eTable 1. Baseline characteristics of controls and participants with clinically diagnosed HD.
eTable 2. Change in motor features in HD over 3 years.
eTable 3. Changes in cognitive measures in HD over 3 years (Mini-Mental State Examination, Verbal Fluency, and Symbol Digit Modalities Test).
eTable 4. Changes in cognitive measures in HD over 3 years (Stroop Interference Test).
eTable 5. Changes in behavioral measures in Huntington disease over three years.
eTable 6. Changes in functional measures in HD over 3 years.
eTable 7. Changes in body mass index and pulse in Huntington disease over three years.
eFigure 1. Participants included in this analysis.
eFigure 2. Distribution of baseline total functional capacity scores by baseline disease burden.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Dorsey ER, Beck CA, Darwin K, Nichols P, Brocht AFD, Biglan KM, Shoulson I, . Natural History of Huntington Disease. JAMA Neurol. 2013;70(12):1520–1530. doi:10.1001/jamaneurol.2013.4408
Copyright 2013 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Understanding the natural history of Huntington disease will inform patients and clinicians on the disease course and researchers on the design of clinical trials.
To determine the longitudinal change in clinical features among individuals with Huntington disease compared with controls.
Design, Setting, and Participants
Prospective, longitudinal cohort study at 44 research sites in Australia (n = 2), Canada (n = 4), and the United States (n = 38). Three hundred thirty-four individuals with clinically manifest Huntington disease who had at least 3 years of annually accrued longitudinal data and 142 controls consisting of caregivers and spouses who had no genetic risk of Huntington disease.
Main Outcomes and Measures
Change in movement, cognition, behavior, and function as measured by the Unified Huntington’s Disease Rating Scale, the Mini-Mental State Examination, and vital signs.
Total motor score worsened by 3.0 points (95% CI, 2.5-3.4) per year and chorea worsened by 0.3 point per year (95% CI, 0.1-0.5). Cognition declined by 0.7 point (95% CI, 0.6-0.8) per year on the Mini-Mental State Examination. Behavior, as measured by the product of frequency and severity score on the Unified Huntington’s Disease Rating Scale, worsened by 0.6 point per year (95% CI, 0.0-1.2). Total functional capacity declined by 0.6 point per year (95% CI, 0.5-0.7). Compared with controls, baseline body mass index was lower in those with Huntington disease (25.8 vs 28.8; P < .001), and average pulse was higher (74.2 vs 69.6 beats/min; P < .001).
Conclusions and Relevance
Over 3 years, the cardinal features of Huntington disease all declined in a monotonic manner. These data quantify the natural history of the disease and may inform the design of trials aimed at reducing its burden.
clinicaltrials.gov Identifier: NCT00313495
Huntington disease (HD) is an autosomal-dominant neurodegenerative disorder resulting from an unstable cytosine-adenine-guanine (CAG) repeat in the Huntingtin gene1 that is clinically characterized by involuntary movements, cognitive decline, and behavioral changes.2-4 Several studies have previously described the natural history of the disease.4-22 Many, however, have been either small, narrowly focused, short in duration, or uncontrolled. High-quality longitudinal data will inform the design of clinical trials aimed at slowing progression and may inform patients and clinicians about the disease course. We, therefore, analyzed longitudinal data from the recently concluded, prospective, multicenter, observational study Cooperative Huntington’s Observational Research Trial (COHORT).
The Johns Hopkins Medicine institutional review board reviewed and approved this analysis of COHORT study data. All COHORT study participants provided written informed consent or, if unable, had an authorized representative provide consent on their behalf. The COHORT project was an observational study that collected phenotypic and genotypic data from individuals with HD and their families from February 14, 2006, until June 30, 2011. The study design and initial baseline characteristics have been previously described.23 In brief, the study, which was conducted at 44 research sites in Australia (n = 2), Canada (n = 4), and the United States (n = 38), enrolled 2636 research participants who underwent standardized clinical assessments at baseline and annually thereafter.
