Correlates of Psychological Distress in Patients with Parkinson's Disease During the COVID‐19 Outbreak

ABSTRACT

Background

Following the severe consequences of the COVID‐19 outbreak, on March 9, 2020, the Italian government implemented extraordinary measures to limit viral transmission, including restrictive quarantine measures. This resulted in a rapid and profound change of people's daily lives.

Objective

We assessed the psychological impact of the 40‐day quarantine in a large cohort of patients with Parkinson's disease (PD) and caregivers. Moreover, we analyzed whether prelockdown clinical features may be associated with subjective response of patients with PD to this traumatic event.

Methods

A total of 94 patients with PD were enrolled in the study. The Impact of Event Scale‐Revised, the Kessler Psychological Distress Scale, and the 12‐item Zarit Burden Inventory were obtained from patients and caregivers by email. A multivariate regression analysis was performed to determine whether prelockdown clinical motor and nonmotor features were associated with the psychological impact of lockdown.

Results

Regression analyses showed that prelockdown levels of anxiety, treatment‐related motor complications, patients' quality of life, and lockdown hours per day were significantly associated with psychological impact measures of the 40‐day quarantine. In addition, we showed that caregiver burden was correlated with overall patient autonomy and attention/memory impairment.

Conclusions

We identified specific PD motor and nonmotor features potentially predisposing to higher psychological impact of stressful situations, such as quarantine. This may help guide postpandemic interventions and preventive strategies to avoid further impairment of psychological well‐being in patients with PD.

In late 2019, a pneumonia cluster of unknown origin was identified in Wuhan, China. ¹ A novel coronavirus termed severe acute respiratory syndrome coronavirus 2 was isolated from human cases, and its genetic sequence was shared publicly in early January 2020. The disease caused by severe acute respiratory syndrome coronavirus 2, named coronavirus disease 2019 (COVID‐19), quickly spread to all Chinese provinces and to several other countries worldwide. Therefore, on March 11, 2020, the World Health Organization announced the COVID‐19 outbreak had reached pandemic levels. ²

The COVID‐19 pandemic has created unprecedented challenges for medical healthcare systems worldwide. Many countries have used rigorous global containment and quarantine efforts to limit infection diffusion. Italy has been significantly affected by COVID‐19 with respect to confirmed cases and related deaths. ³

Notwithstanding the importance of such containment measures for public health, social lockdown could have serious consequences on people's general health. ⁴ For instance, several studies on healthy individuals after previous pandemics reported a high prevalence of psychological symptoms such as mood alterations, ⁵ , ⁶ stress,⁷ and posttraumatic stress symptoms. ⁸

Lockdown measures to mitigate the spread of COVID‐19 have brought about a profound change in people's daily lives in the short term. According to Helmich and colleagues,⁹ successfully coping with such a sudden change requires cognitive operations that depend on the normal functioning of dopaminergic structures. In addition, there is compelling evidence linking impaired dopaminergic signaling and psychological posttraumatic stress. ¹⁰

Nigro‐striatal dopamine depletion is the most relevant pathophysiological process in Parkinson's disease (PD). Other pathways, mainly serotoninergic, adrenergic, GABAergic, and cholinergic, also have been shown to play a role in the pathogenesis of motor and nonmotor symptoms in PD. ¹¹

Similarly, these neurotransmitters have been variously shown to mediate stress‐related responses as well as resilience processes. ¹² Thus, patients with PD may be particularly vulnerable to negative psychological consequences following the COVID‐19 outbreak and lockdown measures. A few cross‐sectional surveys already have provided insights about patients with PD being particularly vulnerable to experiencing COVID‐19 outbreak‐related stress. ¹³ , ¹⁴ Here, we sought to extend this work by assessing the psychological distress and impact of a 40‐day quarantine in a large cohort of Italian patients with PD and their caregivers. We further explored whether prelockdown clinical features may be associated with the psychological impact of this traumatic event.

We hypothesized that specific PD‐related motor and nonmotor symptoms render some patients more prone to experiencing psychological distress following quarantine. These patients should deserve particular attention to target postpandemic treatment strategies.

