Patient With Parkinson Disease and Care Partner Perceptions of Key Domains Affecting Health-Related Quality of Life

Abstract

Background and Objectives

To effectively customize Parkinson disease (PD) programs, it is important to incorporate the “individual's voice” and have a thorough understanding of the symptom priorities of people with PD (PwP) and care partners (CP). In this convergent integrated mixed-method systematic review, we aimed to analyze qualitative and quantitative evidence of PD motor and nonmotor symptoms affecting health-related quality of life (HRQOL) in PwP and CP, comparing priorities across different levels of disease severity.

Methods

We searched MEDLINE, PsycINFO, Web of Science, Embase, and Scopus; ProQuest Dissertations and Theses Global; and the Michael J. Fox Foundation Data Resources for studies published up to June 29, 2022. We included qualitative, quantitative, and mixed-method studies investigating PD symptom priorities among PwP and CP. We critically appraised eligible studies for methodological quality using the Mixed-Methods Appraisal Tool. Derived terms were mapped and coded according to thematic attribution. Independent syntheses of qualitative and quantitative evidence and transformation of quantitative data into qualitative formats were performed.

Results

Of the 7,716 identified studies, we included 70 that provided qualitative (n = 13), quantitative (n = 53), and mixed (n = 4) evidence. We included 604 mapped terms representing 11 PwP-identified and CP-identified motor and nonmotor symptom categories. Across all PD stages, both PwP and CP considered 5 domains more affecting their HRQOL, namely: “motor functionality,” “mood,” “cognition,” “gait, balance, posture, and falls,” and “nighttime sleep disorders.” In early disease, PwP and CP considered “mood” the domain that most affected their HRQOL. In advanced PD, PwP considered “pain” the domain that most affects their HRQOL, while CP considered “psychiatric symptoms.” The domain “gait, balance, posture, and falls” was equally considered by both PwP and CP as the second domain that most affects their HRQOL in the advanced stage of PD.

Discussion

The ranking of the priority of symptoms is largely shared by PwP and CP, and motor symptom priorities dominate the full disease spectrum. However, the nonmotor symptom priorities shift according to the disease severity stage. Tailored care and research require that providers consider these shifting priorities and incorporate the “individual's voice” into treatment decisions.

Introduction

Parkinson disease (PD) is a neurodegenerative condition with both motor and nonmotor symptoms that cause difficulties for people with PD (PwP) and their care partners (CP).^1,2 Variations in the impairment of symptoms' function over disease progression may contribute differently to the health-related quality of life (HRQOL) of PwP and CP.^3,4

HRQOL is a multidimensional concept of an individual's physical and mental well-being. The factors that influence them include their health risks and conditions, functional abilities, level of social support, and socioeconomic status.⁵ HRQOL can change over time as an individual's circumstances change, providing a holistic understanding of their overall health status and helping guide interventions to improve their well-being.⁵

Individualized care is crucial to a high-quality PD health service.^6-9 However, additional efforts are needed to emphasize the “voice” of affected PwP and CP to operationalize it as an outcome measure in clinical care and research.¹⁰ If PwP and CP perceptions of PD symptom priorities differ from the clinician's, such discrepancies can hamper symptoms' effective management.^11,12

In various diseases, including PD, there has been an increasing attempt to understand the best available evidence on target symptoms by combining self-reported outcome qualitative and quantitative data,^13-17 gathering valuable information on the effect of a given treatment or intervention on efficacy and HRQOL. A preliminary search of PROSPERO, MEDLINE, and Joanna Briggs (JBI) Evidence Synthesis confirmed no current or in-progress systematic review. Therefore, we aimed to produce an integrated set of qualitative and quantitative evidence to understand which PD motor and nonmotor symptoms affect the HRQOL of PwP and CP and to study those priorities across PD progression by comparing early disease with advanced disease studies.

Methods

We conducted a convergent integrated mixed-methods systematic review following JBI methodology^18,19 and registered the study protocol in the Open Science Framework. We ran a 3-step strategy to answer the following question and achieve the specific aims: As PD progresses, what motor and nonmotor symptoms most affect the HRQOL of PwP and CP?

Our approach consisted of the following aims:

Aim 1: To analyze qualitative evidence to gain a subjective-based understanding of the effect of PD symptoms on the HRQOL of PwP and CP. This will help us understand how PwP and CP experience PD symptoms through constructivist and personal exploration of their effect on their HRQOL.

Aim 2: To analyze quantitative evidence to establish a comprehensive understanding, based on Clinical Assessment Outcomes (COA), of the PD symptoms that affect the HRQOL of PwP and CP. This will help us understand which PD symptoms significantly affect the HRQOL of PwP and CP through objective exploration derived from COA.

Eligibility Criteria

We searched for reports investigating self-reported PD symptoms affecting the HRQOL of PwP and CP according to the following criteria:

Inclusion Criteria

• Type of publication: peer-reviewed original studies (articles, dissertations, and theses).

• Year and language: We did not apply a year or language restrictions during the search process and selected all studies published until June 29, 2022. However, during the full-text analysis, we included only studies published in English.

• Study Design:

  • Quantitative: We included studies implementing nonrandomized and descriptive approaches, using any clinical outcome measure for data collection of PD symptoms and using patient-reported outcomes (PRO) to collect the effect of an individual's HRQOL. For example, we included studies using the International Parkinson and Movement Disorder Society–Unified Parkinson Disease Rating Scale (MDS-UPDRS) Parts I to IV to assess the severity of PD motor and nonmotor symptoms and the Parkinson Disease Questionnaire (PDQ)–39 or the Parkinson Impact Scale, both PRO, to assess the effect on HRQOL of the PwP or CP.