COHORT research participants represented 4 groups: (1) individuals with clinically diagnosed HD; (2) individuals who underwent DNA testing prior to enrollment and were found to carry an expanded allele but did not have clinically diagnosed HD; (3) first-degree or second-degree relatives of individuals from the first 2 groups; and (4) spouses or caregivers of individuals from the first 2 groups (controls) who had no genetic risk for HD. We limited this analysis to control participants and those with clinically diagnosed HD at baseline, as determined by the site investigator’s answer of yes to question 80 of the Unified Huntington’s Disease Rating Scale (UHDRS)4: “Based on the entire UHDRS [Motor, Cognitive, Behavioral, and Functional components] do you believe with a confidence level of >99% that this subject has manifest HD?” and having a CAG repeat length of at least 36.24-26
At each visit, participants underwent a physical and neurological examination that included height (at baseline only), weight, the UHDRS, and the Mini-Mental State Examination (MMSE). At follow-up visits, new clinical and mental health events and each participant’s current medications were recorded. A blood sample for DNA isolation and CAG repeat genotyping was also collected at baseline.
Our primary analysis focused on the cohort of clinically diagnosed participants and controls who had at least 3 consecutive years of longitudinal data. We investigated motor, cognitive, behavioral, and functional outcomes and other clinical characteristics including body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared), pulse rate, and blood pressure. For motor outcomes, we examined changes in total motor and chorea scores. For cognition, we determined changes in verbal fluency, symbol digit modality, Stroop interference, and MMSE score. For behavior, we calculated the product of the frequency and severity scores for 11 behavioral questions (eg, depressed mood) and summed them. For function, we determined changes in the total functional capacity (TFC) score and functional assessment score.
To determine whether the results varied by baseline functional status or disease burden, we divided participants with HD into 2 different subgroups separated into quartiles. The first was defined by baseline (TFC), which assessed an individual’s capacity in occupation, finances, domestic chores, activities of daily living, and care level. The second was defined by baseline disease burden where disease burden = (CAGn − 35.5) × age.12
Clinical characteristics were analyzed in a cross-sectional manner using summary statistics at each point and longitudinally using mixed-effects models.27 The analysis incorporated data across all points by assuming a linear shape over time for the mean response and included random effects for centers with an unstructured correlation matrix for the repeated measures. These models were used to estimate the mean change per year in each outcome and the corresponding effect size (mean change divided by standard deviation of change) over 3 years. A simpler analysis that used only change from baseline to 3 years to estimate the mean change per year was conducted to confirm the results. Other mixed-effects model assumptions (eg, normality and linearity) were also verified. The focus of this analysis was on estimation, not hypothesis testing. We did not correct for multiple statistical comparisons. We report the number of individuals with missing data for each outcome measure in the corresponding Table. We reported this natural history study in keeping with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.28
During the study, 1514 participants with clinically diagnosed HD were enrolled, and 366 participants had at least 3 years of longitudinal data. Over the course of the 5-year study, 205 participants (13.5%) with clinical diagnosed HD prematurely withdrew from the study. The most common reasons were death (n = 61), withdrawal of consent (n = 36), and institutionalization (n = 31). Of the 366 participants, 32 did not have a Huntingtin allele with a CAG repeat length of at least 36 (24 did not have genotypic data available and 8 had a repeat length less than 36), leaving 334 for the purposes of this analysis (eFigure 1 in the Supplement). Of the 673 controls who enrolled at baseline, 142 participants had at least 3 years of longitudinal data.
Based on the baseline TFC of individuals with clinically diagnosed HD at baseline, the population was divided into the following 4 quartiles: quartile 1 (n = 58) had a TFC score of 13, quartile 2 (n = 95) had scores of 10 to 12; quartile 3 (n = 95) had scores of 7 to 9; and quartile 4 (n = 86) had scores from 0 to 6. Based on the baseline disease burden, the population was also divided into the following 4 quartiles: quartile 1 (n = 85) had scores of 0 to 341; quartile 2 (n = 82) had scores of 342 to 398; quartile 3 (n = 86) had scores of 399 to 456; and quartile 4 (n = 81) had scores of 457 or greater. Individuals with greater disease burden tended to have lower TFC (Spearman r = 0.80; P < .001) (eFigure 2 in the Supplement).
Table 1 and eTable 1 in the Supplement show baseline characteristics of the overall study population. Participants with clinically diagnosed HD were primarily female (56.0%) and white (93.4%), and most had completed at least 12 years of education (89.8%), although few were in the labor force (22.8%). At baseline, BMI was lower in those with HD compared with controls (25.8 vs 28.8; P < .001), and average pulse was higher than in controls (74.2 vs 69.6 beats/min; P < .001).