Methods

Participants

The study sample was recruited from an ongoing longitudinal study enrolling patients with early PD diagnosed according to the modified diagnostic criteria of the UK Parkinson's Disease Society Brain Bank ¹⁵ at the Movement Disorders Unit of the First Division of Neurology at the University of Campania “Luigi Vanvitelli” (Naples, Italy). As previously described, ¹⁶ , ¹⁷ patients enrolled in this longitudinal study underwent an extensive motor and nonmotor assessment (please see “Prelockdown Measures” for further details) at the time they were diagnosed with PD (timepoint 0). After the baseline assessments, patients were prospectively followed with a full clinical evaluation every year (timepoints 1, 2, 3, 4 and 5, to date).

From this cohort, we selected a subset of patients with PD fulfilling the following inclusion criteria at their last timepoint: (1) at least 1‐year of PD diagnosis, to limit a potential misdiagnosis with other parkinsonisms; (2) last clinical follow‐up within 6 months of the current study and stable dopaminergic treatment over this period. Exclusion criteria were: (1) relevant cognitive impairment associated with PD according to consensus criteria ¹⁸ ; (2) major depression, minor depression, and dysthymic disorder according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria; and (3) any other neurological disorder or clinically significant medical condition. For each patient, clinical data from the last available timepoint were considered. Figure 1 shows the selection process for the subset of patients with PD/timepoints included in this study from the original longitudinal cohort.

FIG 1.

Selection process of the study cohort from the original ongoing longitudinal Parkinson's disease (PD) sample. Patients were considered for enrollment if they had at least a 1‐year PD diagnosis and the last timepoint visit was performed within 6 months from the present study. Moreover, the presence of relevant cognitive impairment associated with PD as well as major depression, minor depression, and dysthymic disorder were excluded. For each patient, clinical data from the last available timepoint were considered. Please refer to the text for further details.

The local ethical committee supervised and approved all procedures in accordance with the Declaration of Helsinki. According to the recent statement from the Italian Medicines Agency (www.aifa.gov.it) regarding COVID‐19 emergency‐related studies, informed consent was obtained by email from all participants.

Prelockdown Measures

During their last prelockdown timepoint visit, all patients underwent extensive motor and nonmotor assessments with validated PD‐related scales. Disease severity was assessed in the off state by means of the Unified Parkinson's Disease Rating Scale (UPDRS) Part III ¹⁹ and the Hoehn and Yahr stage. ²⁰ Moreover, the presence of treatment‐related motor complications was assessed by means of the UPDRS Part IV ¹⁹ (A–B subscales). The Schwab and England Scale (S&E Scale) ²¹ was used to measure the impact of PD on the patients' ability to cope with daily life.

Nonmotor symptoms presence and severity were assessed using the Non‐Motor Symptom Scale (NMSS). ²² Global cognitive functioning was assessed with the Montreal Cognitive Assessment, ²³ and the presence of relevant cognitive impairment associated with PD was excluded based on clinical judgment. Moreover, the Beck Depression Inventory (BDI) ²⁴ was used to measure depressive symptoms, the Parkinson Anxiety Scale (PAS) ²⁵ was used to measure anxious symptoms, the Apathy Evaluation Scale (AES) ²⁶ was used to measure apathetic symptoms, the Parkinson Fatigue Scale (PFS) ²⁷ was used to measure fatigue symptoms, and the Epworth Sleepiness Scale ²⁸ was used to measure excessive daytime somnolence. Finally, the Parkinson's Disease Questionnaire (PDQ‐39) ²⁹ was used to assess overall quality of life. The total amount of daily dopaminergic medication was computed using an algorithm adapted from Tomlinson and colleagues. ³⁰

Finally, the use of antidepressants and/or anxiolytics (not related to major psychiatric syndromes) was coded as binary variables (yes/no).

Psychological Impact of Lockdown

After a mean of 40 days of quarantine (38.2 ± 2.7, mean ± SD), we asked patients to fill out and return by email specific scales assessing the psychological stress and impact of the event. Two experienced psychologists (M.S. and V.S.E.) explained project details to the patients by phone calls.