  • Qualitative: We included studies implementing, but not limited to, phenomenology, grounded theory, and descriptive approaches (e.g., in-depth interviews, focus groups, and hybrid thematic analysis [inductive and deductive]).

  • Mixed methods: We included studies of any approach (convergent or sequential exploratory design).

• Topic of the studies: We did not apply restrictions on the topic of the study and selected all studies investigating the HRQOL outcome of perceived motor and/or nonmotor symptoms in PwP and CP. We adopted the definition of HRQOL proposed by the Center for Disease Control and Prevention, which describes HRQOL as “an individual's or group's perceived physical and mental health over time.”⁵ In this sense, we included studies using the 9 HRQOL COA in PD recommended by the MDS.²⁰

Exclusion Criteria

We excluded study protocols, conference proceedings, editorials, and editors' letters. We searched the primary studies for identified reviews and included those that met the eligibility criteria. For dissertations and theses, we searched for published articles related to that study, prioritizing the inclusion of peer-reviewed articles. In this sense, if an article was included, then the dissertation/thesis (the article's primary source) was excluded to avoid data duplication.

Search Strategy and Study Selection

We used 3 steps to locate published and unpublished studies: 1. initial study identification, 2. comprehensive study search, and 3. complementary search of studies.

First, on June 25, 2022, we implemented an initial limited search of MEDLINE (PubMed) to identify studies on the topic. We used the text words in the titles and abstracts of relevant studies and the index terms to develop a full search strategy for MEDLINE (PubMed) (eAppendix 1, links.lww.com/WNL/D332).

Second, on June 29, 2022, we expanded the search strategy by including all identified keywords and index terms found in step 1 and adapting them for each included database: PsycINFO (APA PsycNet), Web of Science (Clarivate Analytics), Embase and Scopus (Elsevier), ProQuest Dissertations and Theses Global (ProQuest), and the Michael J. Fox Foundation (MJFF) Data Resources²¹ (eAppendix 2, links.lww.com/WNL/D332).

Following the search, we collated all identified studies, uploading them into Rayyan (Qatar Computing Research Institute, Doha, Qatar) for the initial screening of abstracts and titles and duplicate removal. Furthermore, 2 independent reviewers screened titles and abstracts using the inclusion and exclusion criteria for the review. Finally, we retrieved potentially relevant studies and imported their citation details into the reference manager Mendeley v.1.19.4 (Mendeley Ltd., Elsevier, Netherlands). The same 2 reviewers assessed the full text of selected citations in detail against the inclusion criteria, recording and reporting all the reasons for excluding full-text studies in this review. Disagreements raised between the reviewers at each step were resolved through discussion or with the third reviewer.

In the third step, conducted on August 1, 2022, we reviewed the reference lists of all selected studies to identify potential additional studies. The final systematic review process is presented in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.²²

Critical Appraisal

We critically appraised all included studies for methodological validity using the standardized Mixed-Methods Appraisal Tool, version 2018.²³ Regardless of their methodological quality, all studies underwent data extraction and synthesis.

Data Extraction, Transformation, Synthesis, and Integration

We extracted the data from the included studies to a Microsoft Excel spreadsheet for Mac (16.66.1), developed according to the JBI data extraction tools.¹⁹ We adapted the categories according to the investigated variables: study characteristics, population, COA, and phenomena of interest (Table 1).

Table 1.

Categorization of the Extracted Data

Variables	Categories	Subcategories
Study characteristics	General information	• Author, title, journal, language, year, continent, country of publicationa • Setting: Inpatient, Outpatient, Nursing home, Community (Association, Foundation, Database)
	Method	• Qualitative: ethnography, phenomenology, narrative, grounded theory, case study, qualitative description • Quantitative—Descriptive: incidence or prevalence without comparison group, survey, case series, case report; nonrandomized: nonrandomized control trials, cohort study, case-control, cross-sectional analytic study • Mixed-methods: convergent design, sequential explanatory, sequential exploratory
	Components	• Qualitative: qualitative data from qualitative and mixed-method studies • Quantitative: quantitative data from quantitative and mixed-method studies
Population	People with PD	• Sample size, mean age, disease durationb • Disease severity: Hoehn & Yahr (HY) Scale medianb • Disease stage: early (HY: 1 a 3), advanced (HY: 3 a 5), all stages (HY: 1 a 5)
	CP	• Sample size, relationship with the patient (spouse, family member, other)
COA	Overview	• Name, type (PRO, ObsRO), construct (PD-specific, non–PD-specific)
Phenomena of interestc	Perceived Symptoms23	• Motor symptoms  ○ Tremor  ○ Rigidity: Micrography, speech and voice problems, mastication*, breathlessness/dyspnea at rest or in effort*  ○ Bradykinesia: Arm swing decrease, hypomimia*  ○ Postural instability (and Gait): Falls, freezing of gait/festination, camptocormia  ○ Motor complications: Fluctuations (ON/OFF, Wearing-off), dyskinesia, dystonia (morning stiffness, blepharospasm), muscle cramps/spasms*  ○ Miscellaneous:   ■ Overall motor symptoms   ■ Motor functionality • Nonmotor symptoms  ○ Dysautonomia   ■ Urogenital: Genitourinary, sexual   ■ Gastrointestinal: Drooling, dysphagia, gastric stasis (decreased appetite*, nausea*), constipation   ■ Cardiovascular: Postural hypotension (dizziness*), supine hypertension   ■ Thermoregulation: Hyperhidrosis and skin disorders, intolerance to heat/cold (hot flushes*), dry eyes/mouth*  ○ Sleep: Behavioral disorder of REM, vivid dreams/nightmares, insomnia, excessive daytime sleep  ○ Neurobehavioral   ■ Mood: Depression, anxiety (strain/stress/distress*, panic attacks*), apathy   ■ Cognition: Minimal cognitive impairment, dementia (attention, memory, executive function, language…)   ■ Other psychiatric: Hallucination, illusions, delusions, impulse control disorder (hypersexuality, disinhibition, impulsivity*)  ○ Sensorial and others: Fatigue, weight gain/loss, sensorial complaints (hyposmia/anosmia/ageusia, pain (migraines/headaches*), paresthesia, visual problems* (intolerance to bright light), restless legs syndrome)  ○ Miscellaneous:   ■ Overall nonmotor symptoms   ■ Nonmotor functionality

