Quality of life in Parkinson's disease: A systematic review and meta‐analysis of comparative studies

Abstract Background Studies regarding the impact of Parkinson's disease (PD) on quality of life (QOL) have reported conflicting results, and the underlying QOL domains require further study. In order to understand the association between PD and QOL, we conducted this meta‐analysis to systematically compare QOL between PD patients and healthy controls. Method The PubMed, PsycINFO, EMBASE, and Web of Science databases were systematically searched. Data were analyzed using the random‐effects model. Results Twenty studies covering 2707 PD patients and 150,661 healthy controls were included in the study. Compared with healthy controls, PD patients had significantly poorer QOL overall and in most domains with moderate to large effects sizes. Different QOL measures varied in their association with quality of life, with the Parkinson's Disease Questionnaire‐39 (PDQ‐39) having the largest effect size (standard mean difference, SMD = −1.384, 95% CI: −1.607, −1.162, Z = 12.189, P < 0.001), followed by the Europe Quality of Life Questionnaire‐visual analogue scale (EQ‐VAS) (SMD = −1.081, 95% CI: −1.578, −0.584, Z = −4.265, P < 0.001), Europe Quality of Life Questionnaire‐5D (EQ‐5D) (SMD = −0.889, 95% CI: −1.181, −0.596, Z = −5.962, P < 0.001), and the Short‐form Health Survey (SF) scales (physical dimension: SMD = −0.826, 95% CI: −1.529, −0.123, Z = −2.303, P = 0.021; mental dimension: SMD = −0.376, 95% CI: −0.732, −0.019, Z = −2.064, P = 0.039). Conclusion PD patients had lower QOL compared with healthy controls in most domains, especially in physical function and mental health. Considering the negative impact of poor QOL on daily life and functional outcomes, effective measures should be developed to improve QOL in this population.


| INTRODUC TI ON
Parkinson's disease (PD) is a neurodegenerative disease having an overall prevalence ranging from 1 to 2 per 1000 people. 1,2 PD is a chronic, progressive, age-related disorder, which is rare in young people, but whose prevalence reaches up to 4% in older adults. 2 PD is characterized by various motor dysfunctions, such as bradykinesia, rigidity, gait freezing, resting tremor, and postural reflex impairment, 3  to their goals, expectations, standards, and concerns." 5 QOL encompasses physical, psychological, autonomy, cognitive, social relations, and environmental factors. 5,6 To improve the QOL of PD patients, it is important to understand how various QOL domains differ in PD patients and healthy controls. Some comparative studies on QOL in PD patients have been conducted, but the findings are mixed, especially the extent of differences between PD patients and controls in different domains. For instance, compared with healthy controls, some studies found that PD patients had an overall lower QOL, 7-12 while other studies did not find significant differences in QOL domains of physical health, 8,13 mental health, 9 emotional function, 10 environment, 11 and social relations. 12 Major correlates of QOL in PD include comorbid depressive symptoms, and PD severity and subtypes. 14 Gait impairments, adverse effects of medications, and psychosocial dysfunction are contributing factors to poor QOL. 15 To the best of our knowledge, no systematic review or meta-analysis has compared QOL between PD patients and healthy controls that also drilled into various domains. The main objectives in this systematic review and meta-analysis were as follows: (a) to compare the overall and domain QOL between PD patients and healthy controls and (b) to quantify QOL differences between groups, with different standardized instruments, using the effect size statistic. We hypothesized that PD patients would have significantly lower QOL compared with healthy controls.

| Inclusion and exclusion criteria
The search for relevant articles was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart, 16  screened the titles and abstracts of relevant literature and then read the full text to further assess eligibility. Any disagreement was discussed by the two above researchers, and if a consensus could not be reached, guidance was sought from a senior researcher (YTX).

| Data extraction and quality assessment
Participant and study information, such as first author, publication year, sampling method, QOL measures, number of PD patients and controls, illness duration, and QOL scores, was extracted. For studies reporting QOL by a patient subgroup (eg, by gender), overall QOL was calculated by combining the QOL subgroup scores using a formula. 20 Study quality was independently assessed by the same two researchers (NZ and YY) using the Newcastle-Ottawa Scale (NOS) in three domains: selection, comparability and exposure. 21,22 The NOS total score was calculated by summing up all item scores.

| Statistical analysis
Data were analyzed with the Comprehensive Meta-analysis software, version 2.0 (CMA; https://www.meta-analy sis.com/). Data were combined across studies using the same QOL measure, which varied from one study to another. Physical and mental/psychological domains were measured separately with the WHOQOL and SF scales; thus, domain scores were pooled for each scale. For studies without SDs for QOL data, the SDs of other studies were averaged as previously done. 23 Standardized mean differences (SMDs) in Q OL between PD patients and healthy controls were calculated to estimate effect size. As a guide, SMDs of 0.2, 0.5, and 0.8 were considered small, moderate, and large effect sizes, respectively. 24 Taking into account differences in sampling methods, study characteristics, and assessment tools, random-effects models were used to synthesize data. 25 Heterogeneity was assessed with Q and I square statistics.
An I 2 value of 50 percent or more 20 indicated significant heterogeneity in which case possible sources of heterogeneity between subgroups were explored based on: (a) QOL measures (WHOQOL vs. SF scales vs. PDQ-39 vs. Europe Quality of Life Questionnaire-5D (EQ-5D) vs. Europe Quality of Life Questionnaire-visual analogue scale (EQ-VAS)) and (b) QOL domains (physical health vs. mental/psychological health). Each subgroup was required to consist of at least 3 studies. If there were 10 or more studies, funnel plots were created and Egger's Rank test was conducted to assess possible publication bias. 26 The significance level for meta-analytic outcomes was set at 0.05 with two-tailed tests.

| Study characteristics and quality assessment
Key characteristics of included studies are summarized in Table 1.
They were published between 1995 and 2020, and the sample size ranged from 33 to 144,692. The details of study quality assessment are presented in Table S1.

| QOL measurements
QOL measures involved in this systematic review are shown in Table 1.

| QOL comparisons by scale
Three studies employing the WHOQOL 12 Figure S4) were meta-analyzed separately. Compared with controls, PD patients had significantly poorer overall QOL in these analyses.

| Subgroup analyses and publication bias
No significant difference was found between the WHOQOL and SF assessments regarding physical and mental QOL ( Table 2). There was a significant difference between QOL measures in effect sizes (Table 2); the PDQ-39 was associated with the largest effect size, followed by the EQ-VAS, EQ-5D and SF scales ( Table 2). Since the minimum number of studies per measure was not met, publication bias analysis could not be undertaken.

| D ISCUSS I ON
To the best of our knowledge, this was the first systemic review and meta-analysis that compared QOL between PD and healthy controls with standardized measures and estimating group differences. PD patients had significantly poorer QOL than healthy controls overall and in most domains. ing factor for QOL in some, 11,12,28 but not all studies. 55 Anxiety, apathy, and pain are also associated with poor QOL in PD, 11,48 with greater effect sizes than motor-symptoms. 48 However, the significant relationship between anxiety and poor QOL was not found in another study. 12 Similarly, the association between sleep disturbances and QOL is contested, with some studies finding a significant relationship between sleep problems and QOL, 7,42,58 but not others. 12,55 In addition, some studies found that REM sleep behavior disorder with reduced striatal dopamine transporter values and increased expression of PD-related pattern may be associated with the occurrence of PD. [65][66][67] The discrepancy between studies could be partly due to differences in instruments, 68,69 sampling methods, disease severity, 14

ACK N OWLED G M ENTS
The study was supported by The National Science and Technology

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created in this study.