The Impact of Event Scale–Revised (IES‐R) ³¹ was used to measure the psychological impact of the lockdown. The IES‐R includes 3 subscales addressing different symptom clusters of trauma phenomena, such as intrusion (ie, repeated thoughts about the event), avoidance (ie, effortful attempts not to think about the event), and hyperarousal (ie, anger, irritability, hypervigilance, difficulty concentrating).

The Kessler Psychological Distress Scale (K‐10) ³² was used as a global measure of psychological distress experienced by patients during the lockdown period.

The 12‐item Zarit Burden Inventory (ZBI) ³³ was used to assess caregiver burden.

Moreover, a brief survey was developed to quantify mean days and hours per day spent at home as well as other lockdown‐related features (eg, presence of the caregiver at home, employment status, work from home, COVID‐19 infection).

Statistical Analysis

First, we used bivariate regression analyses to identify associations between patient characteristics at the last prelockdown visit (ie, demographic, motor, and nonmotor features) as well as patient experiences of lockdown (ie, days of lockdown, lockdown hours per day, time since last visit, and presence of caregiver at home) with the psychological impact of lockdown (ie, IES‐R subscale, K‐10, and ZBI scores). Next, variables exhibiting statistically significant associations were entered as independent variables in multiple regression analyses (forward stepwise method) with the psychological impact measures as dependent variables. Finally, we checked the assumptions of multicollinearity (tolerance >0.20 ³⁴ and largest variance inflation factor ≤ 10 are acceptable ³⁵ , ³⁶ ) to make sure that the defined regression models could generalize beyond our sample. We used Statistical Package for Social Science (version 21) for all analyses, with P values <0.05 considered as statistically significant.

Results

A total of 132 patients with PD were eligible for the study; 38 patients with PD were contacted by phone, but were not able to receive and fill out the forms by email (Fig. 1). Therefore, 94 patients were enrolled in the study, having their last timepoint within 6 months from the survey collection (117.7 ± 49.9 days, mean ± SD; range, 43–193) (Fig. 1, Table 1).

TABLE 1.

Descriptive statistics

Variable	Mean (SD)/Count (%) a
Characteristics at last prelockdown visit
Clinical and motor features
Age at diagnosis, y	63.98 (8.95)
Age at last follow‐up, y	65.47 (8.95)
Sex, male a	62 (66.00%)
Disease duration, mo	39.16 (20.74)
Hoehn and Yahr	1.84 (0.51)
UPDRS Part III	22.34 (10.41)
UPDRS Part IV (A–B)	0.82 (1.77)
Treatment‐related motor complications, yes a	29 (30.85%)
S&E Scale	88.39 (10.45)
LEDD total	310.40 (176.88)
Antidepressants and/or anxiolytics, yes a	8 (8.50%)
Nonmotor measures
Montreal Cognitive Assessment	23.43 (4.05)
NMSS‐domain 1 (cardiovascular)	0.59 (1.41)
NMSS‐domain 2 (sleep/fatigue)	6.09 (6.70)
NMSS‐domain 3 (mood/cognition)	8.64 (12.21)
NMSS‐domain 4 (perceptual problems)	0.12 (0.62)
NMSS‐domain 5 (attention/memory)	2.38 (4.27)
NMSS‐domain 6 (gastrointestinal tract)	4.45 (5.60)
NMSS‐domain 7 (urinary)	6.48 (7.22)
NMSS‐domain 8 (sexual function)	1.38 (3.71)
NMSS‐domain 9 (miscellaneous)	4.33 (4.68)
Apathy Evaluation Scale	30.74 (8.62)
Parkinson Anxiety Scale	9.39 (8.38)
Beck Depression Inventory	7.84 (8.50)
Epworth Sleepiness Scale	4.32 (3.55)
Parkinson Fatigue Scale	2.12 (1.14)
Parkinson's Disease Questionnaire–39	30.24 (23.62)
Characteristics at time of COVID‐19 survey
Psychological impact measures
IES‐R intrusion	4.42 (4.60)
IES‐R avoidance	3.88 (4.95)
IES‐R hyperarousal	3.11 (3.74)
Kessler Psychological Distress Scale	14.73 (6.17)
Zarit Burden Inventory	4.77 (5.42)
Others features
Patients' days of lockdown	38.19 (2.77)
Patients' lockdown hours per day	21.74 (4.71)
Patients' time since last visit, mo	3.92 (1.97)
Caregiver present at home, yes a	85 (90.40%)
Caregivers' age, y	54.61 (15.23)
Caregivers' relationship with patient
Husband/wife a	81 (86.17%)
Son/daughter a	11 (11.70%)
Brother/sister a	2 (2.13%)

