Abstract
Background
Anxiety is the most under‐recognized nonmotor symptom of Parkinson's disease (PD), yet it is unclear whether motor impairment exacerbates anxiety observed in PD, or vice versa. The current study examined: (1) whether movement (i.e., walking vs. standing) elevates distress in PD; (2) how dopaminergic treatment influences anxiety specifically while walking; and (3) whether these responses are worse in PD patients with gait impairments (compared to those without).
Methods
Twenty healthy control participants (HC), 17 PD participants without gait impairments (PD‐GI), and 14 PD participants with gait impairments (PD+GI) performed two tasks (stand vs. walk) in two virtual environments: (1) LOW threat; (2) HIGH threat. This protocol was completed in on and off dopaminergic states (to evaluate the effect of exacerbating motor symptoms).
Results
PD+GI reported greater levels of anxiety compared to PD‐GI and HC overall. All participants reported greater levels of anxiety and had higher skin conductance levels (SCLs) when walking compared to standing. The HIGH threat condition also generated greater levels of anxiety in all participants, compared to LOW threat, especially when required to walk. Notably, only PD+GI reported greater levels of anxiety when walking compared to standing in the LOW threat environment. Dopaminergic medication reduced self‐reported levels of anxiety, but did not significantly change SCL.
Conclusion
This study provides evidence that movement exacerbates anxiety in all older adults, but is particularly influential in those with gait impairments, which emphasizes the importance of optimally treating movement impairments as a method of reducing movement driven anxiety.
Keywords: anxiety, dopaminergic replacement therapy, Parkinson's disease, virtual reality, psychophysiology
Anxiety is becoming a well‐recognized nonmotor symptom of Parkinson's disease (PD) and has been associated with increased motor symptom severity,1 on/off fluctuations,2 and more‐severe gait problems.3 However, to date, anxiety has not been extensively studied considering it is one of the strongest predictors of quality of life in PD.4
Anxiety often arises before PD motor symptom onset and may even be a precursor to motor impairment.5, 6 Previous research has also shown that motor symptoms themselves can be exacerbated by anxiety (e.g., rigidity, bradykinesia, tremor, and, especially, gait deficits increase when patients are feeling distressed).7, 8, 9 Thus, it seems plausible that threatening situations (that induce anxiety) may drive motor impairment in PD. However, there is also evidence to support the opposite, that motor impairments may, in fact, elevate anxiety in PD. For example, anxiety fluctuations have been reported to mirror motor fluctuations,5, 10, 11 and dopaminergic replacement therapies have also been found to normalize motor deficits, leading to an amelioration of anxiety symptoms in PD.12 Taken together, it seems obvious that anxiety and movement impairments are related, although it remains unclear which is the chicken and which is the egg—or whether this is simply a reciprocal relationship?
Although there has been research that has investigated whether anxiety influences movement, to date, there have been no studies that have examined the alternative: whether movement itself leads to greater anxiety in those with PD. Thus, the current study investigated: (1) whether movement (i.e., walking vs. standing) elevates anxiety levels in PD (compared to controls); (2) how dopaminergic treatment influences anxiety specifically while walking; and (3) whether anxiety is worse in PD patients with gait impairments (compared to those without). It was hypothesized that if movement impairments in PD increase distress, then walking might lead to greater anxiety compared to standing, especially in the off dopaminergic state and in those PD participants with more severe gait impairments.
Materials and Methods
Participants
Twenty healthy control participants (HC), 17 PD participants without gait impairments (PD‐GI), and 14 PD participants with gait impairments (PD+GI) were recruited from the Sun Life Financial Movement Disorders Research and Rehabilitation Center at Wilfrid Laurier University (Waterloo, Ontario, Canada) and completed this study (see Table 1). Participants were excluded from the study if they could not walk 10 meters unassisted or had severe kyphosis, dementia, or other neurological disorders other than PD. Patient files were also carefully screened for comorbid conditions (i.e., visual impairments, hearing loss). The UPDRS motor section (UPDRS‐III)13 was administered by a certified clinician to assess disease severity. Additionally, all participants completed a series of baseline measures before completing the experimental protocol. These included the Modified Mini–Mental State Exam (3MS),14 State‐Trait Anxiety Inventory (STAI)15 assessing baseline levels of anxiety, Geriatric Depression Scale (GDS),16 the SCOPA‐AUT: questionnaire assessing autonomic nervous dysfunction (SCOPA‐AUT),17 and the modified clinical test of sensory interaction on balance (m‐CTSIB). Ethical approval was obtained by the research ethics boards at Wilfrid Laurier University and University of Waterloo. Written informed consent was obtained from all participants before participating according to the Declaration of Helsinki.
