Introduction
Assessment of patients’ quality of life has become an important outcome indicator in Parkinson’s disease (PD) [
1]. PD patients show an impaired quality of life [
2], not only due to the motor symptoms, but also several non-motor symptoms such as sensory abnormalities, sleep disorders, autonomic disturbances, cognitive impairment, and some neuropsychiatric aspects, like anxiety, depression, apathy, and fatigue [
3,
4].
Quality of life includes both an individual's perception of themselves in the cultural context and the value systems, and the relationship to their goals, expectations, norms, and concerns [
5]. Specifically, health-related quality of life (HRQoL) [
2] is defined as the patient's life satisfaction, perception, and self-evaluation of the effects of a disease and their experience with it, being the physical and psychological dimensions, functional capacity, and social interactions the most relevant aspects [
2,
6]. The way to assess HRQoL differs among studies and although there are several scales to measure this concept, Parkinson’s Disease Questionnaire-39 (PDQ-39) is the most commonly used instrument for HRQoL assessment in PD [
2]. Most of the studies only analyze the overall HRQoL, while PDQ-39 was designed to assess different aspects of the functioning and well-being of people affected by PD [
7]. This questionnaire is divided into eight dimensions: mobility, activities of daily living (ADL), emotional well-being, stigma, social support, cognition, communication, and bodily discomfort [
7,
8]. Each of these dimensions offers the possibility of analyzing specific aspects that influence the HRQoL of PD patients [
7,
8].
On the other hand, cognitive dysfunction is a non-motor symptom of PD that has an impact on the quality of life of PD patients [
1]. Cognitive impairment is observed in 40% of people with PD during the course of the disease [
9]. The most common deficits in PD involve visuospatial abilities, verbal memory, and frontal executive domains [
10]. The risk of progression of cognitive impairment to dementia has a functional impact on the ADL and HRQoL of people with PD [
6,
11,
12]. The high association between cognitive impairment and difficulties in ADL persists with disease progression [
12] and 83% of PD patients with cognitive impairment develop dementia after 10–20 years [
11].
Anxiety, depression, apathy, and fatigue are other characteristic non-motor symptoms that can affect the quality of life of PD patients [
13‐
16]. Neuropsychiatric problems are present in more than 77% of people with PD [
17]. Specifically, meta-analyses showed that the average prevalence rate of anxiety disorders in PD patients was 31% [
18], depressive symptoms were found in 22.9% [
19], apathy in 40% [
16], and fatigue in 50% [
14]. Although research generally focuses on depressive symptoms, anxiety is a very common symptom in PD [
18] and both have been reported to affect the quality of life of people with PD [
15,
20,
21].
A systematic review showed that HRQoL of PD patients has been increasingly explored through diverse studies investigating the predictors of the overall HRQoL in PD [
22]. It is essential to identify the most important predictors of HRQoL in PD patients in order to set treatment priorities and reduce the effects of the functional and emotional consequences of PD [
22]. However, it should be considered that measures of depression and quality of life are concepts with overlapping items [
23‐
26], because emotional distress is a construct that is present in most quality of life questionnaires [
23,
24], including specific items assessing psychopathology [
24,
25], such as depressive [
23,
25] and anxiety symptoms. As suggested by Hays and Fayers [
23], most of the articles published to date in different pathologies have included measures of depression and HRQoL without considering the empirical and conceptual overlap between the two variables [
23]. Therefore, not taking into account the overlap between these measures may result in tautological inferences about the influence of depression on HRQoL, since, if a large proportion of the items overlap, it is logical that quality of life would be significantly associated with measures of psychopathology [
23,
24]. Recognition of the overlap between these variables is crucial to interpret associations and avoid incorrect causal inference [
23], so it is recommended to control for the presence of psychopathological items in quality of life measures [
24].
Conversely, predictors of HRQoL dimensions have not been studied as much as predictors of overall HRQoL [
13,
17,
27‐
31]. Moreover, cognition has not been studied as often as the clinical and motor variables of the disease. Despite some studies having analyzed cognition as a predictor of overall HRQoL [
6,
21,
28‐
30,
32‐
35], few of them have examined cognition as a predictor of the specific dimensions of HRQoL [
28‐
30]. Most of these studies have analyzed cognition with cognitive functioning screening tests, however, further research is needed to analyze the impact of cognition on overall HRQoL and its dimensions using a comprehensive neuropsychological assessment.
