Social deprivation and exclusion in Parkinsons disease: a cross-sectional and longitudinal study

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • This study is based on the Survey of Health, Ageing and Retirement in Europe (SHARE), a cross national panel database on microdata on health, socioeconomic status and social and family networks of European individuals.

  • The study used internationally comparable cross-sectional and longitudinal data to examine the impact of social deprivation and exclusion on people with Parkinson’s disease (PwPD).

  • The study considered PwPD, prodromal PD and without PD.

  • Common PD-related measures (eg, Hoehn and Yahr Scale) are not available in the SHARE dataset, and the diagnosis is based on self-reporting.

Introduction

Social aspects have a significant impact on health in various chronic conditions.1–3 Motor impairment along with emotional and communicative changes are thought to disrupt social functioning, but little is known about the social determinants of well-being in Parkinson’s disease (PD).4 In particular, material deprivation, social deprivation and social exclusion are associated with poor health, physical inactivity, poor cognitive function, mood disorders and pain in distinct chronic disorders.5–8

Material deprivation refers to a lack of access to material economic resources and describes the situation of an individual who cannot reach the minimum standard of material well-being.9 People struggle daily to feed, clothe and house themselves and their families. Materially deprived people face persistent hardships that affect their physical, social and mental well-being. Whereas social deprivation extends far beyond economic factors and refers to the reduction or prevention of cultural and social interactions between an individual and the rest of society. Socially deprived people lack freedom of choice, opportunity, political voice and dignity. With few options to improve their circumstances, they are exposed to a variety of stressors that deplete individual resources.10 Persistent deprivation can lead to social exclusion, which can be broadly interpreted as an individual’s inability to participate in the basic political, economic and social activities of the society.11 Therefore, the concepts of deprivation and social exclusion share similar focus. However, while deprivation emphasises a lack of resources that contribute to a lack of social participation, social exclusion emphasises a lack of participation in a broader range of social, cultural and political activities.12

Therefore, social exclusion is a multidimensional and relative concept.13 This means that a person can only be socially excluded compared with a certain social group at a certain place and time.11 Social exclusion has two main aspects. First, economic-structural exclusion refers to distributional material (income and goods) and non-material dimensions (social rights). Second, sociocultural exclusion refers to the relational dimensions of social integration (social relations and networks) and normative integration (values and norms).14 15 Social exclusion has rapidly become common in debates on deprivation and policies that combat deprivation. However, to date, there has been limited attention in PD research.16 Though it is highly likely that social exclusion is present in PD, because older and chronically ill individuals can quickly become cut-off from a wider society.17 Additionally, social exclusion has been found to impact health and well-being7 18 and may, therefore, also explain some variance in functional decline and quality of life (QoL) in people with PD (PwPD). Finally, it seems promising to combine PD-related biomedical symptoms with sociocultural dimensions19 to pave the way for new holistic therapeutic approaches.

However, the extent to which deprivation and social exclusion occur in PD and how they influence QoL remain an open question. Especially with respect to the growing population of older PwPD, the measurement and monitoring of deprivation and social exclusion are crucial to design policies and interventions aimed at improving well-being. Therefore, the objectives of the study were: (1) to describe the prevalence of social deprivation and exclusion in PD, (2) to evaluate the risk factors correlated with social deprivation and exclusion, (3) to assess the impact of social deprivation and exclusion on QoL and (4) to evaluate the occurrence of social deprivation and exclusion prior to the onset of PD.

Materials and methods

Data source and participants

Data were taken from waves 5 (2013, n=66 188)20 and 6 (2015, n=68 186)21 of the Survey of Health, Ageing and Retirement in Europe (SHARE); for technical details about sampling, we refer to the SHARE working paper series.22 SHARE is a multidisciplinary, cross-national and longitudinal research project that focuses on community-dwelling adults aged 50 years or older in European countries (www.share-project.org). SHARE is the largest pan-European social science panel study that provides internationally comparable longitudinal data on the public health and socioeconomic living conditions of European individuals.

Of note, not all items in the SHARE questionnaires were assessed in every wave. Our variables of interest, deprivation and social exclusion, were assessed in wave 5; therefore, we considered participants who took part in wave 5 and answered the item asking if they have or do not have PD (SHARE variable ph006d12). No other inclusion or exclusion criteria were defined other than the presence of PD. In addition, we looked at the association of deprivation and the longitudinal course of PD. For this purpose, we selected the SHARE participants of wave 5, who also took part in wave 6, and again, indicated if they have or do not have PD. Subsequently, based on this information, we divided the participants of wave 5 into three subgroups: participants with PD (already reporting PD in wave 5), prodromal PD (newly reporting PD in wave 6 but not in wave 5) and participants without PD (not reporting PD in waves 5 and 6).