Abbreviations: SD, standard deviation; S&E Scale, Schwab and England Scale; UPDRS, Unified Parkinson's Disease Rating Scale; LEDD, levodopa equivalent daily dose; NMSS, Non‐Motor Symptoms Scale; COVID‐19, coronavirus disease 2019; IES‐R, Impact of Event Scale–Revised.

^aThe significance is already placed in the heading (Count/%).

During the quarantine period, patients included in the current analyses did not report any need for treatment of changes in their motor or nonmotor symptoms. They likewise did not report the need for psychiatric consultation.

Descriptive statistics are reported in Table 1.

The results of our bivariate regression analysis (ie, standardized β regression coefficients and P value) are reported in Table 2. We found that higher scores on the IES‐R avoidance subscale were associated with lower age at diagnosis or at last follow‐up, female sex, and higher scores on the UPDRS Part IV (A–B), domain 1 of the NMSS, AES, PAS, BDI, PFS, and PDQ‐39. Higher scores on the IES‐R hyperarousal subscale were associated with higher scores on the UPDRS Part IV (A–B); domains 1, 3, or 5 of the NMSS; AES; PAS; BDI; Epworth Sleepiness Scale; PFS; and PDQ‐39. Higher scores on the IES‐R intrusion subscale were associated with female sex and higher lockdown hours a day as well as higher scores on the UPDRS Part IV (A–B); domains 3, 4, 5, or 7 of the NMSS; AES; PAS; BDI; PDQ‐39; and higher levodopa equivalent daily dose total; domains 3, 4, 5, or 7 of the NMSS; AES; PAS; BDI; and PDQ‐39. Higher scores on the K‐10 were associated with female sex, higher disease duration, and time since last visit as well as higher scores on the UPDRS Part IV (A–B); domains 2, 3, 5, 7, or 9 of the NMSS; AES; PAS; BDI; Epworth Sleepiness Scale; PFS; and PDQ‐39; higher levodopa equivalent daily dose total; and lower scores on the S&E Scale. Higher scores on the ZBI were associated with higher scores on the UPDRS Part III; domains 2, 5, or 6 of the NMSS; BDI; PFS; and PDQ‐39; and lower scores on the S&E Scale.

TABLE 2.

Bivariate regression analyses for exploring variables associated with the IES‐R subscales, K‐10, and ZBI

