Parkinson's disease (PD) is a progressive neurodegenerative disorder with motor symptoms such as tremors,¹⁾ rigidity, bradykinesia,²⁾ and postural instability,³⁾ as well as non-motor symptoms, including cognitive impairment, mood disorders, and sleep disturbances.⁴⁾ The prevalence of PD is rising rapidly, with projections estimating that 1.2 million people in the United States will be diagnosed with PD by 2030.⁵⁾ Among the non-motor symptoms, depression is particularly prevalent, with studies indicating that 17% of individuals with PD experience major depression, 22% have minor depression, and 13% have dysthymia. Furthermore, depressive symptoms are present in 35% of individuals with PD, and 50% of those with depression also experience comorbid anxiety.⁶⁾

Depression in PD is linked to disease severity⁷⁾ and is associated with reduced physical activity (PA), impaired mobility, worsened non-motor symptoms, sleep disturbances, fatigue, and lower quality of life.^8-11) In older adults with PD, PA levels tend to decline compared to age-matched healthy individuals,¹²⁾ potentially exacerbating motor and non-motor symptoms and further impacting quality of life.¹³⁾ Previous studies have indicated significant correlations between depression severity, PA, PD symptoms, and quality of life, suggesting that lower PA levels may contribute to increased motor and non-motor impairments, as well as reduced quality of life.^14-16)

Despite these findings, existing studies often overlook the varying levels of depression severity and their selective effects on PD symptoms.¹⁷⁾ Understanding how different levels of depression severity influence PD motor and non-motor symptoms is crucial for developing targeted interventions. By categorizing participants into four groups according to the existing literature¹⁸⁾—normal, mild, moderate, and severe depression—this study aims to provide a more nuanced analysis of the relationships between depression severity, PA, and PD symptoms compared to the traditional binary model of depressed versus non-depressed. This stratification enables a more detailed analysis of the distinct relationships between depression severity, PA, and PD symptoms, compared to the traditional binary model of depressed versus non-depressed.¹⁹⁾ By examining the impact of varying depression severity on PA levels, PD motor and non-motor symptoms, and quality of life, this study aims to identify symptom profiles and provide insights into tailored interventions for each group.

By employing a cross-sectional study design with a large sample size from Fox Insight (FI) dataset and utilizing validated tools such as the Geriatric Depression Scale Short Form (GDS-S), the Physical Activity Scale for the Elderly (PASE), the Unified Parkinson’s Disease Rating Scale Part 2 (UPDRS-2), Parkinson’s Disease Questionnaire-39 (PDQ-39), and the Non-Motor Symptoms Questionnaire (NMSQ), this research seeks to elucidate the complex interactions between PA, depression severity, PD motor and non-motor symptoms, and PD-related quality of life. Identifying these associations may provide insights into tailored interventions that promote PA as a non-pharmacological strategy to manage depression, improve PD symptoms, and enhance quality of life.

This research employed data from the FI database, an open-access resource for qualified researchers engaged in the study of PD. The FI database is accessible to individuals aged 18 years or older who were able to provide informed consent independently. The data collected through FI encompasses a range of information, including participants’ demographic characteristics, symptoms associated with PD, levels of depression, engagement in PA, performance of daily living activities, and genetic data. As of the most recent data release, the FI database has received data from over 55,000 participants. The study protocol has been reviewed and approved by the New England Institutional Review Board (IRB No. 120160179, Legacy IRB No. 14-236; Sponsor Protocol No. 1; Study Title of Fox Insight). Prior to submitting their data to the FI database, all participants provided informed consent.

Table 1 presents demographic information for participants in four depression severity groups: normal (n=1,767), mild (n=837), moderate (n=424), and severe (n=417). There were significant differences between groups in age, years with PD, and height. Age varied significantly between groups (χ²=24.02, p<0.001), with participants in the normal group being older than those in the moderate and severe groups (pairwise comparison: normal > moderate and severe). Years with PD also showed a significant difference (χ²=18.20, p<0.001), with those in the normal group having fewer years with PD than those in the mild and moderate groups (pairwise comparison: normal < mild and moderate). Differences in height reached statistical significance (χ²=10.01, p=0.018); however, after pairwise comparisons, these differences were no longer significant. Weight (χ²=1.487, p=0.685) and body mass index (χ²=5.905, p=0.116) did not differ significantly between depression severity groups.

The purpose of this study was to examine the associations between depression severity, PA levels, PD motor and non-motor symptoms, and quality of life in older adults with PD using the FI dataset. The findings revealed significant relationships between these variables, highlighting the importance of managing depressive symptoms and the potential role of PA in mitigating PD-related motor and non-motor symptoms, as well as improving quality of life. Additionally, this study explored the impact of depression severity on PA levels, PD symptoms, and quality of life, demonstrating that higher levels of depression are associated with lower PA levels, severe PD motor and non-motor symptom severity, and reduced quality of life.