Abbreviations: ObsRO = observer-reported nonclinical outcome; PD = Parkinson disease; PRO = patient-reported outcome.

Symptoms marked with (*) represent those that may be related to PD and have multifactorial causes.

^aFor multicentric studies, we considered the continent/country of the first author.

^bFor longitudinal studies, we considered the baseline data.

^cThe phenomenon of interest is the primary data obtained from qualitative and quantitative components obtained from the studies.¹⁵.

We also used Microsoft Excel to analyze the characteristics of the studies, population, and COA with statistics of mean values (±), ranges, percentages (%), and cumulative sum. We used the χ² test to determine associations between motor and nonmotor symptom categories and study types (qualitative and quantitative). We set statistical significance at threshold p < 0.05.

We analyzed the phenomena of interest by implementing the JBI convergent integrated approach, integrating the independent syntheses of the qualitative and quantitative components of the included studies following the transformation of quantitative data into qualitative data, which involves a narrative interpretation of the quantitative results (“qualitizing”)^18,19 (eAppendix 3, links.lww.com/WNL/D332).

We synthesized the qualitative findings using JBI-QARI, implementing a 2-stage meta-synthesis process for the identification of the 2 levels of findings¹⁸:

1. Level 1: We mapped and listed the qualitative data (verbatim) considered representative of the phenomena of interest and rated them according to their credibility (eAppendix 4, links.lww.com/WNL/D332).

2. Level 2: We extracted and categorized all the terms identified in the qualitative data based on their similarity in meaning, implementing a meta-synthesis to generate a single comprehensive set of categories that we could use as a basis for evidence-based practice. We coded the findings according to the thematic attribution from a classification of PD's primary motor and nonmotor symptoms described previously (Table 1).²⁴ For instance, responses that mentioned tiredness, lack of energy, fatigue, and weariness were grouped into a broader subcategory of symptom group called “Fatigue,” contained in the category “Sensory and others.” We created an additional category (Miscellaneous) to accommodate general terms that we could not allocate to specific subcategories, such as overall impairment of motor symptoms (eAppendix 5, links.lww.com/WNL/D332).

We synthesized the quantitative findings by implementing a 2-stage meta-synthesis process for the identification of the 2 levels of findings¹⁸:

1. Level 1: We mapped and listed the terms (verbatim) extracting their quantitative data following the criteria for each type of study:

   1. Quantitative descriptive design: We extracted the number and percentages of all terms, regardless of their prevalence, and flagged those cited by ≥10% of the sample.

   2. Quantitative analytical design: We extracted the linear and logistic regression data statistics (β coefficient (95% CI) and p value) and included only those with significant inferential relationships with the phenomena of interest.

2. Level 2: We categorized all the terms identified based on their similarity in meaning, implementing the same meta-synthesis mentioned earlier.

The primary author analyzed and combined the data and presented the synthesis to the other authors, who examined and further refined the results for validation and agreement.

Ultimately (level 3), we converged the findings from each single-method synthesis to generate a set of evidence that represented this aggregation. Finally, to establish the clinical relevance of each symptom to the HRQOL of PwP and CP, we calculated the percentage of each symptom according to the number of included studies citing it.

Standard Protocol Approvals, Registrations, and Patient Consents

This is a secondary data analysis study, and institutional review board approval and patient consent are waived.

Data Availability

Data supporting the findings of this study are available for 5 years after the study publication with the corresponding author on reasonable request.

Results

Study Selection and Critical Appraisal

We screened 7,716 studies and excluded 1741 duplicates and 5,319 that did not meet the eligibility criteria. From the databases, we identified 656 studies as being potentially relevant, and from the MJFF Data Resources and reference lists, we identified 18. During the full-texts analysis, we could not retrieve 1 dissertation,²⁵ physically stored at the University of New Delhi library. We did not find manuscripts related to this dissertation.

We performed the full-text assessment from the 673 retrieved studies and excluded 603 that did not meet eligibility criteria. We identified 24 studies in languages other than English (Spanish n = 9; Chinese n = 5; Turkish n = 4; German and Hungarian n = 2 each; and French and Portuguese n = 1 each), and we excluded them for feasibility reasons (translation costs). We also identified 312 studies investigating only 1 category of symptoms, either motor (n = 70) or nonmotor (n = 214) or investigating both categories but not their effect on an individual's HRQOL (n = 28). We identified 147 randomized clinical trials investigating an individual's HRQOL outcome of perceived symptoms related to a specific intervention or service (e.g., data reporting people's perceptions of using deep brain stimulation). We excluded them to avoid selection bias.