		IES‐R
Variable	Avoidance	Hyperarousal	Intrusion	K‐10	ZBI
Characteristics at last prelockdown visit
Clinical and motor features
Age at diagnosis, y	‐0.29	−0.14	−0.02	−0.16	−0.10
Age at last follow‐up, y	‐0.27	−0.13	0.01	−0.13	−0.08
Sex (coded as: male = 0, female = 1)	0.24	0.20	0.34	0.26	−0.16
Disease duration, mo	0.05	0.02	0.13	0.21	0.11
Hoehn and Yahr	0.01	−0.05	0.12	0.07	0.03
UPDRS Part III	−0.12	−0.11	−0.06	0.09	0.28
UPDRS Part IV (A–B)	0.43	0.45	0.34	0.39	0.06
S&E scale	0.00	−0.06	−0.09	0.22	0.35
LEDD total	0.15	0.07	0.21	0.24	0.10
Antidepressants and/or anxiolytics (coded as: no = 0, yes = 1)	−0.02	−0.07	0.08	0.06	−0.11
Nonmotor measures
Montreal Cognitive Assessment	−0.11	−0.16	−0.10	−0.17	−0.08
NMSS‐domain 1 (cardiovascular)	0.23	0.33	0.20	0.18	0.21
NMSS‐domain 2 (sleep/fatigue)	0.14	0.09	0.02	0.23	0.36
NMSS‐domain 3 (mood/cognition)	0.18	0.22	0.22	0.40	0.13
NMSS‐domain 4 (perceptual problems)	−0.12	−0.02	0.21	0.02	−0.09
NMSS‐domain 5 (attention/memory)	0.10	0.22	0.30	0.26	0.46
NMSS‐domain 6 (gastrointestinal tract)	0.16	0.04	0.07	0.17	0.26
NMSS‐domain 7 (urinary)	0.14	0.04	0.25	0.21	0.21
NMSS‐domain 8 (sexual function)	−0.07	−0.05	−0.15	−0.04	−0.03
NMSS‐domain 9 (miscellaneous)	0.13	0.13	−0.01	0.23	0.11
Apathy Evaluation Scale	0.26	0.33	0.26	0.43	0.10
Parkinson Anxiety Scale	0.47	0.54	0.36	0.57	0.21
Beck Depression Inventory	0.46	0.52	0.33	0.57	0.29
Epworth Sleepiness Scale	0.20	0.33	0.10	0.33	0.21
Parkinson Fatigue Scale	0.24	0.23	0.08	0.38	0.35
Parkinson's Disease Questionnaire–39	0.30	0.38	0.29	0.56	0.39
Characteristics at time of COVID‐19 survey
Others features
Patients' days of lockdown	−0.01	0.02	0.10	0.19	−0.04
Patients' lockdown hours a day	0.19	0.10	0.27	0.17	0.12
Patients' time since last visit, mo	0.11	0.15	0.01	0.23	0.10
Caregiver present at home (coded as: no = 0, yes = 1)	0.08	−0.05	−0.15	0.05	0.00

The shown numbers correspond to standardized β coefficients or β associated to independent variables. Bold values indicate β standardized coefficients with P < 0.05.

Abbreviations: IES‐R, Impact of Event Scale–Revised; K‐10, Kessler Psychological Distress Scale; ZBI, Zarit Burden Inventory; UPDRS, Unified Parkinson's Disease Rating Scale; S&E Scale, Schwab and England Activities of Daily Living Scale; LEDD, levodopa equivalent daily dose; NMSS, Non‐Motor Symptoms Scale; COVID‐19, coronavirus disease 2019.

The multivariate regression analysis confirmed that higher scores on the IES‐R avoidance subscale were associated with higher scores on the UPDRS Part IV (A–B) and PAS. Higher scores on the IES‐R hyperarousal subscale were confirmed to be associated with higher scores on the UPDRS Part IV (A–B) and PAS. Likewise, higher scores on the IES‐R intrusion subscale were associated with higher scores on the UPDRS Part IV (A–B), PAS, and higher patients' lockdown hours a day. Higher scores on the K‐10 were confirmed to be associated with higher scores on the UPDRS Part IV (A–B), PAS, and PDQ‐39. Lastly, higher scores on the ZBI were associated with lower scores on the S&E Scale and higher scores on domain 5 of the NMSS. These results were not biased by multicollinearity as tolerance statistics were above 0.20, and the largest variance inflation factors were below 10 (Table 3).

TABLE 3.

Multivariate regression analyses for exploring the association between the IES‐R subscales, K‐10, and ZBI and independent variables resulting as statistically significant in bivariate analyses