Table 1.
Demographic and clinical measures comparing PD in both dopaminergic states
| HC | PD‐GI | PD+GI | P‐value | |
|---|---|---|---|---|
| Age | 68 (9.5) | 66 (8.7) | 71 (7.8) | P = 0.35 |
| Sex | 6F; 14M | 3F; 14M | 3F; 10M | |
| UPDRS‐III (off) | – | 27 (9.8) | 40 (10.5) | * P = 0.007a |
| UPDRS‐III (on) | – | 20 (10.5) | 34 (10.2) | * P = 0.001a, * P < 0.0001b |
| PIGD score (off) | – | 2.6 (1.4) | 4.5 (1.3) | * P = 0.005a |
| PIGD score (on) | 1.7 (1.1) | 3.9 (2.2) | * P = 0.002a | |
| LDE | – | 223 (99) | 204 (63) | P = 0.54 |
| STAI‐Trait (off) | – | 36 (8.2) | 36 (8.4) | P = 0.94a |
| STAI‐Trait (on) | 31 (5.7) | 33 (6.6) | 33 (6.9) | P = 0.41a, * P = 0.01b |
| STAI‐State (off) | – | 35 (5.0) | 38 (10.3) | P = 0.40a |
| STAI‐State (on) | 28 (6.7) | 30 (5.9) | 34 (8.7) | P = 0.11a, * P = 0.005b |
| 3MS | 98 (2.7) | 96 (4.5) | 95 (7.0) | P = 0.14 |
| # Years of education | 15 (4.6) | 15 (3.5) | 15 (4.3) | P = 0.95 |
| GDS | 3 (3.0) | 7 (5.0) | 7 (3.6) | * P = 0.005 |
| SCOPA‐AUT | 8 (4.9) | 16 (4.4) | 16 (5.6) | * P < 0.0001 |
| Average m‐CTSIB (off) | – | 1.6 (1.2) | 1.5 (1.1) | P = 0.73 |
| Average m‐CTSIB (on) | 1.5 (0.9) | 1.8 (1.3) | 1.97 (1.1) | * P = 0.004a, * P = 0.047b |
aComparison between groups.
bComparison within PD between medication states.
*denotes significant findings at the P < 0.05 level.
PIGD, postural instability and gait disturbance; LDE, levodopa dose equivalence. UPDRS‐III: Unified Parkinson's disease rating scale motor subsection; LDE: Levodopa Dose Equivalence; STAI: State‐Trait Anxiety Inventory; 3MS: Modified Mini Mental State Exam; GDS: Geriatric Depression Scale; SCOPA‐AUT: questionnaire assessing autonomic nervous dysfunction; m‐CTSIB: modified clinical test of sensory interaction on balance.
Design and Procedure
Apparatus
The virtual environment (VE) used in this study was constructed using Vizard (Worldviz LLC, Santa Barbara, CA) and delivered using a high‐definition, low‐latency wireless link to a zSight head‐mounted display (HMD; Sensics Inc., Columbia, MD) that featured a 60‐degree field of view with 100% binocular overlap and a 1,280 × 1,024 full‐color pixels‐per‐eye resolution. The HMD weighed less than 3 pounds, and participants were able to move completely freely around the environment because of the wireless system that was carried by the patient in a backpack. In order to make the VE as immersive as possible, the experiment was completed in a dark room, which prevented participants from seeing the floor, their own feet, or the spotters' feet walking beside them. The viewpoint in the VE was controlled by seven OPTOTRAK Certus cameras (NDI Principles, Waterloo, Ontario, Canada) capturing and synchronizing the participants' position and movement using a rigid body with three infrared light‐emitting diodes that were attached to the HMD, allowing the viewpoint to update in real time creating an immersive virtual setting. The visual focus and eye width settings were adjusted for each participant to display a clear stereoscopic three‐dimensional image.
Q sensor cuffs were secured to the participants' left hand and collected skin conductance levels (SCLs) as a physiological measure of anxiety.18 The Q sensor collected SCLs at a frequency of 8 Hz from the hypothenar eminence, which has been recommended for ambulatory experimental paradigms.18 Synchronization of the start and end of each trial was achieved with a button press from the researcher. The m‐CTSIB was collected at the beginning of the experiment using a BIODEX Balance System SD in order to quantify baseline standing balance deficits.