Variables

Wave 5 provided three compound measures: material deprivation, social deprivation and social exclusion. All items are available at the SHARE homepage SHARE_release_guide_8-0-0.pdf (share-project.org).

First, the Material Deprivation Index relies on material deprivation items, which capture the ability of households to afford specific types of goods and services. These complement income-based measures of material conditions and are increasingly used as indicators to identify insufficient material resources.23 They cover aspects of the economic circumstances of households, such as the ability to afford meat or fruit more often than three times per week, the affordability of a number of specific items such as groceries and holidays away from home, the necessity to limit expenses on a number of items such as shoes or heating to keep living costs down and the inability to see a doctor because of costs.23 The Material Deprivation Index can take values between 0 and 1 (with higher values indicating greater deprivation).

Second, the social deprivation index combines information on items related to participation in everyday life, social activities and neighbourhood quality into a single index.15 The social deprivation index can take values between 0 and 1 (with higher values indicating greater deprivation).

Third, using these two deprivation indices, a two-dimensional proxy for social exclusion was constructed to identify those with high levels of material and social deprivation (severe deprivation indicator). Here, individuals whose deprivation measures were above the 75th percentile of the distribution in both dimensions were classified as severely deprived, indicating a risk of social exclusion.15

Quality of life (QoL)

The Control, Autonomy, Self-realization, and Pleasure Scale (CASP-12)24 is one of the most common internationally used QoL measures. It is composed of subscales of control, autonomy, self-realisation and pleasure. The 12 items are assessed on a 4-point Likert scale (‘often,’ ‘sometimes,’ ‘rarely’ and ‘never’), and the resulting sum score ranges from 12 to 48 with higher values indicating better QoL. SHARE provides the CASP-12 variable (casp) as a generated variable in the gv_health module.

Instrumental activities of daily living (IADL)

We used the measurement of IADL to assess participants’ motor function. A modified version of the IADL was used in SHARE wave 5 with seven activities: using a map to get around in a strange place, preparing a hot meal, shopping for groceries, making telephone calls, taking medications, doing work around the house or garden and managing money. The total score ranged from 0 to 7. The higher the index, the more difficult it is with these activities and the lower the mobility of the respondent. SHARE provides IADL as a generated variable in the gv_health module.

Grip strength

We used grip strength as a general biomarker of poor health status25: Using a dynamometer, maximum grip strength was measured two times for each hand. Grip strength was derived from the gv_health module.

Depressive symptoms

The EURO-D Scale26 consists of the following items: depression, pessimism, suicidality, guilt, sleep, interest, irritability, appetite, fatigue, concentration (on reading or entertainment), enjoyment and tearfulness. It is summed up in the EURO-D variable (eurod) as a generated variable in the gv_health module. The maximum score is 12 for very depressed, and the minimum score is 0 for not depressed.

Cognitive function

The 10 words delayed recall test (ranges 0–10; cf016tot as generated variables in the gv_health module) and verbal fluency (name as many animals as possible; cf010_ as generated variable in the gv_health module) were used as cognitive function measures.

Marital status

The marital status was derived as a dichotomous variable from the nominal coded variable mstat in the gy_imputations module. Thereby, participants were classified as married and living together with spouse or in a registered partnership versus participants who are living separated, never married, divorced or widowed.

Other variables that are known to be related to social exclusion were age, gender, eyesight at distance (higher values indicating poorer vision), eyesight while reading (higher values indicating poorer vision), hearing (higher values indicating poorer hearing), education (duration of school education in years), body mass index (BMI), number of chronic diseases (0–9) and country.7 27–29

Variables were treated as missing and excluded from the analysis in case of missing information (including ‘don’t know’ and ‘refusal’).

Statistical analyses

All analyses were conducted using IBM SPSS Statistics (V.25) and JASP (V.0.16). Statistical significance was set at p<0.05. Descriptive statistics were used to characterise the samples. Normality was tested using the Shapiro-Wilk test. Correlations between variables were tested using Spearman’s correlation rs. The effect sizes were considered low (|rs|=0.1), moderate (|rs|=0.3) or strong (|rs|=0.5). Group comparisons were performed using the Mann-Whitney U test for non-normally or ordinal data, χ2 test for nominal data or analysis of variance (ANOVA) with Bonferroni post hoc test, where appropriate. The effect sizes r of significant group differences were determined using the rank biserial correlation for the Mann-Whitney U test and the phi coefficient for the χ2 test. Exploratory multiple linear regressions were used to analyse the association between CASP-12 (dependent variable), social deprivation, material deprivation and several independent variables using a stepwise selection algorithm and the Akaike information criterion as the selection criterion. For regression analyses, we included all variables correlating with CASP-12 as indicated by Spearman’s correlation |rs| of at least 0.2, and the nominal variable country. Multicollinearity was tested using the variance inflation criterion.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Results

