Optimal Care

The current issue of the Journal of Parkinson’s Disease includes a new feature, “How I examine my patient”, a section designed to help improve the clinical skills of physicians, allied health professionals, and other professionals involved in the care of patients with PD and other movement disorders. But what constitutes optimal care for Parkinson’s disease? And how much difference can it make?

When Joyce A. Oberdorf became the new CEO of the Miami-based National Parkinson Foundation in 2008, she decided to find out. 

Oberdorf, whose previous work resume included stints at Lehman Brothers, Aetna, and the Michael J Fox Foundation for Parkinson's Research, had been stimulated to think about this fundamental issue in part through reading a provocative 2004 article in The New Yorker by Boston surgeon and writer, Atul Gawande. Entitled “The Bell Curve”, the piece reported on the striking differences in the quality of care delivered by different medical institutions — what health policy wonks call “comparative effectiveness”. Gawande’s main example in the article concerned a devastating inherited pediatric disease called cystic fibrosis. Based on data from the Cystic Fibrosis Disease Foundation, his article revealed that the nation’s 117 certified centers of excellence — which all followed the same “best practice” guidelines — achieved alarmingly different results. “In 1997,” wrote Gawande, “patients at an average center were living to be just thirty years old; patients at the top center typically lived to be forty-six.” The variation between the very best and the very worst centers was even more shocking. Patients being treated at the top performing institutions had double the life expectancy as those at the lowest ranked ones. What was true for cystic fibrosis, Gawande reported, held in other medical specialties, from heart surgery to managing chronic respiratory disease. According to Gawande, the data across medicine shows “a handful of teams with disturbingly poor outcomes for their patients, a handful with remarkably good results, and a great undistinguished middle”.

Oberdorf wanted to know if Parkinson’s disease care showed a similar variation in effectiveness between its top centers. After sitting down with leading clinicians and health policy experts it became clear that the top Parkinson’s disease clinicians lacked any scientific data about how good their care was as compared with other practitioners. So, in 2009, the National Parkinson Foundation launched “The Parkinson’s Outcomes Project”.

The initiative set out to follow a group of 7,500 people with Parkinson’s disease and see what happened; that is to carefully document what symptoms patients reported, how slowly or quickly their disease progressed, and which therapies clinicians prescribed. The men and women in the cohort range in age from 25 to 95. Some 400 of them contracted the condition under 40 — according to the National Parkinson’s Foundation, that’s the largest group of people with young-onset Parkinson’s ever studied. The cohort also includes more than 350 people who have had Parkinson’s for more than 20 years, with at least one case living with Parkinson’s for 49 years. The hope is that this elite group of survivors may have clues to share. 

The patients were tracked by neurologists at 20 centers of clinical excellence around the world, namely Baylor College of Medicine Houston, TX, Beth Israel Deaconess Medical Center Boston MA, Georgia Health and Science University Augusta, GA, Georgetown University Washington, DC, Johns Hopkins University Baltimore, MD, Centre for Movement Disorders Toronto, Canada, Mount Sinai Medical Center New York, NY, Muhammad Ali Parkinson Center of Barrow Neurological Institute Phoenix, AZ, Northwestern University Chicago, IL, Health and Science University Portland, OR, Parkinson’s Institute and Clinical Center Sunnyvale, CA, Struthers Parkinson’s Center Golden Valley, MN, University of Florida Gainesville, FL, University of Kansas Medical Center Kansas City, KS, University of South Florida Tampa, FL, Vanderbilt University Nashville, TN, University of Pennsylvania Philadelphia, PA, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands, Tel-Aviv Sourasky Medical Center Tel-Aviv, Israel, Toronto Western Hospital Toronto. 

To help sort out the data, Oberdorf recruited mathematician Peter Schmidt. Schmidt, who has bright eyes and wears a smirky smile, sees beauty in numbers. Those numbers, he says, show that Parkinson's disease is spectrum disease. Says Schmidt, “With smallpox, you either have smallpox or you don't have smallpox. But with Parkinson's, it really is an almost arbitrary lines we draw on the spectrum. Everybody slows with age — developing what a neurologist would call bradykinesis. And it's the outliers on this scale that we define as having Parkinson's disease. 

When patients in the Parkinson’s Disease Outcomes Project are asked to rank in order of seriousness the various symptoms impacting quality of life, there is, as you might expect, a lot of patient-to-patient variation. But overall, patients report two areas as the most troubling — mood and mobility. The symptoms patients report as having the most affect on their health status are “negative mood and depression”. The project finds, as previous studies have reported, that at least 40 percent of people diagnosed with Parkinson’s will suffer at some time from depression and anxiety. The next biggest reported area involves a litany of mobility issues, ranging from gait and balance difficulties to loss of fine motor skills. 

Given the complexity of PD symptomatology, researchers could not use a simple outcome measure like mortality to judge quality of care. So how best to determine when patients are improving or getting worse? Schmidt sought the answer in n-dimensional geometry, the topic, incidentally, of his PhD dissertation. “We decided we were going to use a six dimensional model. We don't assume that there's just one measure (e.g. mortality) that represents quality care in Parkinson's disease, we selected six measures.” The six metrics fall into three domains. The first domain tries to assess how the physician thinks that the patient is doing. This involves the TUG test (the time it takes for a patient to stand up, walk, return and sit down) and certain thinking tasks derived from the Montréal Cognitive Assessment tool. The second metric asks patients how they think they are doing — this is based on questionnaires that patients and the caregivers fill out. The third measure asks how the disease is affecting the ability of the patient to live at home. It takes into account falls, hospitalizations, and other events that affect the patient outside of the clinic.

The project’s researchers are not so much interested in absolute values but in how these six measures change over time. With respect to the animal naming task in the Montreal test of cognitive function, for example, says Schmidt, “It’s the change not the absolute number that matters. We had a retired zookeeper who could name 60 animals. So when that guy can only name 30 animals he will be impaired, but he'll still been able to name more than anyone else.” 

For each of the six measures, Schmidt says, they assigned a maximum score of +3 (that’s the best you can do), an average score of 0, and a minimum possible score of -3 (that’s the worst you can do). So, the best center could get a maximum of +18, and the worst could, in theory, receive a score of -18. 

How did the centers do? What the researchers found was that out of the 20 centers of excellence — all considered elite institutions that claim to follow the “best practice” guidelines for PD care — some did much better than others. The top performing centers had quality scores of 7 or 8 points above average, and the lowest performers had scores 8 or 9 points below average. Researchers also found that there were, as Schmidt puts it, “pretty serious differences in the choices and combinations of medications from center to center”. So a neurologist in one center might prescribe a moderately advanced patient just Sinemet (that’s levodopa-carbidopa) several times a day. Whereas another clinician in a different center prescribed that same type of patient other pills as well, for example, a dopamine agonist, the enzyme blockers seligiline and encapatone, and an antidepressant to aid sleep. Then, some centers appear to make more (and earlier) use of allied professionals such as physical therapists, speech therapists, and occupational therapists than others. Lifestyle factors — promoted with varying intensity at different institutions — also appear to play a significant role. Those patients who exercised regularly, for example, tended to do better all round — having better mobility and mood. 

The project aims to sort out precisely what treatment combinations achieve the best results and eventually to publicize which of the 20 centers of excellence did best and which ones underperformed. Even though this is a highly sensitive issue, the project, says Schmidt, is committed to complete transparency. 

While it’s true that Parkinson’s disease is a highly variable condition, the Parkinson’s Outcomes Project suggests that some neurologists deliver more effective care than others. If all neurologists follow the lead of those clinicians getting the best results, we can surely expect Parkinson’s care to improve dramatically.