Relationship of burden with depressive symptoms, mental disorders and older adults' functional dependence: A study with paid and unpaid caregivers in Brazil.

The objective of this study was to identify the relationship of burden with depressive symptoms, mental disorders and older adults' functional dependence in paid and unpaid caregivers. This is a quantitative and cross‐sectional study, carried out in the inland of the state of São Paulo with 111 caregivers, 60 of whom were unpaid and 51 were paid. The caregivers answered a questionnaire for the assessment of sociodemographic and care characteristics, the Zarit Burden Inventory (ZBI), the Patient Health Questionnaire (PHQ‐9), the Self‐Report Questionnaire (SRQ‐20), and the Functional Assessment Questionnaire (FAQ). Unpaid caregivers were older, married, have been providing care for a longer period of time, and devoted more days and hours per week to care when compared to paid caregivers. Depressive symptoms and mental disorders (non‐psychotic) are related to the levels of burden in both paid and unpaid caregivers (p < 0.001). The care process causes harm such as burden, depressive symptoms, and mental disorders in paid and unpaid caregivers. Thus, it is important to think about coping strategies to minimize the psychological consequences of the care process.

Keywords: aging; burden; caregivers; depressive symptoms; mental disorders; mental health

• Unpaid caregivers present more burden when compared to paid caregivers.
• Older adults' dependence in carrying out activities of daily living contributes negatively to the caregivers' mental health.
• Most caregivers are female.
• Burden was identified in paid and unpaid caregivers and, in both groups, it was associated with depressive symptoms and mental disorders (non‐psychotic).
• Burden was not related to the degree of older adult's dependence for either paid or unpaid caregivers.
• Unpaid caregivers have higher schooling levels.

What is known about this topic What this paper adds

Population aging is a worldwide phenomenon. With the growing number of older adults and an increase in life expectancy, there is also an increase in the prevalence of chronic non‐communicable diseases, including systemic arterial hypertension, stroke, diabetes mellitus, osteoarthritis, and neuropsychiatric diseases such as dementia (World Health Organization, 2012). Thus, it is observed that the increase in the population's life expectancy is accompanied by specific demands for care, in many cases requiring the presence of a caregiver.

Different definitions for individuals devoted to the care of older adults are identified in the literature; however, we present the one proposed by Stall et al. (2019), which characterizes caregivers as paid and unpaid. Paid caregivers are professionals with higher education or who attended a caregiver course, while unpaid caregivers, in turn, provide non‐professional care without remuneration, in most cases being family members, although they can also be friends or neighbors (Stall et al., 2019). Performing the role of caregiver, whether paid or unpaid, can have several consequences. A number of studies show that care implies changes in the caregiver's daily routine, which, in addition to other responsibilities such as work and family, can compromise their physical and psychological health. The burden is one of the main adverse conditions related to the role of caregiver of older adults (Diniz et al., 2018; Pereira & Almeida, 1999).

The burden construct is multidimensional and has several definitions. In the definition adopted in this study, burden is conceptualized as a quantitative or qualitative demand which goes beyond the capabilities of the individual who provides care (Platt, 1985). It can also be characterized as objective and subjective (Schene, 1990). Objective burden refers to the behavior and symptoms that can be observed in everyday caregiving, and subjective burden refers to the negative consequences of caregiving that are perceived by the caregiver, usually of a psychological nature. These harms can generate social, economic, and health consequences (Schulz et al., 2016; Zarit, Todd, & Zarit, 1986), which may imply repercussions in the health assistance provided to the person cared for.

Although many studies have assessed burden in unpaid caregivers, this variable has also been studied in paid caregivers (Kalanlar & Kuru Alici, 2020), since the instruments most frequently used to assess burden, such as the Zarit Burden Inventory (ZBI), consist of questions that contemplate the care context as a whole, not only that of unpaid caregivers but also that of paid caregivers (e.g., Do you feel that the older adult asks for more help than he/she needs?). Comparative studies conducted with paid and unpaid caregivers of older adults identified the presence of burden in both groups of caregivers, and these studies used the ZBI (Diniz et al., 2018; Maximiano‐Barreto, Moretti Luchesi, et al., 2021).