Descriptive statistics

Participants in wave 5 can be divided into the following three subgroups: people with PD (already reporting PD at wave 5; n=559), prodromal PD (newly reporting PD at wave 6; n=215) and people without PD (not developing PD until wave 6; n=46 737). Descriptive statistics for all variables for people with PD (n=559) and without a diagnosis of PD in wave 5 (prodromal PD and no PD; n=46 952) are provided in online supplemental table 1. PwPD had a median age of 77 years with an IQR of 69–83 years and were older than people without PD (median=66 years, IQR=59–74). Of the PwPD, the majority were male participants, while there were more female participants without PD (47.9% vs 56.7%). No differences were seen regarding BMI and marital status.

Supplemental material

Table 1

Comparison of people with prodromal PD, PD and without PD in wave 5 (analysis of variance, ANOVA)

How common is social exclusion in people with PD?

PwPD experienced more social deprivation (median=0.21, IQR=0.13–0.39) than people without PD in wave 5 (median=0.13, IQR=0.08–0.22) (p<0.001, r=0.357). The level of social deprivation was higher than many other chronic conditions in SHARE and comparable to the social deprivation level in people with dementia and affective disorders (online supplemental figure 1). Additionally, PwPD experienced more material deprivation (median=0.08, IQR=0.00–0.27) than people without PD in wave 5 (median=0.00, IQR=0.00–0.22) (p=0.001, r=0.081). Using the severe deprivation indicator, 20% of the PwPD were at risk of social exclusion (online supplemental table 1).

What are the correlated risk factors of deprivation?

In PwPD, univariate analyses showed that social deprivation correlated moderately or strongly with QoL (p<0.001, r=−0.655), EURO-D (p<0.001, r=0.467), IADL (p<0.001, r=0.434), material deprivation (p<0.001, r=0.430), education (p<0.001, r=−0.402), cognitive functioning (verbal fluency: p<0.001, r=−0.368; recall test: p<0.001, r=−0.307), grip strength (p<0.001, r=−0.318) and eyesight at distance (p<0.001, r=0.324) (online supplemental figure 2; online supplemental table 2) and was higher in women (p=0.001, r=0.177) (online supplemental table 3). Likewise, material deprivation moderately correlated with QoL (p<0.001, r=−0.490), education (p<0.001, r=−0.459), social deprivation (p<0.001, r=0.430) and EURO-D (p<0.001, r=0.319) (online supplemental figure 3; online supplemental table 2) and was higher in women (p=0.005, r=0.147) (online supplemental table 3). Group comparisons between PwPD with and without severe deprivation (ie, risk for social exclusion) supported these correlations (online supplemental table 4).

How does deprivation influence QoL?

In PwPD, social deprivation alone accounted for 35% of the variance in QoL (F (1, 468)=250.6, p<0.001), whereas material deprivation alone accounted for 21% (F (1, 468)=122.1, p<0.001). Both social and material deprivation remained significant predictors of QoL after adjusting for cofactors (figure 1, online supplemental table 5). Being severely deprived (ie, being at risk for social exclusion) alone accounted for 15% of the QoL variance (F (1, 468)=81.3, p<0.001) and remained a significant predictor of QoL after adjustment for cofactors (online supplemental table 5).

Figure 1
Figure 1

Predictors of quality of life (CASP-12) in the linear regression in people with PD in wave 5. CASP-12, Control, Autonomy, Self-realization, and Pleasure Scale; IADL, instrumental activities of daily living; PD, Parkinson’s disease.

Does the occurrence of social deprivation and exclusion precede the onset of PD?

Longitudinal data analyses were performed to determine whether social deprivation precedes the onset of PD. From waves 5 to 6 of the SHARE surveys, 215 individuals were newly diagnosed with PD. Compared with non-PD participants, these prodromal PD participants already had higher social deprivation in wave 5 before they reported onset of PD in wave 6 (F (2, 41 939)=118.5, p<0.001, figure 2, table 1). Their social deprivation scores in wave 5 did not differ from those who already had PD in wave 5 (post hoc Bonferroni p>0.05).

Figure 2
Figure 2

Boxplot for social deprivation in people with prodromal Parkinson’s disease (PD), PD and without PD in wave 5.