			Model			Multicollinearity
Variable	β	F Test	P Value	R	R 2	Tolerance	VIF
IES‐R avoidance		21.38	<0.01	0.58	0.34
Age at diagnosis, y	−0.07					0.30	3.33
Age at last follow‐up, y	−0.06					0.30	3.33
Sex (coded as: male = 0, female = 1)	0.14					0.90	1.11
UPDRS Part IV (A–B)	0.35					0.67	1.49
NMSS‐domain 1 (cardiovascular)	−0.03					0.71	1.40
Apathy Evaluation Scale	−0.05					0.48	2.08
Parkinson Anxiety Scale	0.38					0.31	3.22
Beck Depression Inventory II	0.17					0.31	3.22
Parkinson Fatigue Scale	−0.01					0.54	1.85
Parkinson's Disease Questionnaire–39	−0.01					0.42	2.38
IES‐R hyperarousal		27.87	<0.01	0.63	0.40
UPDRS Part IV (A–B)	0.33					0.68	1.47
NMSS‐domain 1 (cardiovascular)	0.02					0.50	2.00
NMSS‐domain 3 (mood/cognition)	−0.06					0.44	2.27
NMSS‐domain 5 (attention/memory)	0.00					0.47	2.12
Apathy Evaluation Scale	0.44					0.52	1.92
Parkinson Anxiety Scale	0.46					0.35	2.85
Beck Depression Inventory II	0.18					0.30	3.33
Epworth Sleepiness Scale	0.06					0.60	1.66
Parkinson Fatigue Scale	−0.05					0.50	2.00
Parkinson's Disease Questionnaire–39	0.14					0.35	2.85
IES‐R intrusion		8.14	<0.01	0.54	0.29
Sex (coded as: male = 0, female = 1)	0.19					0.81	1.23
UPDRS Part IV (A–B)	0.23					0.71	1.40
LEDD total	0.04					0.75	1.33
NMSS‐domain 3 (mood/cognition)	0.05					0.43	2.32
NMSS‐domain 4 (perceptual problems)	0.17					0.88	1.13
NMSS‐domain 5 (attention/memory)	0.15					0.60	1.66
NMSS‐domain 7 (urinary)	0.11					0.75	1.33
Apathy Evaluation Scale	0.09					0.51	1.96
Parkinson Anxiety Scale	0.26					0.40	2.50
Beck Depression Inventory II	0.08					0.31	3.22
Parkinson's Disease Questionnaire–39	0.00					0.39	2.56
Patients' lockdown hours a day	0.21					0.83	1.20
K‐10		22.04	<0.01	0.67	0.45
Sex (coded as: male = 0, female = 1)	0.11					0.77	1.29
Disease duration, mo	0.12					0.59	1.69
UPDRS Part IV (A–B)	0.19					0.51	1.96
S&E Scale	0.00					0.39	2.56
LEDD total	0.00					0.57	1.75
NMSS‐domain 2 (sleep/fatigue)	−0.10					0.39	2.56
NMSS‐domain 3 (mood/cognition)	0.02					0.37	2.70
NMSS‐domain 5 (attention/memory)	−0.07					0.49	2.04
NMSS‐domain 7 (urinary)	0.05					0.67	1.49
NMSS‐domain 9 (miscellaneous)	0.01					0.77	1.29
Apathy Evaluation Scale	0.08					0.48	2.08
Parkinson Anxiety Scale	0.38					0.27	3.70
Beck Depression Inventory II	0.12					0.27	3.70
Epworth Sleepiness Scale	0.05					0.52	1.92
Parkinson Fatigue Scale	0.03					0.40	2.50
Parkinson's Disease Questionnaire–39	0.28					0.19	5.26
Patients' time since last visit, mo	0.16					0.74	1.35
ZBI		10.24	<0.01	0.52	0.27
UPDRS Part III	0.15					0.67	1.49
S&E Scale	‐0.26					0.34	2.94
NMSS‐domain 2 (sleep/fatigue)	0.02					0.61	1.63
NMSS‐domain 5 (attention/memory)	0.40					0.60	1.66
NMSS‐domain 6 (gastrointestinal tract)	0.08					0.68	1.47
Beck Depression Inventory II	0.12					0.42	2.38
Parkinson Fatigue Scale	0.16					0.50	2.00
Parkinson's Disease Questionnaire–39	0.03					0.30	3.33

Bold values indicate β standardized coefficients (β) with P < 0.05.

Abbreviations: IES‐R, Impact of Event Scale–Revised; K‐10, Kessler Psychological Distress Scale; ZBI, Zarit Burden Inventory; VIF, variance inflation factor; UPDRS, Unified Parkinson's Disease Rating Scale; NMSS, Non‐Motor Symptoms Scale; LEDD, levodopa equivalent daily dose; S&E Scale, Schwab and England Activities of Daily Living Scale.