In addition to burden, depressive symptoms and mental disorders, anxiety, stress, tension, sleep disorders, impaired quality of life and occupational and social functioning can be cited as negative consequences of care, as well as social isolation and difficulty performing roles inside and outside the home environment (Ishii et al., 2012; Miyashita et al., 2011).

Studies that seek to relate the harms to the mental health of paid and unpaid caregivers are identified in the literature. However, there is a limitation regarding studies that investigate the relationship between burden and the psychological impacts on both paid and unpaid caregivers who provide care in family homes. Investigating the caregiver's profile and the impact resulting from care can contribute to the planning and implementation of support actions aimed at this population. Thus, our objective in this study was to assess the burden in paid and unpaid caregivers of older adults, as well as to analyze its relationship with depressive and psychiatric symptoms and with the dependence of the older adults cared for.

This is an analytical and cross‐sectional study, with non‐probabilistic sampling, carried out with 111 caregivers of older adults, 60 of whom were unpaid and 51 were paid. The caregivers were recruited from health units, courses for caregivers of older adults, and social networks from the Federal University of São Carlos (public calls) in a city from the inland of the state of São Paulo.

The eligibility criteria for participation in the research were as follows: caregivers who provided care in a family home for at least four hours a day and aged 18 years old or over. Paid caregivers should have specific training (higher education in Nursing, nursing technician and/or caregiver course). Only caregivers who were primarily responsible for the older adults were included. Caregivers who were devoted to care for less than three months were excluded.

The study followed the recommendations set forth in Resolution 466/2012 of the National Health Council and obtained approval from the Committee of Ethics in Research with Human beings (CAAE: 81016717.0.0000.5504, opinion No. 2.498.312) of Federal University of São Carlos. All caregivers signed an informed consent form.

Sociodemographic and care questionnaire: developed and administered by the researchers, it includes data regarding age, gender, schooling, marital status, time devoted to care ‐ in years, paid or unpaid caregiver, and hours per day/days in the week devoted to care.

The Zarit Burden Inventory (ZBI) is an instrument whose objective is to assess the level of burden in caregivers of older adults. It consists of 22 items that assess the caregiver's psychological, physical, economic, and social aspects, evaluating objective and subjective burden. It is a Likert‐type instrument with scores ranging from 0 to 88, where the higher the score, the greater the level of burden. Its translated and adapted version for the Brazilian context was used (Scazufca, 2002).

The Patient Health Questionnaire (PHQ‐9) aims at screening the presence of depressive symptoms according to the criteria set forth in the Diagnostic and Statistical Manual of Mental Disorders (DSM). It consists of nine Likert‐type items with answers ranging from 0 (no days) to 3 (almost every day) with a total of 27 points, where the higher the score, the greater the presence of depressive symptoms (Chagas et al., 2013; Spitzer et al. 1999).

The Self‐Report Questionnaire (SRQ‐20) is an instrument whose objective is to screen mental distress. This questionnaire consists of 20 questions with binary answers (yes/no) capable of screening mental disorders (non‐psychotic). The score varies between 0 and 20 and the higher it is, the greater the presence of mental disorders. For this study, its translated and adapted version for the Brazilian context was used (Harding et al., 1980; Mari & Williams, 1986).

The Functional Assessment Questionnaire (FAQ) was developed to assess the older adults' functional capacity regarding instrumental activities of daily living. Consisting of 11 items with Likert‐type answers, the FAQ has a score ranging from 0 to 33, with higher scores determining a higher level of functional disability in older adults. Its translated and validated version for the Brazilian context was used (Dutra et al., 2015; Pfeffer et al., 1982).