Therefore, the main goal of this study was to simultaneously analyze the predictive value of a wide spectrum of variables such as anxiety, depression, apathy, fatigue, cognition, and motor symptoms that impact HRQoL and its dimensions. The second aim of this study was to observe the predictive value of these variables after identifying and controlling the overlap between HRQoL items and clinical measures.
Discussion
The aim of this study was to investigate the predictors of HRQoL and its dimensions in PD patients. Specifically, this study is a multidimensional predictive model that analyzes the predictor value of motor and non-motor symptoms, including scores of anxiety, depression and fatigue and neurocognitive functions in HRQoL and its dimensions in PD. Moreover, this study analyzed the predictive value of these variables after considering the overlapping items between HRQoL and clinical measures.
This study has identified several characteristics associated with a worse HRQoL in PD patients. Higher scores for anxiety, fatigue, motor symptoms, and depression were the factors that predicted a worse HRQoL, with anxiety being the main predictor. The variance in the predictive value of depression scores was lower than in the other studies, with anxiety, fatigue, and motor symptoms predominating over depression scores. Most studies showed that depression was the main predictor of HRQoL [
22], and it may be due to the absence of anxiety and depression variables joined in the same regression model [
21]. Additionally, the high association between depressive symptoms and quality of life may be due to the overlap between the emotional items of the quality of life measures and depressive symptomatology [
25], so these results should be interpreted with caution. Therefore, in our study, an additional analysis was performed in which the emotional well-being dimension items of the HRQoL total index were removed to consider this overlap. Interestingly, our results showed that after taking into account the overlap, fatigue predominated over anxiety and motor symptoms; but depression was no longer a predictor of HRQoL total index. Most studies have not considered the overlap between clinical and HRQoL measures in their analyses and it may be the main reason why depression has been considered as the main predictor of HRQoL. Future studies should contemplate the overlap between clinical and HRQoL measures to perform more accurate interpretations. Controlling the overlapping, fatigue was the main predictor of HRQoL total index. Fatigue is recognized as one of the most disabling symptoms of PD [
61], present in 50% of PD patients [
14], however, it has not been such a studied predictor as anxiety and depression. In four of the five studies reviewed in a systematic review, fatigue was a strong predictor of HRQoL [
22], demonstrating that it emerges in the early stages of PD and persists throughout the course of the disease, negatively impacting their HRQoL [
14]. Additionally, in our study, neurocognition emerged as a predictor of HRQoL total index when controlling the overlap. The majority of the studies analyzed cognition with cognitive functioning screening tests [
21,
28,
29,
32‐
35]. In contrast, in our study, participants performed an extensive neuropsychological battery comprised of a wide variety of cognitive tests, revealing that neurocognition composite score predicted HRQoL. To the authors´ knowledge, only four studies found cognition as a predictor of HRQoL total index [
6,
29,
30,
33]. Two of these studies analyzed cognition with cognitive functioning screening tests [
29,
33], while the two other studies assessed cognition by specific cognitive measures, showing that working memory [
30], verbal fluency [
30], visual attention/memory [
6], visuospatial [
6], and executive functioning [
6] were predictors of HRQoL. As far as the authors are aware, this is the first study investigating predictors of HRQoL in PD patients in which the possible overlap between HRQoL and clinical outcomes was analyzed. Furthermore, it should be emphasized that the methodology used and the results obtained in this study could be extrapolated to other pathologies.
Regarding the specific HRQoL dimensions, mobility and ADL dimensions were predicted by fatigue and UPDRS-III. Similar to our results, other studies found that reduced activity and motor symptoms were associated with mobility [
27] and ADL dimension [
27,
28,
31], while physical fatigue was also a predictor of mobility dimension [
62]. Consistent with previous studies [
27], our findings showed that fatigue was also a predictor of emotional well-being dimension. Other research suggest that longer disease duration [
27], UPDRS total score [
27], working memory and verbal fluency [
30] were also a predictors of this dimension. Finally, fatigue along with anxiety scores, were predictors of the bodily discomfort dimension. Other studies found that fatigue, UPDRS and female sex [
27], anxiety [
13,
17,
28], depression and hallucinations [
17], and motor fluctuations [
28] were predictors of this dimension. However, although in our study depression scores were included in the bodily discomfort model, anxiety and fatigue scores showed a higher association with this dimension, predominating over depression scores. In line with other studies [
17,
28], social support dimension was predicted by anxiety scores and UPDRS-III. Additionally, anxiety scores also predicted the cognition dimension. PDQ-39 items of the cognition dimension assess concentration problems, the sensation of having a bad memory and hallucinations or nightmares, which generate feelings of anxiety or nervousness in PD patients. Even though we expected to find neurocognition as a predictor of this dimension, it did not happen. This may be due to the fact that the sensations of concentration problems and memory impairment approach a similar interpretation of cognition; while hallucinations or nightmares are more related to anxiety. In fact, one study suggests that the dimension of cognition in PDQ-39 has a stronger relationship with mood states rather than with neurocognitive domains [
63]. Interestingly, neurocognition was found as a predictor of communication dimension, in addition to motor symptoms. The range of communication impairment is very variable in PD (from a lack of problems to inaudible and unintelligible speech) [
64]. Indeed, acoustic speech deficits, use of action verbs and pausing have been shown to be more associated with motor impairment and linguistic deficits with cognitive impairment [
64].
