Increased depressive symptoms in Parkinson’s disease during the COVID‐19 pandemic: Preliminary findings from longitudinal analysis of the PHASE study

Introduction The cumulative number of patients has increased through the four waves of the pandemic in Japan. Many people experienced mental stress due to the fear of infection, and restrictions of leaving the house and leisure activities. No longitudinal study has assessed the fluctuation of neuropsychiatric symptoms during the COVID-19 pandemic using the same scale. We examined changes in non-motor symptoms, and the scores of a Parkinson’s Disease (PD)-specific questionnaire between the early and later periods during the COVID-19 pandemic. Methods We conducted a questionnaire survey during the first wave (from February to April 2020) and the fourth wave of the COVID-19 pandemic (from March to April 2021). We compared the number of symptoms from the two periods. Results Compared with the first wave, the Geriatric Depression Scale score was significantly higher in the fourth wave of the pandemic (median score of GDS: 4.00 vs. 5.50, p = 0.022). Consistently, the scores of symptoms on MDS-UPDRS part 1 in the fourth wave were significantly higher in hygiene (p = 0.033), handwriting (p = 0.033), performing hobbies and other activities (p = 0.035), and turning in bed (p = 0.046) than in the first wave. Conclusions Our observation over a year between the early and later phases of the COVID-19 pandemic showed an increase in the severity of depression in patients with PD.


Introduction
Nara is a local commuter town in Japan.The first wave of the COVID-19 pandemic occurred in Nara in April 2020, and a state of emergency was declared in Japan.The state included self-strict regulations, such as staying at home and refraining from going out as much as possible, wearing a mask, and avoiding three conditions: (1) closed spaces (sealed spaces with poor ventilation), 2) crowded places (many people are crowded together), and 3) close-contact settings (conversations within reach of each other, utterance is made).Since then, the cumulative number of patients has increased through the four waves of the pandemic in Japan (Fig. 1).During the repeated COVID-19 waves, many people experienced mental stress due to the fear of infection, and restrictions of leaving the house and leisure activities.Patients with Parkinson's disease (PD) may be at a higher risk of developing neuropsychiatric symptoms because they have difficulties adapting to a drastic change in the environment due to impaired functioning of the dopaminergic system [1].Previous cross-sectional studies have shown an increase in neuropsychiatric symptoms after the COVID-19 pandemic [2] and after the lockdowns [3][4][5].To our knowledge, no longitudinal study has assessed the fluctuation of neuropsychiatric symptoms during the COVID-19 pandemic using the same scale.We examined changes in non-motor symptoms, including depression, apathy, fatigue, and the scores of a PD-specific questionnaire between the early and later periods during the COVID-19 pandemic.

Study population
We launched a prospective cohort study among 202 PD patients who were enrolled in the PHASE study (Parkinson's disease and the relationships with circadian biological rhythms and sleep).Consecutive patients fulfilled the UK Parkinson's Disease Society Brain Bank criteria [6].The study participants were divided into a first cohort (n = 161) and a second cohort (n = 41) depending on the timing of the baseline survey.These patients had none of the following conditions at baseline: possible or probable multiple system atrophy according to the Gilman criteria [7]; progressive supranuclear palsy according to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [8]; dementia with Lewy bodies as per McKeith and colleagues [9]; any other atypical parkinsonian syndrome such as vascular parkinsonism; or large vessel disease, infarction, or tumors on magnetic resonance imaging.The present study includes 22 patients of the second cohort who completed a questionnaire survey about psychiatric symptoms both in the first and the fourth wave of the COVID-19 pandemic.The timing of the survey of the first wave was when 3 years passed since the inspection of the PHASE study.Until the fourth wave of the COVID-19 pandemic, we excluded 19 patients from this study due to death (n = 5), bone fractures (n = 4), aspiration pneumonia (n = 1), unable to follow our instructions due to dementia (n = 6), and three patients who did not respond to the questionnaire.We identified that the enrolled 22 patients had not been infected with coronavirus-2 (SARS-CoV-2) infection when these patients visited our hospital after the fourth wave of the COVID-19 pandemic.No patient underwent surgical interventions, including deep brain stimulation surgery.

Assessment of symptoms
We conducted a questionnaire survey during the first wave (from February to April 2020) and the fourth wave of the COVID-19 pandemic (from March to April 2021).The questionnaire included the Geriatric Depression Scale (GDS-15) [10], Starkstein's Apathy scale [11], Parkinson's Fatigue scale [12], rapid eye movement (REM) sleep behavior disorder (RBD) screening questionnaire [13], and the Japanese version of the Movement Disorder Society Revision of the Unified PD Rating Scale (MDS-UPDRS) part 1 [14].The disease stage was rated on the Hoehn-Yahr stage [15].Validation study on the basis of a DSM-IV diagnosis for depression showed high discriminant validity of the GDS-15 with comparison of the Hamilton Depression Rating Scale (HDRS) as a screening to distinguish depressed from nondepressed patients in PD [16].