Our findings demonstrated a clear gradient effect of depression severity on both motor and non-motor symptoms in older adults with PD. Participants with more severe depression had significantly higher UPDRS-2 scores (indicating severe motor symptom severity) and NMSQ scores (indicating severe non-motor symptom severity), reflecting a higher overall disease burden. Strong positive correlations were observed between depression severity (GDS-S) and both UPDRS-2 (r=0.506, p<0.01) and NMSQ (r=0.598, p<0.01), suggesting that depression is a key factor influencing PD symptom severity. Regression analysis further supported these associations, with non-motor symptoms (NMSQ) emerging as the strongest predictor of depression severity (β=0.449, p<0.001), followed by motor symptoms (UPDRS-2; β=0.212, p<0.001). These results align with previous research indicating that depression worsens both motor and non-motor PD symptoms. More specifically, PD patients with depression exhibit significantly worse motor function, as measured by UPDRS Motor III,²⁶⁾ and have lower Hoehn and Yahr stage classifications indicating greater disease severity.²⁷⁾ Additionally, depression has been identified as a major risk factor for worsening non-motor PD symptoms.²⁸⁾ The neurophysiological mechanisms underlying these associations may be linked to progressive neurodegeneration and functional brain changes in PD. According to the neuroimaging research, individuals with severe depression in PD exhibit reduced dopamine transporter activity, decreased frontal blood flow, impaired caudate-frontal glucose metabolism,^29-31) and white matter loss in cortical-limbic circuits.³²⁾ Since dopamine deficiency is a primary driver of both motor and non-motor symptoms in PD,³³⁾ dopamine dysfunction also plays a central role in PD-motor and non-motor symptoms, and depression.

A key finding of this study was the significant negative correlation between PA levels (PASE) and depression severity (r=-0.335, p<0.01), motor symptom severity (r=-0.340, p<0.01), and non-motor symptoms (r=-0.212, p<0.01). These results indicate that individuals with higher PA levels exhibit lower depression severity and less severe PD-related symptoms. Further supporting these findings, the regression model identified PASE as a significant negative predictor of depression severity (β=-0.171, p<0.001), indicating that higher PA is associated with lower depression scores. Among all predictors in the model, non-motor symptoms (NMSQ) had the strongest positive association with depression severity (β=0.449), followed by motor symptoms (UPDRS-2; β=0.212), highlighting the particularly strong influence of non-motor burden on depressive outcomes. Years living with PD had a small but significant negative association (β=-0.057), suggesting that symptom burden and behavioral factors may play a more prominent role than disease duration alone. This aligns with previous research suggesting that PA has positive impact on PD-motor symptoms³⁴⁾ and non-motor symptoms by enhances neuroplasticity,^35-37) promotes release of dopamine, brain-derived neurotrophic factors and neurotransmitter, all of which may contribute to mitigating both motor and non-motor PD symptoms.^38,39) Given that PA is a modifiable lifestyle factor, these findings emphasize the importance of exercise-based interventions in PD management. Regular PA is associated with lower depression severity and improve PD motor and non-motor function. This suggests its potential as a holistic management approach to improving overall quality of life.

The stratification of participants by depression severity revealed pronounced differences in perceived functional mobility, as assessed by the EQ-5D mobility item. In the normal depression group, 35.3% of participants reported having "no problems" walking, whereas this percentage declined sharply to 18.3% in the mild group, 14.4% in the moderate group, and just 5.5% in the severe group. Conversely, the percentage of individuals reporting "moderate," "severe," or "unable to walk" progressively increased with greater depression severity. Specifically, reports of "moderate problems" rose from 18.4% in the normal group to 36.2% in the severe group; "severe problems" rose from 4.5% to 25.2%; and "unable to walk" increased from 0.3% to 5.0%. These findings indicate that depression not only exacerbates PD symptoms but also contributes to functional limitations, further reducing independence and increasing fall risk. It also emphasizes the importance of early detection and treatment of depressive symptoms as a potential means to preserve functional mobility in individuals with PD. Additionally, depression severity was strongly associated with sleep disturbances, with 50.4% of participants in the severe depression group reporting sleep disorders, compared to 37.2% in the normal group. This is consistent with previous studies indicating that sleep disturbances, particularly REM sleep behavior disorder and insomnia, are common in PD^34,40) and are closely linked to depressive symptoms. Given that poor sleep quality can further exacerbate both motor and non-motor symptoms, addressing sleep disturbances should be an integral component of PD treatment strategies.

Despite the strengths of this study, including a large sample size and the use of validated assessment tools, several limitations should be acknowledged. First, the cross-sectional design prevents the determination of causal relationships between depression severity, PA levels, and PD symptoms. Future research should adopt a longitudinal approach to track changes in depression, PA, and PD motor and non-motor symptoms over time, allowing for a better understanding of the directionality of these associations. Another limitation is the reliance on self-reported PA using the PASE questionnaire, which may introduce recall or perception bias. Although the PASE has demonstrated validity and reliability in PD populations, self-reported data are inherently vulnerable to bias. Future studies should incorporate objective measures of PA, such as wearable devices (e.g., accelerometers or pedometers) to monitor step counts and activity duration. Additionally, clinical assessments of mobility and motor function could benefit from the use of inertial measurement unit (IMU) sensors, which provide detailed, quantitative analysis of gait and movement patterns. These objective tools can improve measurement precision and reduce bias associated with self-report instruments. Finally, our regression model did not control for several important confounding variables, such as dopaminergic medication use (e.g., levodopa equivalent dosage), cognitive function, and socioeconomic status. These factors can substantially influence both depression severity and PD symptoms, and their omission may affect the precision of the reported associations. Future research should include these variables as covariates in regression analyses to provide a more accurate understanding of the independent contributions of PA and PD-related symptoms to depression severity.

In conclusion, this study provides strong evidence that depression severity is closely associated with PD motor and non-motor symptoms, functional mobility, and sleep disturbances. Furthermore, higher PA levels are linked to lower depression severity and improved PD-related outcomes, underscoring the potential of exercise as a non-pharmacological intervention. These findings highlight the need for comprehensive, multidisciplinary approaches to PD management that address both physical and mental health components. Future research should focus on longitudinal studies to further elucidate causal relationships and develop targeted interventions to optimize quality of life for individuals with PD.