The risk of social exclusion (above the 75th percentile of both deprivation dimensions) was present in 17.4% (n=29) of the prodromal PD patients, but only in 11.1% (n=4460) of the participants without PD in wave 5 (F (2, 40791)=19.2, p<0.001) (post hoc Bonferroni p=0.031). Before the onset of PD, these prodromal PD participants also had poorer QoL, worse cognitive function, higher number of chronic diseases, more depressive symptoms, more severe sensory impairments (vision and hearing), more limitations in IADL and worse handgrip strength compared with people without PD (table 1, online supplemental table 6).

Discussion

Our study has three main findings. First, PwPD are at risk for social deprivation, and about 20% are at risk of social exclusion. Second, social and material deprivation are predictors of impaired QoL. Third, social deprivation can occur before the disease onset in the prodromal PD stage.

To the best of our knowledge, there are no comprehensive data on social deprivation in PD. Therefore, to discuss our findings, we must consider findings from other chronic disorders and thematically related constructs in PD, such as research on social isolation and loneliness. Social isolation has been inconsistently defined and conceptualised. In general, people with social isolation experience a deficiency of fulfilling and quality relationships, may lack a sense of social belonging and have limited social engagement with often few social contacts.30 31 In contrast, social exclusion is a multidimensional construct defined as a deficiency in basic material and social needs.15

Many studies on social isolation came from COVID-19 research, showing the isolation and disruption of social networks and usual activities in PD during the COVID-19 pandemic.32–34 Few studies have examined social isolation or similar constructs before the COVID-19 era.30 These studies have concluded that social isolation can lead to worsening of neuropsychiatric symptoms and motor function,35 36 although these effects are not restricted to PD.37 Evidence for the influence of social exclusion on health outcomes comes from other studies of chronic conditions, such as dementia,38 diabetes,39 HIV40 and rheumatoid arthritis.41

It has been observed that deprived people are less physically active.6 Moreover, perceived deprivation can affect cognitive function leading to a weaker ability to self-regulate one’s own behaviour.8 People who were socially excluded or deprived were more likely to report long-term illness and disability, score lower on self-rated health7 or report more pain.5 Taken together, the literature supports the association observed between deprivation and poorer physical function in PD. This means that social and contextual factors should be considered as important cofactors in patients with PD with poor motor function or a high non-motor burden.

Interestingly, we found that social deprivation had already occurred prior to PD diagnosis. The prodromal phase before disease onset is characterised by a range of symptoms that can antedate the first appearance of classical motor signs in PD. In particular, 2 years before PD diagnosis, motor features (tremor, rigidity, balance impairments, stiffness and shoulder pain), autonomic features and neuropsychiatric disturbances (memory problems, anxiety or depression, cognitive decline and apathy) can occur.42 These prodromal symptoms are not specific to PD and can be caused by other diseases. Several attempts have been made to identify the best predictive set of reliable indicators of prodromal PD derived from clinical data, tissue biopsies, quantitative motor assessment and imaging.43 Interestingly, the prevalence of the most common prodromal symptoms (depression, rapid eye movement sleep behaviour disorder, constipation) is lower than that of social deprivation.44 45 Therefore, future longitudinal studies should analyse whether social deprivation can be regarded as an additional ‘social’ prodromal marker for PD.

We found that social deprivation in patients with PD was mainly associated with depressive symptoms and limitations in IADL. However, deprivation cannot be explained solely by depressive symptoms and limitations in IADL, as many other factors contribute to this multidimensional construct. As depressive symptoms and IADL problems are present before the disease diagnosis, this might partly explain why social deprivation precedes PD onset. In addition, the prodromal PD cohort in wave 5 had worse cognitive function, a higher number of chronic diseases and sensory impairments contributing to social deprivation. However, we cannot determine whether visual problems are due to PD46 or other eye-related changes because the SHARE data do not allow us to draw causal statements about the reasons for visual problems.

Our results suggest that social deprivation and the risk of social exclusion are highly relevant and should be acknowledged by neurologists treating PwPD. Previous studies have investigated the influence of single sociodemographic, disease-related, motor and non-motor symptoms on health-related QoL. Some of these single factors were combined in the construct and model of deprivation here.15 23 The most important determinant of QoL in PwPD is the presence of depressive symptoms. This result parallels the findings of many other studies.47 There is a bidirectional relationship between depression and health and socioeconomic parameters. Higher depressive symptomatology is associated with poorer self-perception of physical health, female sex, economic difficulties, widowhood, lower levels of activity and exercise and lower educational level.48 In addition, age, living status (household members), education, comorbidities, disease disability and cognition were important determinants of QoL in our and other previous studies.47 Given the large influence of social deprivation on QoL, we argue that QoL research in PD should not only focus on PD-related biomedical measures but also acknowledge the impact of social determinants.19 49 Another important aspect of this field is the relationship between social deprivation and QoL outcomes within and across societies and countries.50–52 As demonstrated in our analysis and other previous studies, the country had a significant impact on measures of deprivation and, consequently, on QoL.15 23 Moreover, living conditions also affect QoL in PD; in rural areas, neighbourhood exclusion explains more variance in self-rated health and psychological well-being, while in urban areas, exclusion from services explains more variance in well-being.28 This finding underlines the need to incorporate social determinants into PD research and future treatment approaches.