Statistical analysis
We present the median and interquartile range (IQR) for continuous variables.To compare the number of symptoms from the two periods, we used the paired t-test and Wilcoxon signed-rank test for continuous variables with normal distribution and those with skewed distribution, respectively.We conducted the McNemar test to compare the proportion of patients with GDS ≥ 6, Starkstein's Apathy scale ≥ 16, Parkinson's Fatigue scale ≥ 3.3, RBDSQ ≥ 5, and MDS-UPDRS part 1 ≥ 2. All reported p values are two-sided, and those < 0.05 were considered to be statistically significant.SPSS (version 24) was used for all the statistical analyses.We did not correct p values for multiple testing.
This study protocol was reviewed and approved by [Medical Ethics Committee of Nara Medical University], approval number [No. 2415].The study has been granted an exemption from requiring written informed consent from Medical Ethics Committee of Nara Medical University, and we obtained the written informed consent of all patients.
Compared with the first wave, the depression score was significantly higher in the fourth wave of the pandemic (median score of GDS: 4.00 vs. 5.50, p = 0.022).Consistently, the scores of symptoms on MDS-UPDRS part 1 in the fourth wave were significantly higher in hygiene (p = 0.033), handwriting (p = 0.033), performing hobbies and other activities (p = 0.035), and turning in bed (p = 0.046) than in the first wave (Table 2).
The number of GDS scores ≥ 6 in the first and fourth waves of the pandemic were 9 and 11, respectively.Wearing off and dyskinesia were evident in 15 and 5 patients, respectively, in the first wave.Patient 9, shown in Table 1, received antidepressants in both the first and fourth waves, and the other 21 patients did not receive antidepressants.
There was no significantly different factor between PD patients who became depressed defined ≥ 6 on GDS scores (n = 3) and those who did not become depressed during the COVID-19 pandemic.Between PD patients with worsened GDS scores (n = 13) and those whose GDS score was not deteriorated (n = 9), the score of daytime sleepiness on the MDS-UPDRS part 1 Patient Questionnaire differed in the first (median 2.09 and 1.2, respectively, p = 0.052) and fourth (median 2.3 and 0.67, respectively, p = 0.016) waves.Basic demographic and clinical data between the enrolled 22 subjects and excluded 19 subjects when starting the second cohort of PHASE study showed a difference in age (68.0 ± 7.4 and 73.4 ± 6.7, respectively, p = 0.049) or the score for Mini Mental State Examination (28.8 ± 1.3 and 25.5 ± 3.8, respectively, p < 0.001), but other variables did not differ (Table 3).The UPDRS part 3 score at the beginning of the second cohort of the PHASE study was significantly lower in 3 patients who developed depression (mean, IQR; 8 (13,6)) than in 19 patients who did not develop (mean, IQR; 23 (49, 5)) (p = 0.039), but the difference between patients with worsened GDS scores (n = 13; mean, IQR: 20 (47, 6)) and those whose GDS score had not worsened (n = 9; mean, IQR: 23 (49, 5)) was not statistically significant (p = 0.385).In terms of the UPDRS part 4 score, there was no difference between patients who developed depression and those who did not (p = 0.801) or between patients whose GDS scores worsened and those whose scores did not decrease (p = 0.232).