Experimental Procedure
Participants were instructed to stand on a GAITRite sensor carpet, which was calibrated to visually display the start position (on a platform) in virtual reality (VR; technical specifications about the VR are reported in a previous work8). For each trial, participants were asked to either: (1) remain standing on the starting platform (3 m long × 3 m wide) for 60 seconds (STAND) or (2) walk across the virtual plank (6 m long × 1 m wide) to the opposite platform (WALK). This “Stand” or “Walk” instruction was also posted on a large black screen within VR. These tasks were carried out in two different threat conditions: (1) LOW threat: platform and plank was located on the floor; and (2) HIGH threat: when the floor surrounding the platform and plank descended creating a deep pit (~8 m) below. Because participants did not have any previous experience with virtual reality, everyone was provided with at least two practice walking trials through the LOW virtual environment. After these practice trials, participants completed 20 trials that were randomized for task and condition. After each trial, a 9‐point Self‐Assessment Manikin (SAM) scale was displayed and participants were asked to rate their feelings of stress and anxiety. Then, the VR HMD would present a black screen and the patient returned to the start position for the next trial. A standing rest period of 30 seconds (while screen was black) was given before each trial to prevent carryover effects from anxiety on the previous trial. It should be noted that all participants completed the experiment without the use of assistive walking devices.
In order to assess how dopaminergic treatment influenced anxiety and motor symptom severity, PD participants were asked to complete the full protocol on two separate occasions: once off their dopaminergic medication (minimum 12 hours withdrawal) and on their optimal dosage of medication. Completion of this study on and off dopaminergic medication was counterbalanced across participants. Only 7 of the 14 PD+GI participants were able to complete the experiment in their off state.
In addition to assessing baseline state and trait anxiety using the STAI before completing the experiment, participants' level of anxiety was also measured during the experimental protocol using a subjective self‐reported measure, where participants were asked to rate their feelings of stress and anxiety after each trial (which were described to each participant at the beginning of the study as: physical feelings such as rapid heartbeat, queasiness, shaking or quivering, and sweating, as well as emotional feelings such as worry, fear, and nervousness) by using a SAM scale,19 which visually displayed these features. Additionally, physiological arousal (i.e., SCL) was also collected during each trial, which represented an objective measure of stress and anxiety. SCLs were measured using the Q sensor, 5 seconds after the presentation of the virtual environment until the end of the trial. SCLs were calculated by taking the average SCL over that time period and then normalized by subtracting the average SCL during the 30‐second baseline “rest” period before each trial.18
Statistical Analysis
Baseline measures are reported in Table 1 and were statistically compared between groups with a one‐way analysis of variance (ANOVA) and also compared within PD participants (off vs. on) using dependent t tests in order to evaluate the effects of their dopaminergic replacement therapy. Three factor mixed repeated‐measures ANOVAs (group [HC, PD‐GI {on}, PD+GI {on}] × condition [LOW, HIGH] × task [Stand, Walk]) and four factor mixed repeated‐measures ANOVAs (group × medication state × condition × task) were used to examine the influence of motor impairment on two different outcome measures of anxiety. The primary outcome measure was participants' self‐reported anxiety levels using the SAMs after each trial, and the secondary outcome measure was average SCL on each trial. One PD‐GI participant and 1 PD+GI were removed from the skin conductance analysis because of equipment failure.
Results
Comparison Between HC, PD‐GI, and PD+GI
There was a main effect of group (F (2,48) = 19.58; P < 0.0001) for self‐reported anxiety. Tukey's post‐hoc analysis revealed that PD+GI participants reported significantly greater levels of anxiety during the experiment compared to PD‐GI (P = 0.0001) and HC participants (P = 0.0002). A main effect of condition (F (1,48) = 53.23; P < 0.0001) revealed that all participants reported significantly greater levels of anxiety in the HIGH threat environment compared to LOW threat environment (regardless of task, i.e., standing or walking). A main effect of task (F (1,48) = 53.03; P < 0.0001) showed that participants also reported greater anxiety when walking compared to standing. Importantly, there was also a significant interaction between condition and task (F (1,48) = 28.02; P < 0.0001) for self‐reported anxiety. Tukey's post‐hoc analyses revealed that all participants reported significantly greater levels of anxiety when required to walk compared to stand in the HIGH threat environment (P = 0.0002). However, a planned comparison revealed that PD+GI reported significantly greater anxiety when walking compared to standing in the LOW threat environment (t (13) = 2.15; P = 0.025), whereas the PD‐GI reported similar levels of anxiety when walking compared to standing in the LOW threat environment (t(16) = 1.48; P = 0.08; see Fig. 1).