Discussion

In the present study, we examined the psychological impact of a COVID‐19 quarantine in a large cohort of patients with PD and their caregivers and explored associations between quarantine‐related psychological stress and a wide range of prelockdown motor and nonmotor clinical features.

Our analyses provide novel evidence that the psychological impact of a 40‐day quarantine was significantly associated with prelockdown levels of anxiety symptoms, treatment‐related motor complications, patients' quality of life, and lockdown hours per day. In addition, we revealed that caregivers' burden during lockdown was strongly associated with overall prelockdown patients' physical independence and attention/memory impairment.

Stressful and traumatic events may trigger the development of 3 main different symptom clusters, such as intrusion (ie, flashbacks, repetitive intrusive memories, and distressing dreams), avoidance (ie, active attemps to not think and talk about the event), and hyperarousal (ie, exaggerated startle, sleep disturbance, hypervigilance, impulsivity, and aggressiveness). ¹⁰

We found that patients with PD with higher prelockdown levels of anxiety symptoms experienced more frequent thoughts about the lockdown (ie, higher IES‐R intrusion subscale scores), found it more effortful not to think about the lockdown (ie, higher IES‐R avoidance subscale scores), and felt more marked anger and irritability (ie, higher IES‐R hyperarousal subscale scores). Dopaminergic as well as serotoninergic and adrenergic pathways have been involved in both anxiety symptoms in PD ³⁷ and stress‐related responses. ¹² Interestingly, impairment of dopamine‐driven, top‐down inhibitory control from the prefrontal cortex over limbic structures has been shown to trigger intrusion and hyperarousal symptoms. ³⁸ Moreover, stimulation of nigro‐striatal dopaminergic neurons has been shown to enhance active avoidance behaviors in animal models. ³⁹

In previous pandemic situations, the prevalence of long‐term anxiety symptoms in healthy subjects forced to isolation have been shown. ⁴⁰ , ⁴¹ Moreover, patients with PD with anxiety symptoms and impaired quality of life have been shown to be less resilient to traumatic events. ⁴² This highlights the relevance of interventional strategies in the postlockdown phase aimed at reinforcing patients' psychological abilities for rapidly recovering from this stressful experience.

We also found that more severe prelockdown treatment‐related PD motor complications were associated with an overall higher psychological impact and greater experience of distress related to quarantine.

It has been shown that patients with posttraumatic stress disorder have increased striatal dopamine transporter density in comparison with trauma‐exposed asymptomatic individuals, ⁴³ potentially reflecting higher dopamine turnover. It has been proposed that higher dopamine turnover may contribute to the potentiation of exaggerated fear responses to a given event. ⁴³ As a consequence, situations that resemble a traumatic event may be interpreted as highly salient (driving attention, arousal, and motivation), leading to problems in everyday functioning. ⁴³ An increased dopamine transporter density within the striatum may be related to different mechanisms, such as reduced nigro‐striatal denervation as well as compensatory/maladaptive dopamine transporter up‐regulation ⁴⁴ or a combination thereof, potentially reflecting the presence of a high dopaminergic state at the synaptic level. Converging evidence supports the presence of high synaptic dopaminergic concentrations before and after levodopa administration in patients with PD with treatment‐related motor complications. ⁴⁵ , ⁴⁶ This may support the presence of a link between PD‐related pathophysiological mechanisms and dopaminergic‐driven maladaptive processing of traumatic events. However, pathophysiological processes underlying the experience of posttraumatic stress disorder symptoms in patients with PD have not been properly addressed and need further investigation.

It has to be acknowledged that other factors may potentially explain our observed associations. First, one may infer that patients with motor complications are more advanced, and thereby stressful situations may more readily trigger psychological impairment. However, it is important to note that our PD sample does not include patients with advanced PD. Moreover, we did not find any association between disease duration and severity (ie, disease duration in months, UPDRS Part III, and Hoehn and Yahr stage) and psychological well‐being.