Ultimately, we included 70 studies of qualitative (n = 13), quantitative (analytical, n = 52; descriptive, n = 1), and mixed-method (convergent, n = 2; sequential, n = 2) designs (Figure 1). Critical appraisal for 8 (61.5%) qualitative and 4 (100%) mixed-method studies was 7 of 7 possible scores. Critical appraisal for quantitative studies ranged from 4 to 7 of 7 possible scores. Of these, 51 (72.5%) scored 7 of 7 possible scores (eAppendix 6, links.lww.com/WNL/D332).

Figure 1. Search Results and Study Selection and Inclusion Process.

Study Characteristics

The 70 studies were published between 1997 and 2022, with an expected increase in publications in the next decade (total expected = 101) (Table 2; additional data are listed in eAppendix 7, links.lww.com/WNL/D332 in the Supplement).

Table 2.

Characteristics of the Studies and Sample

Variable (n = 70, 100%)	n (%)
Characteristics of the studies
Year of publication
1997–2001	2 (2.9)
2002–2006	4 (5.7)
2007–2011	16 (22.9)
2012–2016	22 (31.4)
2017–2022	26 (37.1)
Continent
Europe	33 (47.1)
North America	19 (27.1)
Asia	15 (21.4)
South America	2 (2.9)
Australia/Oceania	1 (1.4)
Design
Quantitative	53 (75.7)
Cross-sectional analytic	43 (81.1)
COA	42 (97.7)
Research questionnaire	1 (2.3)
Cohort (follow-ups ranging from 1 to 8 y)	8 (15.1)
COA (baseline data)	8 (100.0)
Descriptive	2 (100.0)
Survey	1 (50.0)
Research questionnaire	1 (50.0)
Qualitative	13 (18.6)
Descriptive	13 (100.0)
In-depth interview	9 (69.2)
In-depth interview, focus group	2 (15.4)
Survey and focus group	1 (7.7)
Focus group	1 (7.7)
Mixed-method	4 (5.7)
Sequential exploratory design	2 (50.0)
Convergent	2 (50.0)
Setting
Outpatient clinic	68 (97.1)
Community (Association, Foundation, Database, Nursing Home)	2 (2.9)
COAa
PDQ-39	21 (27.6)
ZBI	13 (17.1)
EQ-5D	12 (15.8)
SF-36	6 (7.9)
RQ	5 (6.6)
PDQ-8	4 (5.3)
CSI, CBI, SCOPA-PS, NHP (each)	2 (2.6)
HUI, WHOQOL-BREF, BAS, PDQ-Carer, FCI, LiSat-11, MCSI (each)	1 (1.3)
Characteristics of the PwP sample of the studies
Age of the patients (mean)b	67.3 (−)
Disease duration (mean)c	7.5 (−)
Participants and PD stage
Studies with people with PDd	48 (68.6)
All stages	37 (76.6)
Earlyf	10 (21.3)
Advancedf	1 (2.1)
Studies with care partnerse	19 (27.1)
All stages	16 (84.2)
Early	1 (5.6)
Advanced	2 (11.1)
Studies with both people with PD and care partners	3 (7.1)
All stages	3 (100.0)

Abbreviations: BAS = the Burden Assessment Schedule; CBI = Caregiver Burden Inventory; COA = Clinical Assessment Outcomes; CSI = Caregiver Strain Index; EQ-5D = European Quality of Life Five Dimension; EUROHIS-QOL-8 = European Health Interview Survey–Quality of Life–8-Item Index; FCI = Family Care Inventory; HUI = Health Utilities Index; LiSat-11 = Life Satisfaction Questionnaire; MCSI = Multidimensional Caregiver Strain Index; NPH = Nottingham Health Profile; PD = Parkinson disease; PDQ-39 = the Parkinson Disease Questionnaire; PDQ-8 = the Parkinson Disease Questionnaire; PDQ-Carer = Parkinson Disease Questionnaire for Caregivers; SCOPA-PS = The SCales for Outcomes in PArkinson disease–Psychosocial Functioning; SF-36 = Medical Outcome Study Short Form; WHOQOL-BREF = World Health Organization Quality of Life Questionnaire; ZBI = Zarit Caregiver Burden Inventory.

^aSixteen studies used more than 1 COA.

^bTwo studies did not report the mean age of the sample.

^cNine studies did not report the mean disease duration.

^dTotal sample of patients in the 70 studies, n = 42,974.

^eTotal sample of CP in the 70 studies, n = 5,530.

^fTwo qualitative studies and 2 quantitative studies had samples of people with PD in the early and advanced stages of PD.

We identified studies investigating self-perception of the HRQOL effect of PD symptoms on n = 42,974 PwP and n = 5,530 CP (composed of spouses and other family members) on 5 continents (except for Africa and Antarctica), totaling 27 countries. While there was prevalence of studies conducted in developed countries from Europe (n = 33, 47.1%) and North America (n = 19, 27.1%), our review covered the self-perception of 1777 PwP and 659 CP from 13 studies from 9 developing countries: Mexico (n = 3), Brazil (n = 2), Turkey (n = 2), Iran (n = 1), Malaysia (n = 1), India (n = 1), Russia (n = 1), China (n = 1), and Jordan (n = 1) (Table 2; eAppendix 8, links.lww.com/WNL/D332).