Discussion
The present study suggested that the number of depressive symptoms determined by the same depression scale tended to be higher in the fourth wave compared with the first wave of the pandemic in Japan.There are only limited data available from previous longitudinal studies.One study showed that depressed mood worsened in 31.6 % of patients during the COVID-19 pandemic based on the comparison of different questionnaires a month before the pandemic started [2].The event scale assessed by a psychological instrument during the lockdown period was associated with a higher Beck Depression Inventory (BDI) score [4].Prelockdown BDI scores were significantly correlated with depression, anxiety, and stress Scale-21 (DASS) depression during the lockdown period [5].These studies obtained psychological symptoms only once before or after the lockdown.Only one study evaluated psychological distress from two time points during the COVID-19 pandemic [3].Among patients with advanced PD, the prevalence of depression during the lockdown (39 %) decreased to 34.1 % at the end of the lockdown.However, the reduction was not significant [3].
Our findings are consistent with the results of the previous studies that psychological symptoms had worsen after the COVID-19 waves.Patients with PD may have become anxious and, in some cases, prone to depression during the COVID-19 pandemic because of inaccessibility or interruptions to their family hospital, fear of SARS-CoV-2 infection, PD complications related to COVID-19, and interruption of nonpharmacological treatments [1][2][3].Thirty-five percent of patients worried about the interruption of physiotherapy, psychological support, or cognitive stimulation, and 48 % felt the possible worsening of PD symptoms after the limitation of outdoor physical activity [3].This aspect was reported by a cross-sectional study using data obtained via telephone interviews about the psychological symptoms before and after COVID-19 waves [17].Depression reduces the frequency of going out and leisure activities, and diminishes motivation to improve daily life.This impairs motor abilities and accompanying non-motor symptoms [18].Our study also showed that the number of hobbies and other activities was lower at the end of the fourth COVID-19 wave.The psychological impact during the COVID-19 pandemic was significantly associated with treatment-related motor complications [4].Hygiene, handwriting, and turning in bed deteriorated parallel to the decrease in the BDI score.The Hoehn-Yahr stage increased slightly.Depression is a risk factor for PD progression [19].Interaction between motor and non-motor symptoms can create a vicious cycle in patients with PD and may affect their quality of life.Since the COVID-19 pandemic is becoming protracted due to the appearance of mutant strains, it is important to know the change in the severity of depression during the phases of the COVID-19 pandemic.
In the non-PD population, GDS scores of 5-9 were considered mild depression.However, a validation study of the depression screening questions recommended by the UK National Institute for Health and Clinical Excellence (NICE) in PD patients attending a hospital outpatient PD clinic demonstrated that the optimal cut-off of the GDS-15 for the identification of major depression on DSM-IV diagnostic criteria was 5/ 6, which gave a sensitivity of 86% and specificity of 86%, while the specificity of the cut-off 8/9 was 95% [20].In addition, if the GDS score exceeds the cutoff value of 5, it corresponds to major depression in the DSM-IV; thus, it seems difficult to consider that the depression is mild, even with a rise of only two points.Longitudinal evolution of non-motor  symptoms using Parkinson's Progression Markers Initiative data showed that young-onset PD compared to middle-and late-onset PD had more progression in depression by a significant increases of 0.32 points on the GDS-15 [21].
We did not measure MDS-UPDRS Part 3. One reason for this was that patients were unable to visit hospitals because of restrictions on going out or fear of COVID-19 infection during the COVID-19 pandemic.Song et al. showed worsening of both motor and non-motor PD symptoms after a follow-up of 4.6 months during the COVID-19 pandemic, especially in slowness, tremors, feeling stress, fatigue, and pain [22].Similar to Japan, the Korean government placed social distancing, staying at home, and refraining from going out as much as possible.However, the UPDRS Part 3 score did not show a significant difference during the COVID-19 pandemic.Moreover, both the GDS score and the Hoehn-Yahr stage worsened in the present study.Further longitudinal studies are needed to determine whether UPDRS changes can lead to worsening of depression.
The present study had several limitations.Sample size is small, and all participants were from a single hospital in a local place, which may limit the generalizability of our findings.We did not evaluate the symptoms of anxiety, which have been reported to increase during the COVID-19 pandemic.An increase in anxiety may be a mediating factor in depressive symptoms.The GDS-15 distinguish patients with PD with and without depression as a screening instrument [16].One validation study on PD suggests that the GDS and its subscales appear to primarily measure anxiety, apathy, and fatigue, which may be predominant in PDdepression [23].Second, depressive symptoms could develop in the normal population during the COVID-19 pandemic; however, the normal control group was not enrolled in the present study.Patients with PD experience more depression during the COVID-19 pandemic than the normal population [24].Finally, we did not adjust the p-values for the following reasons.First, this research was not confirmative but exploratory [25].Second, precisely adjusting the p-value is difficult because the outcomes are related [26].However, we should pay attention to the type 1 error due to repeated testing, and further study based on adequate statistical power with a large sample size is needed.
In conclusion, our observation over a year between the early and later phases of the COVID-19 pandemic showed an increase in the severity of depression in patients with PD.This was consistent with the previous knowledge that depressive symptoms among patients with PD developed within two months before and after the lockdown [5].

Ethical approval and consent to participation
This study protocol was reviewed and approved by [Medical Ethics Committee of Nara Medical University], approval number [No. 2415].The study has been granted an exemption from requiring written informed consent from Medical Ethics Committee of Nara Medical University, and we obtained the written informed consent of all patients.The appropriate participants' informed consent in compliance with the Helsinki Declaration was obtained.No investigations or interventions were performed outside routine clinical care.As this is without experimental intervention into routine care.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.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.

Availability of data and material
All data generated or analyzed during this study are included in this article.Further enquiries can be directed to the corresponding author.

Fig. 1 .
Fig.1.The cumulative number of patients has increased through the four waves of the pandemic in Nara, Japan.

Table 1
Detailed clinical independent variables of 22 patients with Parkinson's disease.

Table 2
Difference of clinical variable between the first and fourth COVID-19 wave.

Table 3
Basic clinical characteristics between the enrolled subjects and excluded subjects when starting the second cohort of PHASE study (Parkinson's disease and the relationships with circadian biological rhythms and sleep).
H.Kataoka et al.