- To support the Parkinson’s Outcomes Project operations in the collection of data;
- To empower patients to understand their own status and to set goals for their care;
- To guide physicians delivering care in understanding the status of their patients compared to similar patients seen elsewhere; and
- To deliver evidence-based clinical decision support to clinicians.
- Everything is ok, and we need to think about the future;
- Everything is pretty good, but there are some minor issues that need tweaking; or
- It is a crisis and we need to focus on addressing a set of specific issues right now.
- Continue patient enrollment and analysis.
- Evaluate differences on area scanned in different static environments and the translation into dynamic tasks (walking). Additionally, assess the influence of therapeutic visual exploration strategies.
- Development of detailed hypothesis based on our current explorations.
- We are continuing additional data analysis. In addition to the results mentioned here, we are further parsing the patterns of visual exploration, the relation with obstacles, correlations with other clinical features etc. A manuscript is in preparation reporting these results.
- We plan to recruit additional subjects and refine the paradigm while exploring various subtypes of PD, as well as perform testing with and without various treatments.
- We plan to combine the visual exploration experiment, similar to the one performed in this project, with instrumented analyses of gait.
- We are considering the development of rehabilitation and training paradigms based on further findings with this experimental approach.
Funding from The Parkinson Alliance helped to finance the following Parkinson’s research. Grantees were selected by scientific review committees of participating organizations. Updates will be posted, when available.
Project Title: Speech and Deep Brain Stimulation: Treatment Options
Investigators/Authors: Elina Tripoliti, PhD; Lorraine Ramig, PhD, CCC-SLP; Cynthia Fox, PhD, CCC-SLP
Objective/Rationale: Speech can be adversely affected by bilateral subthalamic nucleus deep brain stimulation (STN-DBS) with significant impact on patients’ Quality of Life. The information available to speech clinicians who assess and treat the speech disorders related to STN-DBS is not well-consolidated or presented for the practicing clinician. Further, information regarding the array of treatment options for speech disorders following STN-DBS are not well categorized. This online course will serve to educate speech clinicians, physicians, researchers and patients on key variables related to speech and DBS.
Project Description/Methods/Design: This course will discuss three areas of the speech-DBS literature including: 1) A brief overview of basal ganglia anatomy and physiology and description of deep brain stimulation as a treatment for PD; 2) The various clinical and surgical factors that can affect speech as well as predictive factors on speech response; and 3) Treatment options, classified in two broad categories: behavioral theories (such as the Lee Silverman Voice Treatment) and biofeedback, and multidisciplinary team work coordinating adjustment of electrical parameters (frequency or amplitude of stimulation) and /or repositioning of the electrodes. The paramount importance of team work when working with patients post STN-DBS will be emphasized.
Relevance to Treatment of Parkinson’s Disease: Educating professionals who work with people with PD post STN-DBS is an important step to improving the care of speech following this surgery. This education course may impact treatment management of speech following STN-DBS in a positive way. Further, this course may encourage clinicians to seek further knowledge and/or professional collaborations in their care of people with PD post DBS.
Expected Outcome: We anticipate that speech clinicians, physicians, researchers, and people with PD from around the world will have access to this ONLINE educational course. This will help improve the quality of care of speech following DBS.
October 2014 Project Update:
The course has been designed and is available for health care professionals as well as the general public on the website: www.LSVTGLOBAL.com.
Project Title: National Parkinson Foundation’s Parkinson’s Outcomes Project Clinical Dashboard
Principal Investigator(s): Peter Schmidt, PhD
Objective/Rationale: The Parkinson’s Outcomes Project is a study of care delivered by Parkinson’s expert neurologists. The study currently includes over 7,000 patients, 150 neurologists, and 13,000 clinical evaluations at twenty centers. Data collected includes demographics, care, and both patient-reported and clinically measured outcomes with the goal of identifying best care practices.
All patients deserve excellent, evidence-based care. The majority of patients with Parkinson’s disease do not have access to a specialist and there is a critical and immediate need to develop, prove, and deliver tools that can help these patients get the care that they deserve. A review of outcomes showed that neurologist care reduces the death rate for Parkinson’s disease by 22%: over 6,000 people with Parkinson’s die annually from preventable complications. To address this critical problem, NPF will develop a feed-forward data system that will enable patients to understand their status and clinicians to focus care on issues most important to their patients.
The goals of this system – the Parkinson’s Disease Clinical Dashboard – are as follows:
Project Description/Methods/Design: Designed with the goal of addressing the complexity of Parkinson’s disease without driving information overload, the Clinical Dashboard will collect and display outcome data, presented versus population-based standards. In addition, the Clinical Dashboard will include an associated patient portal that will allow people with Parkinson’s to record data about themselves and then to see how they compare to similar people who participate in the study.