Quantitative cross-sectional analytic studies were the most common study design (n = 43, 81.1%) and performed in outpatient settings (n = 68, 97.1%). In the quantitative and mixed-method studies, we identified 17 PROs used to measure HRQOL. The PDQ-39 was the most used in PwP (n = 21, 27.6%), while the Zarit Caregiver Burden Inventory was the most used in CP (n = 13, 17.1%).

In qualitative and mixed-method studies, we identified the prevalence of in-depth interviews for data collection (n = 11, 84.6%), followed by the focus group approach (n = 4, 30.8%). In mixed-method studies, there was an equal distribution of studies using sequential and convergent approaches.

PwP had a mean age of 67.3 years and a mean of 7.5 years of disease duration. Most of the studies had a sample of PwP (n = 48, 68.6%) representing all stages of PD (n = 37, 76.6%) (Table 2; eAppendix 9, links.lww.com/WNL/D332).

Findings of the Review

From the 70 studies, we mapped a set of 792 terms and included 604 originating from the qualitative (n = 251, 41.6%) and quantitative (n = 353, 58.4%) evidence of the studies. We excluded 188 terms because they affected the HRQOL of less than 10% of the sample in quantitative descriptive studies or did not show a significant inferential relationship with the sample's HRQOL in quantitative analytical studies.

We cross-mapped the 604 terms with all 11 predefined thematic categories of perceived motor (n = 282, 46.7%) and nonmotor symptoms (n = 322, 53.3%) (Table 3). We did not find a significant difference in the sample data variation between the qualitative and quantitative groups across the categories of motor and nonmotor symptoms (χ² = 0.216, df = 1, p value = 0.642).

Table 3.

Distribution of Included Terms According to the Source Component

Self-perceived symptoms	Components		Total
	Qualitativea	Quantitativeb
	n (%)	n (%)	n (%)
Motor	120 (19.9)	162 (26.8)	282 (47)
Miscellaneous	38 (6.3)	84 (13.9)	122 (20)
Motor functionality	31 (5.1)	79 (13.1)	110 (18.2)
Overall motor symptoms	7 (1.2)	5 (0.8)	12 (1.99)
Postural instability (and gait)	29 (4.8)	30 (5.0)	59 (9.8)
Gait, balance, posture, and falls	29 (4.8)	30 (5.0)	59 (9.77)
Rigidity	26 (4.3)	7 (1.2)	33 (5.5)
Speech disorder	10 (1.7)	3 (0.5)	13 (2.15)
Rigidity	8 (1.3)	3 (0.5)	11 (1.82)
Difficulty with handwriting	6 (1.0)	1 (0.2)	7 (1.16)
Breathlessness/dyspnea at rest or in effort	2 (0.3)	—	2 (0.33)
Tremor	15 (2.5)	9 (1.5)	24 (4)
Tremor	15 (2.5)	9 (1.5)	24 (3.97)
Bradykinesia	8 (1.3)	8 (1.3)	16 (2.6)
Bradykinesia	7 (1.2)	8 (1.3)	15 (2.48)
Hypomimia	1 (0.2)	—	1 (0.17)
Motor complications	4 (0.7)	24 (4.0)	28 (4.6)
Muscle cramps, spasms	2 (0.3)	1 (0.2)	3 (0.5)
Motor fluctuations	2 (0.3)	17 (2.8)	19 (3.15)
Dyskinesia	—	5 (0.8)	5 (0.83)
Dystonia	—	1 (0.2)	1 (0.17)
Nonmotor	131 (21.7)	191 (31.6)	322 (53)
Neurobehavioral	41 (6.8)	95 (15.7)	136 (23)
Mood	22 (3.6)	57 (9.4)	79 (13.1)
Cognition	17 (2.8)	25 (4.1)	42 (6.95)
Other psychiatric	2 (0.3)	13 (2.2)	15 (2.48)
Sensorial and others	36 (6.0)	25 (4.1)	61 (10)
Fatigue	16 (2.6)	9 (1.5)	25 (4.14)
Pain	10 (1.7)	12 (2.0)	22 (3.64)
Weight loss/gain	3 (0.5)	—	3 (0.5)
Hyposmia, anosmia, ageusia	3 (0.5)	1 (0.2)	4 (0.66)
Visual problems	2 (0.3)	—	2 (0.33)
Sensory complaints	2 (0.3)	2 (0.3)	4 (0.66)
Restless legs syndrome	—	1 (0.2)	1 (0.17)
Dysautonomia	22 (3.6)	35 (5.8)	57 (9.4)
Gastrointestinal	15 (2.5)	17 (2.8)	32 (5.3)
Urogenital	5 (0.8)	12 (2.0)	17 (2.81)
Thermoregulation	1 (0.2)	2 (0.3)	3 (0.5)
Cardiovascular	1 (0.2)	2 (0.3)	3 (0.5)
Autonomic dysfunction	—	2 (0.3)	2 (0.33)
Miscellaneous	21 (3.5)	16 (2.6)	37 (6.1)
Nonmotor functionality	21 (3.5)	12 (2.0)	33 (5.46)
Overall nonmotor symptoms	0.0)	4 (0.7)	4 (0.66)
Sleep	11 (1.8)	20 (3.3)	31 (5.1)
Nighttime sleep disorder	8 (1.3)	16 (2.6)	24 (3.97)
Daytime sleepiness	2 (0.3)	4 (0.7)	6 (0.99)
Sleep disorder	1 (0.2)	—	1 (0.17)
Grand total	251 (41.6)	353 (58.4)	604 (100)

^aOriginated from all 13 qualitative and all 4 mixed-method studies.

