, and exercise, as well as myriad other treatments for non-motor symptoms. Taken together, these treatments are good enough to give patients additional decades of high-quality life and turn PD into a condition that people die with and not of. It seems clear that there are scientific advances yet to come. What is needed now is a concentration of expertise into an IPU to allow the medical and technical expertise that has been gained in the last century to be systematically delivered to patients in a highly reliable fashion.
This vision is not just a pipe dream. Over the course of my fellowship, I saw that when I had the luxury to spend extra time with my patients, as I did with Ms. M, I could provide compassionate care and achieve the outcomes that mattered most to them. Taking the time needed to deliver a new diagnosis of PD and then checking in by phone the next day or week helped to mitigate the experience of receiving an overwhelming diagnosis. Delivering individualized, specific recommendations about exercise helped patients follow through on this underutilized form of treatment. Referring patients to other specialists with a phone call or face-to-face conversation — rather than just through the electronic record — helped to bridge the gap that normally divides providers.
The question, of course, is how to make this kind of care sustainable. The answer is not to have individual physicians or nurses work harder or longer for individual patients, which would lead to clinician burnout and decreased patient access. Rather, this kind of personalized care must be systematized.
Beacons of Integrated Care
Such systematized care already exists in organizations around the world. In The Netherlands, ParkinsonNet started in 2004 as a network of specially trained physical therapists and today has grown into an infrastructure of 70 regional networks and approximately 3,000 trained professionals from 12 disciplines who provide previously inaccessible specialized care to individuals with PD.
Most studies of this network have demonstrated improved health outcomes. For example, an observational study showed that patients who accessed ParkinsonNet providers had a 50% lower annual rate of hip fractures and fewer inpatient admissions than those who did not. A randomized, controlled trial comparing ParkinsonNet occupational therapy with usual care demonstrated significant improvement in terms of self-perceived performance of acts of daily living. ParkinsonNet also has demonstrated annual cost savings of US$439 to US$1,675 per patient. Although modest, these savings easily offset the organizational start-up costs.
In the U.S., several health care systems have inched toward an IPU model of care. Kaiser Permanente has adopted aspects of ParkinsonNet, and plans are underway to expand the program to the state of Michigan. Cleveland Clinic has completely reorganized care around disease conditions rather than traditional academic departments.
Beyond these major endeavors, there are initiatives at various movement disorder centers that have incorporated elements of an IPU, ranging from having a dedicated Parkinson’s social worker or physical therapist on staff to offering group educational classes. However, it has become clear that while the pieces are there, stitching them together is the hard part. Bas Bloem, MD, PhD, founder of ParkinsonNet, in an interview focusing on the international dissemination of ParkinsonNet, noted that “We’ve . . . been surprised by the level of knowledge that was already out there, both at Kaiser and also in Germany, but it wasn’t always well organized or visible.”
Pulling the pieces together requires implementation of the two critical components of an IPU: (1) utilization of patient-reported outcome measures (PROMs) to guide structural change and (2) a simultaneous focus on the value of care (defined as quality of care divided by cost). Validated PROMs for PD already exist (e.g., the International Consortium for Health Outcomes Measurement [ICHOM] outcome set; the American Academy of Neurology [AAN] PD Quality Measurement Set; and ParkinsonNet’s ParkinsonInzicht), but a lack of consensus on which measure set to use has led to fragmented adoption and an inability to assimilate data across centers. Just as movement disorder physicians have achieved international consensus on a standardized measure of motor function in PD (the Movement Disorder Society–sponsored revision of the Unified Parkinson’s Disease Rating Scale [MDS-UPDRS]), a consensus must quickly be forged around a set of PROMs. The chosen measure set must focus not on process metrics, but on true outcome measures.
Overcoming the Problem of Payment
And what about value? Unfortunately, many of the examples cited above have failed to develop a sustainable funding scheme. Currently, ParkinsonNet is financed by a temporary health innovation grant and a small annual membership fee paid by participating health care professionals. Many of the individual movement center initiatives — such as having a dedicated physical therapist in clinic — are entirely philanthropically funded.
Why hasn’t there been a demonstration of improved quality and decreased cost over the long run? Under the current fee-for-service system, it will be nearly impossible to achieve this goal; only when the same entity that shoulders the cost for coordination of care also reaps the savings from reduced hip fractures and hospital admissions will the hypothesized savings be realized.
This type of risk arrangement theoretically can be achieved in an Accountable Care Organization (ACO), in which health care systems or groups of physicians adopt a cohort of patients for whom they will share the financial risk of care with the insurance provider (payer). Both the health system and its physicians and the payer share in savings when high-quality care (monitored through chosen metrics) is delivered at lower cost. All parties share in losses when the rate of health care spending is not curtailed.
PD represents an ideal condition to manage with the use of an IPU structure that is financed through a shared-risk arrangement such as an ACO. Some systems, such as Kaiser Permanente, employ an ACO model (in their case, through a fully integrated system) and have branched into specialty disease management. Yet no specialty-based chronic diseases have been selected for quality improvement targets under current Medicare ACO models; rather, Medicare and most commercial ACOs focus exclusively on population health management of primary care conditions. There is a huge opportunity here.
To date, IPUs have been implemented frequently for conditions requiring surgical care, but their greatest value is likely in treating longitudinal conditions — the savings for which can finally be realized under shared-risk arrangements. With proof of concept provided by ParkinsonNet, PD is an ideal specialty condition around which to organize an IPU. Mr. J and Ms. M deserved to benefit from the much-needed innovation of our current system. We must do better for current and future PD patients, and we have the tools to do it now.