Figure 1.
Overall PD+GI reported greater levels of anxiety compared to HC and PD‐GI (P < 0.001). All participants reported greater anxiety when required to walk compared to stand in the HIGH threat environment (P < 0.001). Notably, only the PD+GI group reported significantly greater anxiety when walking compared to standing in the LOW threat environment (P = 0.05).
Similar results were found for SCLs, where a significant interaction between condition and task (F (1,45) = 5.57; P = 0.023) was found. Tukey's post‐hoc analyses revealed that all participants had significantly higher SCLs when walking compared to standing in both the LOW (P = 0.002) and HIGH (P = 0.0002) threat environments. Additionally, Tukey's post‐hoc analyses revealed that all participants reported significantly greater levels anxiety when walking (but not standing) in the HIGH threat environment compared to the LOW threat environment (P = 0.016). There was also a significant interaction between condition and group (F (2,45) = 7.61; P = 0.001) for SCLs. Tukey's post‐hoc analysis revealed that only the HC group increased SCLs significantly (P = 0.005) when in the HIGH threat environment compared to the LOW threat environment. Finally, there was also a significant interaction between task and group (F (2,45) = 3.51; P = 0.038). Tukey's post‐hoc analysis revealed that HC and PD‐GI groups both increased SCLs significantly when walking compared to standing (HC: P = 0.0003; PD‐GI: P = 0.012), whereas PD+GI did not significantly increase SCLs when walking compared to standing (PD+GI: P = 0.98). Further analyses confirmed that this difference was not the result of a ceiling effect in the PD+GI group (please see Data S1 and Figure S1 for more details).
Off Versus On Dopaminergic Medication
When examining only PD‐GI and PD+GI groups, a significant interaction between condition and task (F (1,22) = 27.66; P < 0.0001) for self‐reported anxiety was found. Tukey's post‐hoc analyses revealed that all participants reported the greatest amount of anxiety when walking across the HIGH plank (P = 0.0002; Fig. 2). There were also interactions between condition and group (F (1,22) = 4.30; P = 0.05) and between task and group (F (1,22 = 7.42; P = 0.012). These significant interactions were followed up with Tukey's post‐hoc analyses, which revealed that both PD groups reported greater levels of anxiety in the HIGH threat environment compared to the LOW (PD‐GI: P = 0.03; PD+GI: P = 0.001) and greater levels of anxiety when required to walk compared to stand (PD‐GI: P = 0.002; PD+GI: P = 0.0002). When collapsed across all PD participants, a main effect of medication state (F (1,23) = 4.83; P = 0.03) revealed that dopaminergic medication reduced self‐reported levels of anxiety overall in PD.
Figure 2.
Only half of the PD+GI participants were able to complete the experiment in their off state, so statistical analysis was collapsed across all PD. Dopaminergic medication reduces self‐reported levels of anxiety overall in PD (P = 0.03). Although speculative, it appears that both PD groups do not receive the same benefit, but rather PD‐GI had greater change when on dopaminergic medication.
Results from skin conductance showed a nonsignificant effect of group (F (1,22) = 3.18; P = 0.088), which revealed a trend that suggests that PD+GI had slightly higher levels of skin conductance compared to PD‐GI. There was also a significant interaction between condition and task (F (1,22) = 13.81; P = 0.001). Tukey's post‐hoc analyses showed that both PD‐GI and PD+GI had significant increases when required to walk in the HIGH threat environment compared to the LOW threat environment (P = 0.001); however, there was no difference in skin conductance between standing in the HIGH and LOW threat environments (P = 0.99). Finally, there was a nonsignificant interaction between condition, task, and group (F (1,22) = 4.02; P = 0.057), which revealed a trend that only the PD‐GI group demonstrated a significant increase in SCLs when walking compared to standing during the LOW threat environment, whereas the PD+GI group had similar SCLs when standing and walking in the LOW threat environment. Notably, there were no group differences for SCL in either condition or task. Dopaminergic medication did not significantly influence SCLs (F (1,22) = 1.07; P = 0.31).
Discussion
Although studies have established that anxiety influences movement impairments,7, 8, 9, 20 this is the first study to provide evidence that movement may also exacerbate anxiety. Results from the current study showed that walking in a virtual environment compared to standing elicited greater levels of anxiety across all participants, specifically walking in the HIGH threat environment. Thus, it appears that movement contributes to anxiety in PD (as well as healthy older adults) when walking in highly threatening situations.