Phase 1: October, 2013-December 2013: Display definition. The Parkinson’s Outcomes Project investigators will work together with outside advisors on defining the feed-forward, feedback display. In addition, the participant (patient and caregiver) data entry portal will be designed.
Phase 2: January 2014-May 2014: Implementation. The Clinical Dashboard will be built according to the specifications developed during phase 1.
Relevance to Treatment of Parkinson’s Disease: The goal of the dashboard, in the order that we will develop these things, are to deliver:
1. Benchmarking. The ability to evaluate a person and be able to understand how that person is doing compared with other, similar people.
2. Goal setting. Tools to help people to understand how similar people have fared over time – not just where are you today, but where might you be next year?
3. Clinical guidelines. Tools to show what the centers that have achieved the best results in similar patients have done to treat them.
Expected Outcome: The results of this evaluation is reviewed before the clinic visit and is used to inform the structure of the clinic visit, for example, there might be separate structured visits for each of the following situations:
Today, a physician might spend half a clinical encounter just figuring out if most things are pretty good or not. Sometimes, physicians are distracted from helping to set a care plan by the need to address specific complaints, but then they discover that the complaints are minor issues. On the other hand, sometimes a patient complaint might seem like a minor issue and be dismissed by the physician, only to discover that it prevents the patient from doing something that is important to him or her.
September 2014 Project Update:
In January, 2014, a set of six key metrics were presented to leaders of NPF Centers of Excellence as the basis for center-center comparisons and as the basis for ongoing evaluation. These six metrics were: health-related quality of life (HRQL), caregiver strain, mobility, cognition, falls, and hospital admissions. To establish their validity for use in informing patients of their relative status, an analytical technique was developed for comparing across diverse patient sets by creating tranches across the population, computing results within each tranche, and then computing an equally-weighted sum across a center’s cohort. Example results from this approach are available on request.
These metrics were subsequently applied to analyze individual results. A software system was developed to select individuals by proximity to a selected candidate by minimizing the norm (equivalent to distance in a carteasian representation) in age and disease duration difference. Using this technique, an arbitrary number of the subjects most similar to the candidate can be selected. To date, the typical number used in comparison sets has been 100, providing good resolution when preparing a percentile scale.
Using this approach, the characteristics of populations and individuals can be compared. First a range of observed outcomes for each variable or a weighted sum of the variables in the comparison patient set is computed. Subsets may be computed to represent patients who do or do not exercise, who are on or not on levodopa or other medications, or other variables. An individual may be compared to the entire comparison set or to the subset that represents that individual’s treatment plan. Then, longitudinal results can be examined to anticipate the effect of a change in the treatment plan. This model has been implemented manually. The coding of this model into an automatic algorithm is the next step.
This study is not yet complete however these results represent the scientific foundation of and, we believe, good progress towards the project goal.
September 2015 Project Update:
Soon after NPF started work on this project, 23andMe was sued by the FDA for providing medical advice. A patient focus group meeting identified little interest in the patient-facing dashboard feature. However, the technology to underly the dashboard concept was completed: given an age, sex, and Parkinson’s duration, software was developed to produce a cohort of 100 closely matched individuals for comparison. Sensitivity analysis resulted in the ratio of the age and duration windows (e.g., +/- 2 years in age, +/- 1 year in duration) being variable depending on duration, where earlier stage patients were better matched across a wider age window but narrower duration window.
Aim 1: Data collection in the Parkinson’s Outcomes Project has continued and the dataset includes over 8,000 individuals and nearly 20,000 clinical evaluations, resulting in the largest clinical study of Parkinson’s disease ever conducted.
Aim 2: While the design of the dashboard was completed and a prototype developed, the dashboard was not constructed. Reports have been made upon request, an automated comparison dashboard has been deferred until a better understanding of the regulatory implications can be developed.
Aim 3: NPF centers of excellence have been provided a report detailing their site’s performance versus other sites. These reports offered detailed information about prescribing patterns, use of DBS, and other clinical parameters and gives a report on each site’s performance across six outcome domains: health-related quality of life, mobility, cognition, falls, caregiver strain, and hospitalizations.
Results: We believe that these results represent good progress towards the overall objectives of the program. It is unfortunate that the Dashboard project was delayed by regulatory concerns, but we believe that the philosophy underlying the dashboard, that of transparency and quality improvement to drive a higher quality of life, is embodied in the over-arching Parkinson’s Outcomes Project.