Through the caregivers' contact information (email address/telephone number) and interest in participating in the research, interviews were scheduled according to the participants' availability regarding dates and times. Filling‐in of the instruments took place in the homes of the older adults cared for by the caregivers or in an appropriate space for the research at the University. The meeting to fill in the instruments lasted approximately 40 min. Data collection took place from March 2018 to January 2019.

The data were tabulated and analyzed using the Statistical Package for the Social Sciences software (version 23.0). The sociodemographic and clinical data were presented descriptively as percentages, mean values (x̅), and standard deviations (σ). An analysis was performed regarding the normality of the variables using the Kolmogorov‐Smirnov test. The Mann‐Whitney and Chi‐square (χ²) tests were performed for comparisons between the groups. Spearman's Rho Correlation Coefficient was used to analyze the correlation between the numerical variables. To analyze the classification of the magnitude of Pearson's correlation coefficient (r) values, the following values were used (weak: from 0.10 to 0.39; moderate: from 0.40 to 0.69; and strong: from 0.70 to 1), either positive or negative (Dancey & Reidy, 2006). The significance level adopted for all analyses was p ≤0.05.

Table 1 presents the sociodemographic and care characterization of the sample interviewed. The participants in this study were mostly female (91%), with a mean age of 45.36 years old (±14.53). Statistically significant differences among the caregivers were found in relation to age (p = 0.001), marital status (p = 0.038), living with the older adult (p = 0.001), time devoted to care in years (p = 0.027), days in the week devoted to care (p = 0.001) and daily hours devoted to care (p = 0.001). The unpaid caregivers were older, with a higher percentage of married individuals and living with the older adult, had been caregivers for a longer period of time, and devoted more days in the week and more daily hours to care.

1 TABLESociodemographic and care characterization of paid and unpaid caregivers of older adults

The clinical characterization of the caregivers of older adults is presented in Table 2. Statistically significant differences between the groups were identified in all the variables analyzed. The results point to greater burden (28.11 [±13.25]), more depressive symptoms (8.55 [±6.56]) and psychiatric symptoms (7.56 [±4.42]) among unpaid caregivers when compared to paid caregivers (burden: 17.13 [±10.17]; depressive symptoms: 5.15 [±4.83]; and mental disorders: 4.15 [±3.51]). However, functional disability in the older adults cared for by the unpaid caregivers is lower (23.30 [±9.12]), when compared to disability in the aged individuals cared for by paid caregivers: 27.84 (±8.96). According to the data, we sought to assess the relationship of burden with depressive symptoms, mental disorders, and functional dependence of the person cared for in the two groups of caregivers, separately.

2 TABLEClinical characterization of the groups of paid and unpaid caregivers of older adults

1 Abbreviations

2 X̅, Mean; σ, Standard Deviation; FAQ, Functional Assessment Questionnaire; PHQ‐9, Patient Health Questionnaire; SRQ‐20, Self‐Report Questionnaire; ZBI, Zarit Burden Inventory.

Figure 1(a–c) presents the relationship of burden with depressive symptoms, mental disorders, and older adults' functional capacity among unpaid caregivers. A moderate and positive correlation was identified between the unpaid caregivers' burden and the presence of depressive symptoms (r = 0.557) and mental disorders (r = 0.619). The relationship between burden and functional dependence of the person cared for was not significant.

In the group of paid caregivers, the results are similar, with the burden being positively correlated with more depressive symptoms (r = 0.573) and mental disorders (r = 0.610). Regarding functional dependence of the person cared for, no relationship was identified. These results are presented in Figure 2(a–c).

Our results showed that unpaid caregivers present higher levels of burden, depressive symptoms and mental disorders than paid caregivers, even with greater functional dependence of the older adults cared for by paid caregivers. The burden was related to depressive symptoms and mental disorders in paid and unpaid caregivers.