Perhaps the most interesting result of the study was that, even in nonthreatening environments, walking led to greater levels of anxiety compared to standing specifically in PD patients with gait impairments. These findings are discussed in greater detail below, along with the implications for clinical practice in PD. Notably, at baseline, PD+GI had worse symptom severity, but were matched for baseline state and trait anxiety levels as well as all other baseline measures. Yet, during the experiment, PD+GI reported greater levels of anxiety (regardless of task or condition) compared to HC and PD‐GI. Given that the only difference between these two PD groups at baseline was symptom severity, these findings suggest that their movement impairment, namely, their gait impairment, contributed to their elevated anxiety levels during the experiment. In support of this, anxiety levels (i.e., state and trait) at baseline, across all PD, were worse in their off state when motor impairment was also exacerbated. Dopaminergic medication was also found to reduce both baseline anxiety, self‐reported levels of anxiety during the experiment, and motor symptom severity in all PD participants. However, conclusions should be made with caution with respect to the influence of dopaminergic replacement therapy on anxiety in PD because of the high number of dropouts from the PD+GI group and the well‐known gait improvements as a result of dopaminergic therapy. Alternatively, it may be that the off state in PD may aggravate already present anxiety symptoms rather than “cause” anxiety, which has also been previously suggested.21 Further research is needed to fully understand the influence of dopaminergic medication on anxiety, and whether it simply mediates the influence of movement symptoms or actually alleviates or aggravates anxiety.
Another important limitation in this study was the comparison of skin conductance between groups, given that individuals with PD had greater autonomic dysfunction compared to healthy age‐matched adults, which can influence and often dampen the skin conductance signal. Therefore, these results should be interpreted with caution and might explain why healthy controls had greater SCLs compared to the PD+GI group, whereas PD+GI reported greater levels of anxiety. For this reason, SCLs were used in this study as a secondary measure to support the within‐subject self‐report ratings participants gave after each trail.
In conclusion, the use of a VR paradigm and a novel skin conductance measure was able to help demonstrate that movement impairments influence anxiety levels in PD. Overall, these results support that motor disability is among the strongest predictors of anxiety in PD,22 while emphasizing that PD patients with gait impairments are especially susceptible and sensitive to movement‐induced anxiety. It appears that the degree of one's movement impairment might scale their anxious reaction to movement depending on the threat of the environment. Thus, it is important in clinical practice to carefully consider how to optimally treat anxiety and whether there may be a difference in responsiveness depending on the degree of motor impairment. Clinicians might also consider the importance of optimally treating movement impairments as a method of reducing movement driven anxiety, which has the potential to increase the effectiveness of treatment for anxiety in PD and thus improve their quality of life.
Author Roles
(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique.
K.E.M.: 1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B
C.G.E.: 1A, 2C, 3B
Q.J.A.: 1A, 2C, 3B
Disclosures
Funding Sources and Conflicts of Interest: K.E.M. is supported by the Canadian Institute for Health Research and Parkinson Society Canada. C.G.E. is supported by the Natural Sciences and Engineering Research Council of Canada and the Canadian Foundation for Innovation. Q.J.A. is supported the Natural Sciences and Engineering Research Council of Canada and the Canadian Foundation for Innovation. The authors report no conflicts of interest.
Financial Disclosures for previous 12 months: The authors declare that there are no additional disclosures to report.
Supporting information
Data S1. Skin conductance: investigating a ceiling effect.
Figure S1. Skin conductance: ceiling effect comparison within PD+GI group.
Relevant disclosures and conflicts of interest are listed at the end of this article.