Our study has some limitations. Common PD-related measures (eg, Hoehn and Yahr Scale) are not available in the SHARE dataset, and the diagnosis is based on self-reporting. Therefore, we cannot rule out that some people ‘newly’ reporting PD in wave 6 forgot to report their PD diagnosis in wave 5 (recall bias). For this reason, the longitudinal analysis has to be interpreted with caution. Thus, we cannot make statements regarding their influence as potential cofactors. On the other hand, there is already a large body of literature investigating the impact of PD-related markers on QoL47 while neglecting social determinants. We hope our work will help close this gap and provide the basis for more comprehensive studies on QoL in people with PD. We used IADL as a proxy for motor function because earlier studies have shown that IADL contributes to QoL in PD.47 53 However, we cannot fully address the causal relationships between health problems and deprivation, as both influence each other in a bidirectional manner. For example, whether social isolation can accelerate cognitive decline in PD is largely an open question and needs further investigation.54 In this context, further longitudinal studies are needed. At best, these studies should also consider the impact of social and economic circumstances of PwPD on different countries.

Conclusion

Our results suggest that social deprivation and the risk of social exclusion have a large influence on QoL in PwPD. Accordingly, QoL research in PD should not only focus on PD-related biomedical measures but also acknowledge the impact of social determinants to design policies and interventions aimed at improving well-being. Moreover, future longitudinal studies should analyse whether social deprivation can be regarded as an additional ‘social’́ prodromal marker for early PD. This might be partly explained in particular by the presence of depressive symptoms prior to the disease diagnosis. However, social deprivation cannot be explained solely by depressive symptoms, as many other factors contribute to this multidimensional construct.

Data availability statement

Data are available upon reasonable request. Restrictions apply to the availability of these data. Data was obtained from the Survey of Health, Ageing and Retirement in Europe and are available at http://www.share-project.org after successful application.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants. The Survey of Health, Ageing and Retirement in Europe (SHARE) data collection procedures are subject to continuous ethics review by responsible ethics committees (University of Mannheim and Max Planck Society, Germany), as well as national ethics committees in participating countries. The reviews refer to all aspects of the project, from the study design to informed consent. The reviews confirm that the project conforms to international ethical standards, such as the Respect Code of Practice for Socio-Economic Research and the ‘Declaration of Helsinki’. For the present study, appropriate permission was obtained for using the SHARE data after successful application. Participants gave informed consent to participate in the study before taking part.

Acknowledgments

This paper used data from Survey of Health, Ageing and Retirement in Europe (SHARE) waves 5 and 6 (DOIs: 10.6103/SHARE.w5.800, 10.6103/SHARE.w6.800). The European Commission has funded the SHARE data collection, DG RTD through FP5 (QLK6-CT-2001-00360), FP6 (SHARE-I3: RII-CT-2006-062193, COMPARE: CIT5-CT-2005-028857, SHARELIFE: CIT4-CT-2006-028812), FP7 (SHARE-PREP: GA N°211909, SHARE-LEAP: GA N°227822, SHARE M4: GA N°261982, DASISH: GA N°283646) and Horizon 2020 (SHARE-DEV3: GA N°676536, SHARE-COHESION: GA N°870628, SERISS: GA N°654221, SSHOC: GA N°823782, SHARE-COVID19: GA N°101015924) and by DG Employment, Social Affairs & Inclusion through VS 2015/0195, VS 2016/0135, VS 2018/0285, VS 2019/0332 and VS 2020/0313. Additional funding from the German Ministry of Education and Research, the Max Planck Society for the Advancement of Science, the US National Institute on Aging (U01_AG09740-13S2, P01_AG005842, P01_AG08291, P30_AG12815, R21_AG025169, Y1-AG-4553-01, IAG_BSR06-11, OGHA_04-064, HHSN271201300071C, RAG052527A) and from various national funding sources is gratefully acknowledged (see www.share-project.org).

This post was originally published on https://bmjopen.bmj.com