^bOriginated from all 53 quantitative and two of 4 mixed-method studies.

Effect of PD on HRQOL of PwP

Quantitative evidence on the effect of PD on HRQOL of PwP came from the studies investigating all stages, n = 33; early PD, n = 5; and advanced PD, n = 3. Qualitative evidence came from studies investigating all stages, n = 7; early PD, n = 5; and advanced PD, n = 2.

Effect of PD on HRQOL of CP

Quantitative evidence on the effect of PD on HRQOL of CP came from studies investigating all stages, n = 16; early PD, n = 1; and advanced PD, n = 2. Qualitative evidence came from studies investigating all stages, n = 3; early PD, n = 0; and advanced PD, n = 1.

Effect of PD on HRQOL of PwP and CP

One study provided qualitative evidence on the effect of PD on HRQOL of PwP (early PD only) and CP, and 2 studies provided quantitative evidence (only on all stages of PD).

Meta-Synthesis of Qualitative Evidence

We included qualitative components from 17 studies (all 13 qualitative^3,4,11,26-35 and all 4 mixed-method study designs^36-39), from which we extracted 251 (41.6%) terms (nonmotor = 131, 21.7%). We mapped the findings into the 11 categories of symptoms. We tabulated the symptoms according to 3 disease severity categories (all stages, early, and advanced PD) and constructed figures according to the source of information (PwP, care partner, both in combination).

For studies assessing symptoms covering all PD severities, 4 symptom subcategories—two motor (“tremor,” “gait, balance, posture, falls”) and 2 nonmotor (“cognition” and “fatigue”) were cited in 100% of studies affecting HRQOL of PwP And CP.

For studies assessing symptoms covering the early stage of PD, 4 symptom subcategories were most impactful on HRQOL of PwP: 2 motor (“tremor” and “gait, balance, posture, falls”) and 2 nonmotor (“mood” and “cognition”). Unfortunately, our sample did not have studies assessing the perception of CP and the HRQOL effect on early-stage PD symptoms.

For studies assessing symptoms covering advanced PD stages, the symptoms that most affect the HRQOL of PwP and CP were “gait, balance, posture, and falls” (motor), cited in 100% of studies (Figure 2) (additional data are listed in eAppendices 10, 11, 12, and 13, links.lww.com/WNL/D332).

Figure 2. Heatmap of Self-Reported Symptoms Extracted From Qualitative Components of Qualitative and Mixed-Method Studies.

CP = care partners; PwP = People with Parkinson disease. *2 qualitative studies had samples of people with PD in the early and advanced stages of PD.

Meta-Synthesis of Quantitative Evidence

We identified quantitative components in 55 studies (all 53 quantitative^40-51,e1-e41 and 2 mixed-method study designs^38,39), from which we extracted 353 (58.4%) terms (nonmotor = 191, 31.6%). We mapped the findings into the 11 categories of motor and nonmotor symptoms according to the source of information (PwP and/or CP) in studies involving all stages of PD, early or advanced stages of PD only (Figure 3) (eAppendix 14, links.lww.com/WNL/D332).

Figure 3. Heatmap of Self-Reported Symptoms Extracted From Quantitative Components of Quantitative and Mixed-Method Studies.

CP = care partners; PwP = people with Parkinson disease. a: 1 quantitative study had samples of people with PD and CP in all stages of PD. b: 1 quantitative study had samples of people with PD in all, early, and advanced stages of PD. c: 1 quantitative study had samples of people with PD in the early and advanced stages of PD. *: In one study with a sample of 443 PwP and CP, less than 10% of the sample cited the symptom as the one that most affected their HRQOL.

For studies assessing symptoms covering all PD severities, 3 symptom subcategories—one motor (“motor functionality”) and 2 nonmotor (“mood” and “cognition”)—similarly affected PwP and CP (Figure 3).

For studies focusing on the early PD stage, 8 symptom subcategories were significantly cited as having the most effect on HRQOL of PwP and CP, according to the criteria applied for descriptive quantitative studies and analytical. Five of the symptoms were motor (“tremor,” “rigidity,” “bradykinesia,” “overall motor symptoms,” “motor functionality”), and 3 were nonmotor (“mood,” “nonmotor functionality,” and “cognition”). The nonmotor symptom “mood” was cited in 6 studies, “nonmotor functionality” was cited in 2 studies, and all the others were cited in 1 study with a sample of 443 PwP and CP.³⁹ This same study also had samples smaller than 10% of PwP and CP, citing 18 symptom subcategories as most impactful for HRQOL of PwP and CP: 7 motor (“difficult handwriting,” “speech disorder,” “arm swing decrease,” “gait, balance, posture, and falls,” “dyskinesia,” “dystonia,” and “motor fluctuations”) and 11 nonmotor (“urogenital,” “gastrointestinal,” “nighttime sleep disorder,” “mood,” “cognition,” “fatigue,” “hyposmia, anosmia, ageusia,” “pain,” “restless legs syndrome,” “sensory complaints,” “visual problems”) (Figure 3).