The caregivers participating in this research present the following characteristics: they are mostly female, married, young adults, and live with the person cared for. We identified similarities with other studies carried out with caregivers of older adults regarding the high proportion of females (Maximiano‐Barreto, Bomfim, et al., 2021; Maximiano‐Barreto, Moretti Luchesi, et al., 2021; Principi et al., 2014; Tana et al., 2019), age around 45–50 years old (Anjos et al., 2015; Araújo et al., 2013; Maximiano‐Barreto, Moretti Luchesi, et al., 2021), married individuals (Gratão et al., 2013; Martins et al., 2019), living with the older adult cared for (Amendola et al., 2011; Anjos et al., 2015), time devoted to care between four and five years (Souza et al., 2015), and providing care almost every day of the week and for more than 12 hr a day (Diniz et al., 2018). Regarding schooling, the unpaid caregivers in this study presented higher levels of education when compared to other studies which showed that unpaid caregivers have lower schooling levels (Amendola et al., 2011; Gonçalves et al., 2011; Guedea et al., 2009). Possibly, what can justify the high schooling levels in unpaid caregivers is that the municipality where the study was developed is home to two public universities and that many of these caregivers may be linked to them.

The clinical variables assessed in the paid and unpaid caregivers participating in our study were burden, depressive and neuropsychiatric symptoms, and dependence of the older adult cared for. Our results are in line with other studies that indicate high levels of burden (Diniz et al., 2018; López‐Martínez et al. 2017; Pereira et al., 2013) and mental disorders in unpaid caregivers, when compared to paid caregivers (Diniz et al., 2018; Henriques et al., 2018). In addition, there is greater dependence on instrumental activities of daily living in the older adults who are cared for by paid caregivers (Martins et al., 2019).

We evaluated the relationship between burden and depressive symptoms in the caregivers and identified that higher levels of burden are positively correlated with depressive symptoms in both paid and unpaid caregivers. This result is compatible with other studies conducted with unpaid caregivers (Hu et al., 2018; Sun et al., 2019). Regarding paid caregivers, a meta‐analysis identified that there is a relationship between Burnout and depression (Koutsimani et al., 2019). Burden and Burnout are two overlapping constructs, as both are related to the impact of daily work, which can lead to these psychological harms in the individual (Lamothe et al., 2014; Pavarini et al., 2017). These results are also identified in another meta‐analysis carried out with nurses (Molina‐Praena et al., 2018), which can possibly justify this relationship identified in the current research.

Another variable that was positively correlated with the burden in paid and unpaid caregivers was mental disorders. Gratão et al. (2013) conducted a survey with 124 unpaid caregivers of older adults and identified a positive relationship between these variables. Regarding paid caregivers, a study carried out with 50 caregivers, 35 of whom were unpaid and 15 were paid, did not identify any correlation between these variables (Diniz et al., 2018). Another research study with the participation of 130 nurses, nursing technicians, and nursing assistants identified an association between Burnout syndrome and psychiatric symptoms (Silva et al., 2015). Psychological harm in both groups can be justified by the engagement in care, as well as by the unfolding of activities parallel to care (Aires et al., 2017; Zarit & Zarit, 1987). Another important point is that, in the case of paid caregivers, this relationship with psychological harm can be increased considering care aimed at more than one aged person.

Our data did not evidence any relationship between burden in the paid and unpaid caregivers and functional capacity of the older adult cared for. In the opposite direction, a study carried out with 124 caregivers of older adults with the objective of identifying the older adults' functional capacity and the caregivers' burden did show a relationship between those variables (Gratão et al., 2013). Consonant results were identified in other studies carried out with caregivers of older adults (Fuhrmann et al., 2015; Nardi et al., 2013; Nunes et al., 2019). This relationship between burden and older adults' dependence is confirmed in an integrative review (Baptista et al., 2012).