The total motor score increased by 3.0 points per year, and both chorea and dystonia scores increased by 0.3 point per year (Table 2 and eTable 2 in the Supplement). Total motor scores, chorea scores, and dystonia scores generally increased in all subgroups over time, although the rate of increase in chorea slowed substantially in the subpopulations with the most advanced disease.
Cognitive measures all declined over time (Table 3 and eTable 3 in the Supplement). Mean MMSE scores and mean verbal fluency scores both decreased by 0.7 point per year and mean symbol digit modality test scores decreased by 1.5 points per year. On the Stroop test, color naming scores declined by 1.8 points per year, word reading scores declined by 3.1 points per year, and interference scores declined by 1.3 points per year (Table 4 and eTable 4 in the Supplement). All these changes were significantly greater than the changes observed among controls (P < .001).
The mean product of behavior frequency and severity scores increased by 0.6 point per year as compared with the control group’s mean increase of 0.2 point per year (eTable 5 in the Supplement).
The mean TFC score among individuals with HD decreased by 0.6 point per year, and the mean functional assessment score decreased by 1.1 points per year (Table 5 and eTable 6 in the Supplement).
The average BMI among individuals with clinically diagnosed HD decreased by 0.1 unit per year. The higher pulse rate of individuals with HD did not change over 3 years (eTable 7 in the Supplement).
In the COHORT study, individuals with clinically diagnosed HD experienced a steady decline in the cardinal features of the disease. These results quantify the course of HD, draw attention to potential limitations in a commonly used behavioral measure, detail BMI changes, identify heart rate differences among individuals with HD, and may aid in designing and conducting HD clinical trials.
The COHORT study extends previous reports (Table 6) examining the natural history of HD.5-7,16-20,29,31-35 Our observed annual decline in total motor score of 3.0 points was less than the decline observed in some previous studies that ranged from 4.8 to 6.4 points.4,5,19,29 However, the recent 36-month analysis of TRACK-HD found an annual change in total motor score (2.9) similar to that observed in this study.18 Ravina et al29 found an annual worsening in chorea of 1.0 point among individuals with early HD that was greater than observed (0.3 point) in the present study that included individuals with early and advanced HD. Our analyses indicate that the annual rate of increase in chorea is greater among individuals with earlier-stage HD (0.6-0.7 point annual worsening) than advanced HD (0.0-0.1 point annual worsening). Thus, chorea likely increases early in the disease before reaching a plateau in more advanced HD.
The observed magnitude of cognitive changes was similar to previously published studies, including TRACK-HD.16-18 We found substantial differences between controls and individuals in the early stages of HD. For example, among all participants with unimpaired TFC scores of 13, those with HD (n = 58) scored on average 1 point worse on the MMSE and 10% to 20% worse on the verbal fluency, symbol digit, and Stroop tests than controls (n = 142) who had no HD risk. All cognitive metrics used to evaluate change in cognition showed progressive decline, and the MMSE had the smallest variance. Other cognitive measures including verbal fluency, symbol digit, and the Stroop test all showed progressive decline with only modest variance in participants with HD and improved scores (possibly due to learning effects) in the control population.
Unlike the motor and cognitive measures of HD, behavior changes, as measured by the UHDRS, were more difficult to assess. Perhaps because of a selection bias in enrolling participants with few disruptive behavioral features, we found only a modest worsening in behavior over time and wide confidence intervals that included zero (or no change). Behavioral impairment may not worsen as progressively or uniformly as motor and cognitive impairments. For example, despite depression being twice as prevalent compared with the general population,36 depression among our cohort did not progress over time.37 Thus, commonly used depression scales may have limited utility in HD.38 Additionally, measures of behavior in the UHDRS may not be sensitive to change. Other behavioral assessments, such as the Problem Behaviors Assessment for Huntington Disease,39 may have greater utility, especially because they can assess behavioral features of HD, such as apathy, that progress over time.18,37
The annual rate of decline in total functional capacity, a widely used assessment of overall functional capacity and a common primary outcome measure in HD clinical trials, ranges between 0.4 and 0.8 point per year among this large HD population. For example, Marder and colleagues6 observed a 0.7-point annual change in total functional capacity in a large (n = 960) study. In the 1980s, Penney and colleagues20 examined individuals with HD in Venezuela and found a TFC score decline of 1.4 points per year; this principal outlier may be related to the younger age of this population, the relative inaccessibility of these patients to health care, comorbid disorders, and cross-cultural application challenges of this assessment tool.