For studies assessing symptoms covering advanced PD stages, the symptom that most affected HRQOL of PwP was “pain” (nonmotor), cited in all studies. By contrast, other psychiatric symptoms (nonmotor) affected the HRQOL of CP (Figure 3; eAppendix 14, links.lww.com/WNL/D332).

Convergence of Qualitative With Quantitative Studies

We integrated all findings, and across all PD stages, PwP and CP considered 5 domains more affecting their HRQOL, namely: “motor functionality,” “mood,” “cognition,” “gait, balance, posture, and falls,” and “nighttime sleep disorders.” In early disease, PwP and CP considered “mood” the domain that most affected their HRQOL. In advanced PD, PwP considered “pain” the domain that most affects their HRQOL, while CP considered “psychiatric symptoms.” The domain “gait, balance, posture, and falls” was equally considered by PwP and CP as the second domain most affecting HRQOL in the advanced PD stage (Figure 4; eAppendices 15a, b, and c, links.lww.com/WNL/D332).

Figure 4. Ranking of Domains That PwP and Care Partners Considered More Impactful For Their HRQOL According to the Disease Stage.

HRQOL = health-related quality of life; PwP = people with Parkinson disease.

Discussion

Despite the growing interest in how PD symptoms affect individuals, most PD HRQOL studies have quantitatively explored, at a single point in time, the effect of specific symptoms on the lives of PwP and CP who visit outpatient care settings.^40-51,e1-e41

Although relevant, single time point data limit the understanding of evolving HRQOL issues in PD and, consequently, its treatment outcomes. In addition, implementing the traditional investigator-driven symptom inventory to the detriment of approaches using the patient's voice and disregarding the PD burden on CP poses a risk to the care quality.^e41 Because physical and emotional needs change during PD, management adaptations can be anchored in populational reviews and individualized visit-to-visit assessments.^e19 We emphasize the need for further longitudinal studies investigating what matters to PwP and CP. Longitudinal studies offer advantages such as increased statistical power and the capacity to estimate a wider array of conditional probabilities compared with repeated cross-sectional studies. Through more extensive analysis, including evaluating causal links and effects, researchers can identify shifts or modifications in the traits of the target population, both collectively and individually.^e42

By integrating quantitative and qualitative data from studies investigating this multidimensional construct worldwide, including those implemented in developing countries, we could map the PD symptoms that most affect the HRQOL of PwP and CP. Based on the individual's perspective reflective of the stage of PD, these data can help better tailor the assessment of health care and treatment in this population. This approach converges with the recommendations of health regulatory agencies, such as the US Food and Drug Administration (FDA), which recognizes the value of gathering information from individuals about the specific experiences that matter most to them and their perspectives on the significant benefits of treatment.¹⁰

In this review, the PD stages fall into 3 categories, of which 2 showed that motor and nonmotor symptom groups equally affect the HRQOL of PwP and CP. This phenomenon could be attributed to changes in the inherent characteristics of PD and the limited effectiveness of therapies on certain core PD symptoms. While motor symptoms are commonly addressed through dopaminergic replacement therapy, nonmotor symptoms may become more pronounced because they are less responsive and necessitate diverse approaches.^e43 For instance, addressing the emotional aspects such as mood swings and sleep disturbances and enhancing independence and socialization hold greater significance in alleviating CP burden and chronic sadness rather than focusing solely on physical care.^e44 The convergence of priorities between PwP and CP is evident across studies spanning various disease severities and early stages, implying the potential for integrated treatment strategies benefiting both parties.

In advanced PD stages, the emergence of pain as an issue for PwP was unexpected. This observation underscores a vital area for further research and exploration of treatment options. PD-related pain, a complex issue with multiple contributing factors, is insufficiently studied.^e45-e48 Still, its recognition by PwP, despite its comparatively lower recognition by CP, suggests that it might go unnoticed or be perceived differently within a household. Explicit inquiries directed at the PwP are crucial for its accurate assessment.

An intriguing finding is the transition in dominant nonmotor symptoms from mood-related concerns in early PD stages to cognitive and behavioral issues, including psychosis, in advanced stages. Research indicates that CP undergo considerable challenges (such as isolation, depression, and suicidal ideation) when tending to individuals with psychosis.^e49,e50 This underscores the necessity for comprehensive medical guidance as the disease advances. Strikingly, despite these established findings, certain CP note that neurologists seldom inquire about psychosis during routine consultations, though evaluating those symptoms is the established norm in PD.^e51,e52 This highlights a significant practical oversight that warrants deeper investigation and consideration.

Addressing this point of view raised by the CP has supported evidence that they are crucial allies in the administration of health regimens positively and negatively affecting the care recipient. On the positive side, an allied CP is associated with a decreased likelihood of nursing home placement, improved medication adherence and treatment regimens, and higher quality of care.^e53–e55 Conversely, a distressed CP has been associated with higher institutionalization rates, reduced care recipient HRQOL, and decreased psychological well-being.^e56,e57 Research indicates that CP living with and looking after someone with a chronic illness such as PD can experience limitations in their activities and social engagement.^e58 Hence, it is crucial to extend the care focus beyond the person diagnosed with PD, aiming to uphold their overall health and well-being. Implementing interventions that guide CP in addressing their diverse needs can foster healthier and more positive relationships, enabling them to support their loved ones while staying socially connected.^e59

Although the included studies were generally of good methodological quality and represented a sizeable PwP and CP sample, we identified some limitations. First, we tallied but did not include non-English language studies, which may have affected our findings because language and culture are linked, and we had few non-North American or non-European studies. Second, the variety of outcome measures was too diverse to conduct a meta-analysis, and the data from the quantitative components had to be “qualitized.”