Some points can be emphasized to justify the absence of a relationship between burden and functional dependence of the individuals cared for. The paid caregivers' burden may not be directly related to the older adults' dependence on activities of daily living, as identified in other studies (Baptista et al., 2012; Martins et al., 2019). It is likely that other variables such as wage, time devoted to care and care ability, among others, may justify the absence of a relationship between the variables. With the unpaid caregivers, the older adults presented less functional dependence; however, the caregivers had higher levels of burden. A study that assessed the relationship between financial situation and burden in unpaid caregivers verified that lower financial support is associated with higher levels of burden (Lai, 2012). In addition to that, more time devoted to the person cared for (Diniz et al., 2018; Souza et al., 2015), as well as lack of social support (Del‐Pino‐Casado et al., 2018), may justify the higher levels of burden among unpaid caregivers.

Given the psychological harms such as burden, depressive symptoms, and mental disorders triggered by the care process (Gratão et al., 2013; Koutsimani et al., 2019; Luchesi et al., 2015), interventions have been carried out to assist caregivers in facing these consequences (Azman et al., 2017; Williams et al., 2014). A number of systematic reviews have identified that interventions such as informative and/or educational lectures, cognitive techniques, physiological adjustment, conversation circles, and psychological support assist caregivers in facing the consequences of care. In addition, there are also strategies from both the healthcare network and family support (Hawken et al., 2018; Santos et al., 2011) to screen, minimize and identify burden, depressive symptoms, and emotional discomfort in this population.

In addition to these activities highlighted above, it is important to emphasize the need to plan public policies aimed at maintaining the caregivers' health and quality of life. These policies can assist in minimizing psychological harms, changes in daily routines, and in the caregivers' daily lives. The unpaid caregivers need to reorganize their entire home and family dynamics to meet the care demands of older adults. It is important to reinforce these caregivers' support network and investigate other factors, in addition to the older adults' dependence, that may contribute to a higher burden. Financial aid policies for unpaid caregivers have also been investigated (Adamek, 1992; Lai, 2012); and this support could possibly minimize the negative consequences of care (Chen et al., 2010). In the case of paid caregivers, it is important to consider their family and social environment, since these can also cause burden, which ends up being added to the paid caregivers' burden. These issues are usually taken into consideration only in the surveys carried out with unpaid caregivers; however, it is important to also think about the paid caregivers, as these factors can be deleterious for the care of older adults.

The limitations of this study include the following: its cross‐sectional design that does not allow identifying cause‐effect in the relationship between burden and psychological harms among caregivers of older adults; the use of ZBI to screen burden in paid caregivers, as it is an instrument aimed at unpaid caregivers; and that selection of the sample was carried out in a non‐probabilistic manner.

Our findings showed that higher levels of burden were correlated with depressive symptoms and mental disorders among paid and unpaid caregivers. There was no relationship between the caregiver's burden and dependence of the older adults under their care. Our results supported the development and implementation of strategies to cope with psychological harms in paid and unpaid caregivers.

This study was financed in part by the Brazilian fostering agencies: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES [Coordination for the Advancement of Higher Education Personnel])—Finance Code 001). DMF is a recipient of a scholarship from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES [Coordination for the Advancement of Higher Education Personnel])—(grant: # 88,887.338752/2019‐00), ABM is a recipient of a scholarship from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq [National Council for Scientific and Technological Development]), and MAMB is a recipient of a scholarship from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP [State of São Paulo Research Assistance Foundation (process: 20/04936‐4)]).

The authors have no conflicts of interest to declare.

Chagas, MHN and Luchesi, BM made substantial contributions to conception and design of the study. Maximiano‐Barreto, MA and Moura, BM has recruited participants and conducted interviews. Maximiano‐Barreto, MA and Fabricio, DM drafted the initial manuscript. Brito, TRP with contributions on theme generation and made manuscript review. All authors have given final approval of the version to be published and agreed to be accountable for all aspects of the work.

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.