The COHORT study also found a substantial difference in the BMIs of individuals with clinically diagnosed HD (25.8) vs controls (28.8). A prior study found a difference of 1.8 to 2.3 units between individuals with HD and age- and sex-matched controls from the general population.40 The longitudinal decline in BMI among individuals with HD was 0.06 unit per year compared with an increase of 0.16 unit per year among controls. In a clinical trial of riluzole for HD, Aziz and colleagues41 found that BMI in HD decreased by 0.15 unit per year. A recent longitudinal study42 and the wide confidence intervals in this study indicate, however, that weight loss over time is not universal in HD. Studies of metabolites in both murine and human models of HD suggest that a hypermetabolic state may precede the occurrence of motor symptoms and contribute to weight loss.43-45 In support of that hypothesis, Marder and colleagues46 found that among 675 individuals, those premanifest for HD had higher caloric intake than controls but not higher BMI.
In our analysis, we found that pulse rate, but not blood pressure, was higher in individuals with HD (74.2 beats/min) compared with controls (69.6 beats/min) and at baseline was higher by 7 to 9 beats/min in individuals with advanced HD compared with individuals with less advanced HD. However, in longitudinal analyses, we observed no change in pulse over 3 years. This finding is new, and its mechanism and clinical implications are uncertain. Previous studies have suggested the presence of mild autonomic dysfunction as measured by sympathetic skin responses, heart rate variability, and autonomic cardiovascular tests.47,48 In early HD, no previous studies have clearly identified baseline differences in heart rate compared with controls or as a function of disease severity. In a small study, Kobal and colleagues48 concluded that autonomic dysfunction in presymptomatic and early symptomatic HD likely exists. Whether the differences observed are part of the underlying disease, a secondary phenomenon (eg, autonomic dysfunction, response to relative hypovolemia, or catabolic state), or even are replicable remains to be established. However, a recent longitudinal study suggests that a higher resting heart rate is an independent predictor of mortality among 2798 healthy individuals followed up for 16 years.49 Confirmation and a better understanding of our findings showing an unexplained relative tachycardia in HD may offer effective interventions to combat this associated comorbid phenomenon.
The COHORT study may also help investigators in designing therapeutic trials aimed at slowing the clinical and functional decline of HD. Our findings may assist in the selection of the study population, appropriate outcome measures, statistical power estimates, and evaluation of the clinical significance of outcomes. COHORT examined changes in relevant clinical outcomes for groups defined by functional capacity and by disease burden. With some notable exceptions, the rates of change in the cardinal clinical features were similar across subgroups. Among the exceptions was a plateau or slowing of chorea severity with more advanced disease as measured by functional capacity or disease burden and an accelerated decline in cognitive performance in those with greater disease burden. Interventions aimed at slowing chorea may be better focused on individuals with less advanced disease; however, treatments seeking to demonstrate a symptomatic effect on chorea may be best studied in those with prominent chorea who generally have more advanced disease. For interventions aimed at cognition, disease-modifying interventions could evaluate individuals in the earliest symptomatic stages of the disease when deficits begin to develop. Symptomatic interventions seeking to improve cognitive function could include those with more advanced disease, where the deficits and rates of decline are greatest.
The study results can also readily assist in estimating statistical power in designing trials. For example, a hypothetical 3-year placebo-controlled study aimed at reducing progression in the total motor score by 50% would have an expected treatment effect of about 4.5 points on the total motor score (from Table 2, total motor score increase by 3.0 points/y × 3 years × 50%). The SD of the change in the total motor score is about 12 points, which can be calculated by calculating the product of the standard error of change per year (0.22 points/y), the duration of observation (3 years), and the square root of the sample size (334). Assuming 90% power and a 2-sided significance level of 5% using a t test, the sample size per arm would be 158. For a more modest 20% improvement in the total motor score (treatment effect of 1.8 points), the sample size per arm would be 935. This analysis suggests, for example, that large sample sizes are likely required for identifying a treatment of moderate effect on motor function.
Finally, researchers, regulators, and clinicians might use these findings to evaluate the clinical significance of changes in different outcome measures observed in recent50,51 and future trials. For example, a recent clinical trial of pridopidine in HD found a treatment effect on 1 outcome of 2.8 points on the total motor score,51 which according to this study would equate to approximately a 1-year change among individuals with HD.