Project Title: What Patients Look at When Walking
Investigators/Authors: Mark Hallett, MD; Codrin Lungu, MD
Objective/Rationale: Difficulties with walking is often a significant problem for patients with Parkinson disease. Not only is walking slow, but there are also occasional falls and freezing (getting “stuck” in place). Freezing, in particular, is often refractory to medication and even Deep Brain Stimulation (DBS). Freezing is sometimes precipitated by visual stimuli (like encountering a doorway) or relieved by visual stimuli. One of the difficulties in understanding this is that there is no information on where patients are looking while they are walking. With this project, we will study where patients are looking and determine how that influences walking.
Project Description/Methods/Design: The aim is to see where patients with Parkinson disease look when walking. An eye tracking device will be used that allows patients to walk freely. The system will record a video of what is in front of the patient and then indicate exactly what the eyes are focused on in that scene. A comparison will be done between patients and age/sex matched healthy subjects. Exploratory studies will be done first to develop detailed hypotheses. We will study subjects walking in various situations including a plain corridor, a corridor with many objects, a corridor with horizontal lines on the floor, and a corridor with a door to go through. A door often makes walking more difficult, while horizontal lines on the floor might be helpful. Some variability of influence might depend on where the patient is looking.
Relevance to Treatment of Parkinson’s Disease: It is clearly critical in walking to know what lies ahead and how to deal with it. If it is determined that there are systematic abnormalities in where patients are looking, then it might be possible to improve walking by training patients to look differently. The biggest impact could be on freezing of gait where visual stimuli are known to be important.
Expected Outcome: The first outcome is basic knowledge about the coupling of vision and walking in patient with Parkinson disease. Certain types of visual-motor coordination are known to be impaired in patients, and this might be another example, with bad consequences. The second outcome is a set of ideas of how to improve physical therapy gait training. It might also be possible to develop new ideas for devices that would aid vision to help improve gait.
September 2014 Project Update:
Study Progress: By using an eye-tracker while walking, our project explores the presence of visual exploration differences between patients with Parkinson’s disease (PD) and healthy volunteers (HV). The participants are asked to wear the SMI Eye Tracking Glasses while waking through a corridor. Information regarding the areas of visual fixation is obtained and the percentages of time spent looking at specific areas are compared between different populations (Healthy Volunteers, patients with Parkinson’s Disease with and without Freezing of Gait).
Our study has enrolled a total of 24 participants:
15 patients with PD without freezing of gait
2 patients with PD and freezing of gait
7 healthy volunteers
Preliminary results: Our preliminary analysis demonstrates increased fixation duration and decreased saccade velocity on patients with PD when compared with HVs. More importantly, there are sight differences on the percentages of time spent looking at specific areas between the two populations. The differences are not statistically significant at this time, likely due to the small sample size.
September 2015 Project Update:
Methods Used and Research Subjects: We used the SensoMotoric Instruments Eye Tracking Glasses (SMI ETG) set-up. This records eye position binocularly, and superimposes the results on the recording from a front-facing camera. The result is a map of the explored visual environment. We used custom created software scripts to separate the visual scene into areas of interest (AOI). The analysis was performed using SPSS and MATLAB software, analyzing the relative percentage of time spent fixating on different AOIs, over the duration of the task. The resulting output offers an effective summary of where the patients are looking during gait.
A total of 19 PD patients without freezing of gait (PD), 9 PD patients with freezing of gait (FG) and 18 healthy volunteers (HV) participated in the study. The study had two components: for the dynamic task, patients walked in a realistic environment, consisting of a corridor in the NIH clinical center, including a doorway – which is particularly relevant, as freezing episodes often occur in doorways. In the “Static” experiment, patients explored static images in a fixed position.
Results: The data analysis is still being finalized and the manuscript is in preparation.
The crucial finding is that in patients with freezing of gait there is significantly less fixation on the “target” region, defined as the area of the opposing wall towards which the subjects are walking. This is specific only to FG subjects, i.e., occurring only in patients with PD with a history of freezing (determined by the FOG questionnaire), and not in the patients with PD who do not experience freezing (~30% of time vs ~65% of time).
On further analysis, it appears that FG patients fixate more on the ground. This does not appear related to posture or ability to explore relevant areas of the target – interestingly, the same pattern of visual exploration occurs in the static task, where the head is fixated in the same position for all subjects.
Significance: Our data may identify a clinic marker specific to freezing of gait. This is the first time a visual exploration paradigm has been applied in this manner. The results highlight specific visual exploration differences in patients with freezing of gait, that separate them not only from healthy volunteers, but also from PD patients who do not experience freezing of gait. This supports the notion that freezing is characterized by a distinct abnormality of gaze that may be a physiological factor contributing to the disorder.