References
- 1. Siemers ER, Shekhar A, Quaid K, Dickson H. Anxiety and motor performance in Parkinson's disease. Mov Disord 1993;8:501–506. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8232361. [DOI] [PubMed] [Google Scholar]
- 2. Henderson R, Kurlan R, Kersun JM, Como P. Preliminary examination of the comorbidity of anxiety and depression in Parkinson's disease. J Neurophyschiatry Clin Neurosci 1992;4:257–264. [DOI] [PubMed] [Google Scholar]
- 3. Vazquez A, Jimenez‐Jimenez FJ, Garcia‐Ruiz P, Garcia‐Urra D. “Panic attacks” in Parkinson's disease. Acta Neurol Scand 1993;87:14–18. [PubMed] [Google Scholar]
- 4. Hanna KK, Cronin‐Golomb A. Impact of anxiety on quality of life in Parkinson's disease. Parkinsons Dis 2012;2012:640707 Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3236448&tool=pmcentrez&rendertype=abstract. Accessed November 22, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Pontone GM, Williams JR, Anderson KE, et al. Prevalence of anxiety disorders and anxiety subtypes in patients with Parkinson's disease. Mov Disord 2009; 24:1333–1338. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2830642&tool=pmcentrez&rendertype=abstract. Accessed December 5, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Jacob E, Gatto N, Thompson A, Bordelon Y, Ritz B. Occurrence of depression and anxiety prior to Parkinson's disease. Parkinsonism Relat Disord 2010;16:576–581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nuti A, Ceravolo R, Piccinni A, et al. Psychiatric comorbidity in a population of Parkinson's disease patients. Eur J Neurol 2004;11:315–320. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15142224. [DOI] [PubMed] [Google Scholar]
- 8. Ehgoetz Martens KA, Ellard CG, Almeida QJ. Does anxiety cause freezing of gait in Parkinson's disease?. PLoS One 2014;9:e106561 Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=4175083&tool=pmcentrez&rendertype=abstract. Accessed December 4, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Knight R, Godfrey H, Shelton E. The psychological deficits associated with Parkinson's disease. Clin Psychol Rev 1988;8:391–410. [Google Scholar]
- 10. Richard IH. Anxiety disorders in Parkinson's disease. Adv Neurol 2005;96:42–55. [PubMed] [Google Scholar]
- 11. Witjas T, Kaphan E, Azulay JP, et al. Nonmotor fluctuations in Parkinson's disease: frequent and disabling. Neurology 2002;59:408–413. [DOI] [PubMed] [Google Scholar]
- 12. Maricle RA, Nutt JG, Valentine RJ, Carter JH. Dose‐response relationship of levodopa with mood and anxiety in fluctuating Parkinson's disease: a double‐blind, placebo‐controlled study. Neurology 1995;45:1757–1760. [DOI] [PubMed] [Google Scholar]
- 13. Goetz CG, Fahn S, Martinez‐Martin P, et al. Movement Disorder Society‐sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS‐UPDRS): process, format, and clinimetric testing plan. Mov Disord 2007;22:41–47. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17115387. Accessed October 7, 2014. [DOI] [PubMed] [Google Scholar]
- 14. Teng EL, Chui HC. The Modified Mini‐Mental State (3MS) examination. J Clin Psychiatry 1987;48:314–318. [PubMed] [Google Scholar]
- 15. Spielberger C. State‐Trait Anxiety Inventory (Form Y). Redwood City, CA: Mind Garden; 1987. [Google Scholar]
- 16. Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 1983;17:37–49. [DOI] [PubMed] [Google Scholar]
- 17. Visser M, Marinus J, Stiggelbout AM, Van Hilten JJ. Assessment of autonomic dysfunction in Parkinson's disease: the SCOPA‐AUT. Mov Disord 2004;19:1306–1312. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15390007. Accessed January 14, 2015. [DOI] [PubMed] [Google Scholar]
- 18. Boucsein W. Electrodermal Activity. 2nd ed New York: Springer, US; 2012. [Google Scholar]
- 19. Bradley M, Lang PJ. Measuring emotion: the Self‐Assessment Manikin and the Semantic Differential. J Behav Ther Exp Psychiatry 1994;25:49–59. [DOI] [PubMed] [Google Scholar]
- 20. Ehgoetz Martens KA, Ellard CG, Almeida QJ. Anxiety‐provoked gait changes are selectively dopa‐responsive in Parkinson's disease. Eur J Neurosci 2015;42:2028–2035. [DOI] [PubMed] [Google Scholar]
- 21. Leentjens AFG, Dujardin K, Marsh L, Martinez‐Martin P, Richard IH, Starkstein SE. Anxiety and motor fluctuations in Parkinson's disease: a cross‐sectional observational study. Parkinsonism Relat Disord 2012;18:1084–1088. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22771284. Accessed November 30, 2014. [DOI] [PubMed] [Google Scholar]
- 22. Stefanova E, Ziropadja L, Petrovic M, Stojkovic T, Kostic V. Screening for anxiety symptoms in Parkinson disease: a cross‐sectional study. J Geriatr Psychiatry Neurol 2013;26:34–40. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23407399. Accessed December 5, 2014. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1. Skin conductance: investigating a ceiling effect.
Figure S1. Skin conductance: ceiling effect comparison within PD+GI group.