This controlled, prospective COHORT study has its limitations. First, the participants in this study are not representative of the entire HD population. As with other observational studies and clinical trials, participation in this study was voluntary and required the ability to travel to a participating research center. As such, the study population likely represents a more educated, more affluent, and less disabled population, limiting the generalizability of the study. Second, while the observational period was as long or longer than most previous studies, 3 years still represents only about 15% of the average duration of HD.52 High-quality data on disease progression over longer periods of observation are not yet available. Third, the frequency of observation in this study was annual. Shorter-duration clinical trials with more frequent and closer observation may have different, and potentially smaller, rates of decline than those found in this study. Fourth, the observed data are on individuals with HD who were taking medications, often prescribed by HD specialists. The rate of change in this study may not reflect that of the broader HD population who may have less access to such care. Fifth, this analysis was restricted to participants who had at least 3 follow-up visits. To the extent that those who prematurely withdrew from COHORT or those with missing data (especially from cognitive metrics) had more progressive disease, which is quite possible, the estimates provided here on disease progression would be understated. Finally, assessments in this study were unblinded and may differ from those conducted in clinical trials.
Notwithstanding these limitations, this study because of its design, size, geographic scope, prospective nature, rigorous assessments, and inclusion of contemporaneous controls provides valuable natural history information to patients and providers and may inform the design of clinical trials aimed at slowing functional decline and reducing the burden of HD.
Corresponding Author: E. Ray Dorsey, MD, MBA, 265 Crittenden Blvd, CU420694, Rochester, NY 14642 (firstname.lastname@example.org).
Accepted for Publication: July 22, 2013.
Published Online: October 14, 2013. doi:10.1001/jamaneurol.2013.4408.
Author Contributions: Drs Dorsey and Shoulson had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors were involved in the decision to submit the manuscript for publication.
Study concept and design: Dorsey, Beck, Biglan, Shoulson.
Analysis and interpretation of data: All authors.
Drafting of the manuscript: Dorsey, Darwin.
Critical revision of the manuscript for important intellectual content: Beck, Nichols, Brocht, Biglan, Shoulson.
Statistical analysis: Beck, Brocht.
Obtained funding: Dorsey, Shoulson.
Administrative, technical, and material support: Darwin, Nichols.
Study supervision: Dorsey.
Conflict of Interest Disclosures: Dr Dorsey is a consultant to Amgen, Clintrex, Lundbeck, and Medtronic and receives research support from Lundbeck and Prana Biotechnology. Dr Beck receives research support from Prana Biotechnology. Dr Biglan is a consultant to Lundbeck and receives research support from Lundbeck. Dr Shoulson is a consultant to Auspex, Lundbeck, and Prana Biotechnology and receives research support from Lundbeck and the US Food and Drug Administration. No other conflicts were reported.
Funding/Support: This analysis was funded by Lundbeck. CHDI Foundation, Inc funded the COHORT study.
Role of the Sponsor: The sponsor of this analysis (Lundbeck) had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation or approval of the manuscript; and decision to submit the manuscript for publication. The sponsor did review the manuscript prior to submission.
Group Information: The Huntington Study Group COHORT Investigators include the following individuals: Steering Committee: Ira Shoulson (principal investigator), University of Rochester, Rochester, NY; James Gusella (co–principal investigator), Massachusetts General Hospital, Boston; Tatiana Foroud (co–principal investigator), Indiana University School of Medicine, Indianapolis; Daniel P. Van Kammen, CHDI Foundation Inc, Los Angeles, California. Publications and Data Use Committee: Tatiana Foroud (chair), Indiana University School of Medicine; Ray Dorsey (co-chair), Johns Hopkins University, Baltimore, Maryland; John Warner and Joseph Giuliano, CHDI Foundation Inc; Louise Vetter, Huntington Disease Society of America, New York, New