Third, although our purpose was to highlight the PwP and CP voices regarding the HRQOL effect of PD symptoms, the findings (mainly coming from qualitative evidence) needed to be categorized according to scientific language so that they could be analyzed. This “medicalization” process assumes that the professionals understand the intent of the informant and place an interpretive barrier to the direct voice. The most salient example of such a threat is when PwP or CP identify “shaking” or “jerking” as a symptom of importance. Is this a tremor (a disease) or dyskinesia (a medication side effect)? The implications are profound; in this case, the voice is obscured by forced classification into terms meaningful to clinicians.

On the contrary, PwP have long complained of “fatigue.” Before the MDS-UPDRS, this symptom, confidently voiced by PwP and CP, was forced by physicians into categories of sleep abnormalities, bradykinesia, apathy, or depression. Fatigue is considered a specific symptom of importance, and this reality has mainly emerged because quantitative studies have prioritized the voice of PwP. However, our review precludes reconsidering symptom lists or clusters.

Fourth, most studies did not stratify PwP by disease stage, so we had to analyze them as a single group called “all stages.” This shows the importance of separating disease stages or following longitudinal evolutions of symptoms in studies that could be performed with the “open-field” technique and with item-by-item MDS-UPDRS Parts I and II evolutionary patterns over time.

Ultimately, the greater variety of symptoms affecting the HRQOL of PwP than that of CP can be explained by the smaller number of studies that involved this population. However, given the rising number of publications in this field, we might consider giving CP a more robust voice by “adding a few more faces behind the why we do what we do”^e60 and consistently listening to each care team member at each visit.

While different approaches have been used to capture the PwP and CP “voice,” a recurring observation is the wide variety of responses so that even the “most endorsed” symptom is still only one among many. In such studies, design and data interpretation must consider bias issues and cultural differences. As personalized medicine is increasingly prominent in PD, studies focusing on PwP and CP can be used as guides. Still, they do not replace the need for health care providers to hear each individual and evolving concerns at each visit.

Despite some similarities, the effect of PD symptoms on the HRQOL of PwP and CP differs mainly in its variety and the advanced disease stage. Therefore, symptom management needs to consider the experiences of all individuals so that high HRQOL levels can be achieved.

Glossary

COA

Clinical Assessment Outcomes

CP

care partners

HRQOL

health-related quality of life

MDS-UPDRS

Movement Disorder Society–Unified Parkinson Disease Rating Scale

PD

Parkinson disease

PDQ-39

Parkinson Disease Questionnaire

PRO

patient-reported outcomes

PwP

people with Parkinson disease

Appendix. Authors

Name	Location	Contribution
Michelle H. Tosin, PhD	Department of Neurological Sciences, Rush University	Drafting/revision of the article for content, including medical writing for content; major role in the acquisition of data; study concept or design; and analysis or interpretation of data
Christopher G. Goetz, MD	Department of Neurological Sciences, Rush University	Drafting/revision of the article for content, including medical writing for content; major role in the acquisition of data; study concept or design; and analysis or interpretation of data
Glenn T. Stebbins, PhD	Department of Neurological Sciences, Rush University	Drafting/revision of the article for content, including medical writing for content; study concept or design; and analysis or interpretation of data

Study Funding

This manuscript has been funded by the Michael J. Fox Foundation for Parkinson's Research (MJFF) (MJFF-022391).

Disclosure

M.H. Tosin: During the reporting time, Dr. Tosin led the working group supported by the International Parkinson and Movement Disorder Society (M.D.S). Some funds from the mentioned study supported Dr. Tosin's salary and her research efforts at Rush University. C.G. Goetz: Consulting or Advisory Board Membership with honoraria: Genentech (Creative Group), Psychogenics Inc, and CHDI. Grants/Research: Funding to Rush University Medical Center from N.I.H., Department of Defense, and the Michael J. Fox Foundation for research conducted by Dr. Goetz. Honoraria: Faculty and Program Management stipend from the International Parkinson and Movement Disorder Society. Volume Editor stipend from Elsevier Publishers. Guest faculty lecturer at Arizona State University. Intellectual Property Rights: None. Ownership Interests: None. Salary: Rush University Medical Center. G.T. Stebbins: reports consulting and advisory board membership with honoraria from Acadia, Pharmaceuticals, Adamas Pharmaceuticals, Inc., Biogen, Inc., Ceregene, Inc., CHDI Management, Inc., Cleveland Clinic Foundation, Ingenix Pharmaceutical Services (i3 Research), MedGenesis Therapeutix, Inc., Neurocrine Biosciences, Inc., Pfizer, Inc., Tools-4-Patients, Ultragenyx, Inc., and the Sunshine Care Foundation. G.T. Stebbins received grants and research from the NIH, Department of Defense, Michael J. Fox Foundation for Parkinson's Research, Dystonia Coalition, CHDI, Cleveland Clinic Foundation, International Parkinson and Movement Disorder Society, and CBD Solutions. G.T. Stebbins reports honoraria from the International Parkinson and Movement Disorder Society, American Academy of Neurology, Michael J. Fox Foundation for Parkinson's Research, Food and Drug Administration, the NIH, and Alzheimer's Association. G.T. Stebbins received a salary from Rush University Medical Center. Go to Neurology.org/N for full disclosures.