An association between prelockdown severity of anxiety and motor complications may also be proposed. To our knowledge, this link has not been specifically addressed in PD yet. Interestingly, a large prospective cohort study showed that anxiety levels along with other motor and nonmotor features may predict future development of treatment‐related motor complications. ⁴⁷ This is consistent with our findings. However, our multiple regression analysis showed that UPDRS Part IV and PAS were associated with IES‐R and K‐10 independently, thereby suggesting that they may play distinct roles in moderating the impact of traumatic events on the psychological well‐being of patients with PD.

Finally, we found that caregivers' burden level under lockdown conditions (ie, higher ZBI) was associated with higher patients' impairment in daily activities (ie, lower S&E Scale). This is not surprising, as when patients' physical independence decreases, full‐time caregiving is required to deliver increasingly complex care. In line with our findings, previous studies showed that, along with mood disturbances, overall patients' disability predicts both caregiver stress and burden. ⁴⁸ Thus, postpandemic interventions should be targeted not only to improve the well‐being of patients with PD but also their caregivers with specifically designed therapeutic approaches.

It was also expected that patients' prelockdown quality of life would be associated with the psychological impact of quarantine.

We acknowledge that the present study has some limitations. To our knowledge, this is the first study aimed at investigating the COVID‐19 outbreak impact on psychological well‐being of patients with PD. However, psychological distress was only assessed postlockdown. Although our statistical analyses accounted for several clinically relevant features, we cannot rule out the possibility that other factors may have affected the psychological well‐being of our PD cohort during the quarantine period. Moreover, because we did not collect the same psychological distress scales in a control group, we cannot draw more solid conclusions about the pathophysiological processes potentially underlying our results. The association between scales specifically designed for patients with PD (eg, UPDRS Part IV) and psychological distress indicators (eg, IES and K‐10) may offer some exploratory insights about the presence of PD‐specific processes in the development of posttraumatic stress. However, we acknowledge that this is beyond the scope of our study. Future studies with repeated longitudinal measurement of psychological well‐being are needed to address this crucial issue.

The use of telemedicine in movement disorders has been suggested by the Movement Disorders Society under extraordinary conditions. ⁴⁹ However, cultural and technical issues may currently limit the implementation of this practice. For example, remote collection of clinical data may exclude patients lacking technological resources (eg, internet) or competency. Thus, our findings may be only relevant to patients for whom these are not limitations. Similarly, low‐quality video tools precluded us from performing a remote motor assessment in our PD cohort. This would have allowed us to monitor quarantine effects on global motor outcome as well as motor complications. The implementation of telemedicine tools for both patients and clinicians is no longer a deferrable need for providing better assistance to homebound patients with chronic neurological diseases. ⁵⁰

These limitations notwithstanding, we believe that our observations provide crucial insights about potential determinants of posttraumatic stress in patients with PD during the COVID‐19 outbreak. Patients with higher levels of anxiety as well as more severe treatment‐related motor complications should be particularly targeted for postpandemic psychological interventions.

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.

R.D.M.: 1A, 1B, 2A, 2C, 3A, 3B

M.S.: 1A, 2A, 2B, 2C, 3B

V.S.E.: 1C, 3A

A.G.: 2C, 3B

A.R.: 2C, 3B

G.T.: 1A, 3B

A.T.: 1A, 1B, 2C, 3B

Disclosures

Ethical Compliance Statement: The local ethical committee (University of Campania “Luigi Vanvitelli”) supervised and approved all procedures in accordance with the Declaration of Helsinki. According to the recent statement from the Italian Medicines Agency (www.aifa.gov.it) regarding COVID‐19 emergency‐related studies, informed consent was obtained by email from all participants. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest: There was no specific funding for this work. The authors declare that there are no conflicts of interest relevant to this work.

Financial Disclosures for the Previous 12 Months: The authors report no disclosures.

Supporting information

Table S1 Bivariate regression analyses exploring motor and nonmotor correlates of the Impact of Event Scale–Revised (IES‐R) subscales, Kessler Psychological Distress Scale (K‐10), and Zarit Burden Inventory (ZBI); numbers correspond to unstandardized coefficients for each predictor.