Concerning stigma, no significant associations were found, so they could not be included in the model. In contrast, other studies showed that anxiety [
13,
28], reduced activity and motivation [
27], and depression [
31] predicted the stigma dimension in PD. Low scores on the stigma dimension may explain the lack of association with the rest of the variables. Furthermore, as Tu and colleagues noted [
31], understanding stigma requires a broader consideration from a more social context, and not only focusing on motor and non-motor symptoms. Regarding the rest of the variables, our results did not show a significant association between apathy scores and HRQoL or any of its dimensions. However, other studies revealed an association of apathy with poorer HRQoL [
17,
60] and with an increased self-rating of cognition and communication difficulties [
17].
Recognition and treatment of motor and non-motor symptoms in the early stages of the disease may improve the HRQoL of PD patients. Considering that PD is a physically and psychologically debilitating disorder, health care professionals should adopt a holistic approach to PD rehabilitation [
65]. Treatment objectives vary among individuals, highlighting the need for personalized intervention; so, a variety of non-pharmacological therapies should also be taken into account [
66], such as, mindfulness yoga [
65], psychological interventions [
67], spiritual resilience [
68], physical therapy [
69], cognitive training [
70], and speech therapies [
71]. These therapies could be included in the clinical manage of PD patients.
It should be noted that the results of our study were based on cross-sectional data, limiting the ability to observe the impact of motor and non-motor symptoms on HRQoL in the disease course. For future research, longitudinal analyses would be needed to determine if the predictive symptom intervention could affect HRQoL. In fact, a recent longitudinal study examining predictors of HRQoL impairment in PD patients showed that, after a 2-year follow-up, age, sex, mood, and non-motor impairment were associated with clinically significant HRQoL impairment in PD patients [
72]. On the other hand, the HRQoL evaluated using PDQ-39 does not allow to contrast the results with other pathologies, since the PDQ-39 is a specific PD scale. In this study, PD patients did not have a long disease duration and were mildly to moderately affected, without major impairments in their HRQoL, so it could be interesting to analyze the predictors of HRQoL and its dimensions in more advanced stages of the disease. Additionally, the low sample size limits generalization, so it would be interesting to conduct future studies with larger sample sizes. In fact, our study was carried out in a Spanish population and there are studies using PDQ-8, which is an abbreviated scale of PDQ-39, suggesting that geographic location has an effect on the non-motor symptoms affecting HRQoL [
73]. Specifically, mood and sleep were the major predictors in European patients, whereas, non-motor symptoms were not significant predictors of HRQoL in Indian and Japanese patients [
73] and in Chinese population anxiety, depression, motor symptoms, and marital status were the main predictors of HRQoL [
74].
In conclusion, our results showed that anxiety, fatigue, motor symptoms, and depression were the main predictors of HRQoL total index in PD patients, whereas, after removing emotional well-being overlapping items, fatigue, anxiety, motor symptoms, and neurocognition were the predictors of HRQoL total index. These findings indicate the importance of identifying and controlling the overlap of items between HRQoL measures and clinical variables in order to perform an accurate interpretation of the results. Our results and methodology would be extrapolated to other pathologies. Additionally, results showed the impact of fatigue, anxiety, motor symptoms, and neurocognition scores in HRQoL dimensions in PD patients. Consequently, results suggest the importance of the appropriate assessment of HRQoL and its specific dimensions, because predictors are different for each dimension of HRQoL. Recognition and management of motor and non-motor symptoms in the early stages of the disease are essential, as these features have a greater impact on the HRQoL of people with PD. Therefore, intervention on fatigue, anxiety, motor symptoms, and cognitive processes may be a crucial target to improving HRQoL in PD patients.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.