York; Oksana Suchowersky, University of Calgary, Calgary, Alberta, Canada; Christopher Beck and David Oakes, University of Rochester. Participating investigators and coordinators: Frederick Marshall and Charlyne Hickey, University of Rochester; Karen Marder, Steven Frucht, Carol Moskowitz, Ronda Clouse, Paula Wasserman, Lisa Muratori, and Elan Louis, Columbia University Medical Center, New York; Kathleen Shannon and Jeana Jaglin, Rush University Medical Center, Chicago, Illinois; Joseph Jankovic and Alicia Palao, Baylor College of Medicine, Houston, Texas; Madaline Harrison, Robert Davis, and Susan Dietrich, University of Virginia, Charlottesville; Carlos Singer and Monica Quesada, University of Miami, Miami, Florida; Steven Hersch, Diana Rosas, Kaloyan Tanev, Keith Malarick, and Robert McInnis, Massachusetts General Hospital; Amy Colcher, University of Pennsylvania, Philadelphia; Juan Sanchez-Ramos, University of South Florida, Tampa; Sandra Kostyk and Nick Doucette, Ohio State University, Columbus; Jane Paulsen, Ergun Uc, Robert Rodnitzky, Anne Leserman, and Stacie Vik, University of Iowa, Des Moines; Joel Perlmutter, Samer Tabbal, Amy Schmidt, and Stacey Barton, Washington University, St Louis, Missouri; Christopher Ross, Ray Dorsey, Frederick Nucifora, and Claire Welsh, Johns Hopkins University; Richard Dubinsky, Hilary Dubinsky, and Janice Broyles, University of Kansas Medical Center, Kansas City; Oksana Suchowersky and Mary Lou Klimek, University of Calgary; Randi Jones, Claudia Testa, and Stewart Factor, Emory University School of Medicine, Atlanta, Georgia; Elaine Sperin and Claudia Testa, Emory University, Atlanta; Dana Jennings, Institute for Neurological Disorders, John Morgan Medical College of Georgia, Augusta; Donald Higgins and Eric Molho, Albany Medical College, Albany, New York; John Adams, The Centre for Addiction and Mental Health, Toronto, Ontario, Canada; Samuel Frank, Marie Saint-Hilaire, and Melissa Diggin, Boston University, Boston; Sarah Furtado, University of Alberta, Edmonton, Alberta, Canada; Francis Walker, Christine O'Neill, and Vicki Hunt, Wake Forest University School of Medicine, Winston-Salem, North Carolina; Kimberly Quaid and Melissa Wesson, Indiana University School of Medicine; S. Elizabeth Zauber and Melissa Wesson, Indiana University, Indianapolis; Mark LeDoux, University of Tennessee Health Science Center, Memphis; Lynn Raymond, Blair Leavitt, and Joji Decolongon, University of British Columbia, Vancouver, British Columbia, Canada; Susan Perlman, University of California, Los Angeles; Jody Corey-Bloom, Guerry Peavy, and Jodi Goldstein, University of California, San Diego; Rajeev Kumar, Vicki Segro, and Diane Erickson, Colorado Neurological Institute, Englewood; Elizabeth McCusker, Jane Griffith, and Clement Loy, Westmead Hospital, New South Wales, Australia; Vicki Wheelock, Terry Tempkin, and Amanda Martin, University of California, Davis; Martha Nance, Hennepin County Medical Center, Minneapolis, Minnesota; Un Kang, University of Chicago, Chicago; William Mallonee and Greg Suter, Hereditary Neurological Disease Center, Wichita, Kansas; Fred Revilla and Maureen Gartner, University of Cincinnati/Cincinnati Children's Hospital, Cincinnati, Ohio; Carolyn Drazinic and Mary Jane Fitzpatrick, University of Connecticut, Storrs; Michel Panisset, Hôtel-Dieu Hôpital–CHUM, Montreal, Quebec, Canada; Kevin Duff, University of Utah, Salt Lake City; Burton Scott, Duke University Medical Center, Durham, North Carolina; William Weiner and Bradley Robottom, University of Maryland School of Medicine, Baltimore; Edmond Chiu, Olga Yastrubetskaya, and Andrew Churchyard, St Vincent's Aged Mental Health Service, Melbourne, Australia; Timothy John Greenamyre and Nancy Lucarelli, University of Pittsburgh, Pittsburgh, Pennsylvania; Pinky Agarwal, Booth Gardner Parkinson's Care Center, Kirkland, Washington; Irina Antonijevic, CHDI Foundation, Inc; Jeanen DeLaRosa, University of Texas, Galveston; Peter Panegyres, Neurodegenerative Disorders Research, Western Australia, Australia; Allison Coleman, University of British Columbia, Vancouver. Biostatistics/coordination center: David Oakes, Christopher Beck, Suzanne Robertson, Dustina Holt, Ken Eaton, Patricia Lindsay, and Lisa Deuel, University of Rochester. DNA genotyping center: Marcy MacDonald, Center for Human Genetic Research, Massachusetts General Hospital.
Additional Information: Dr Beck conducted and is responsible for the data analysis.