Mental health effects of adult children’s outmigration on older parents in Central and Eastern Europe

Abstract

Objective

To examine the association between adult children’s migration and depression among older parents in Central and Eastern Europe (CEE) and explore the role of intergenerational support in contributing to their depression.

Methods

Data are from the eighth wave of the Survey of Health, Ageing and Retirement in Europe (SHARE), pooling a study sample of 11 CEE countries, with a cross-sectional design. Analysis of covariance (ANCOVA) and hierarchical linear regression were conducted using a study sample of 9133 respondents.

Results

Older adults whose children migrated over 500 km were more likely to experience depression compared to those with no migrant child or all children within 500 km. Among intergenerational support, frequent parent-child contact mitigated the effects of migration on depression in older parents with all their children who migrated over 500 km.

Conclusion

This study suggests that older parents with migrant children over 500 km away should be considered a vulnerable population at risk for mental health in CEE countries. It is crucial for local governments and policymakers to address these challenges through improving integrated mental health and social programs for better mental health outcomes among older adults in CEE countries.

Keywords:

• Depression
• migration
• older adults
• intergenerational support
• Central and Eastern Europe

Introduction

Depression in later life is an increasingly pressing public health issue with aging population trends. Among European countries, Central and Eastern European (CEE) countries have a higher prevalence of late-life depression, with a rate of 42% compared to the average European rate of 39%, as revealed by recent studies (Grundy et al., 2019; Hansen et al., 2017; Horackova et al., 2019; Naghavi, 2019). Unlike depression in younger age groups, risk factors leading to the development of late-life depression include chronic diseases, disability, lower socio-economic status, weak social networks, disrupted family relations, and lack of intergenerational support (Blazer, 2003; Fiske et al, 2009). In CEE countries, where filial care norms remain prevalent and community-based mental health care is less available (Gedvilaite-Kordušienė, 2015; Kureková, 2013), adult children are often a crucial source of instrumental and emotional support for older adults (Bengtson & Roberts, 1991; Guo et al., 2009).

CEE countries have also been experiencing significant outmigration over the past two decades (Kahanec & Zimmermann, 2010). As of 2018, per 1000 working-age citizens of Romania and Lithuania had an annual outflow rate of 1.4% and 1.3%, respectively (European Commission, 2020). Outmigration has raised concerns about the potential increase in loneliness, social isolation, and poorer mental health of older parents left behind (Démurger, 2015; Giles et al., 2010; Knodel et al., 2000, 2010; Kreager, 2006). Geographic proximity of adult children to their aging parents is however generally believed to be beneficial for the physical and mental health of older parents. Living closer may allow for providing intergenerational support, such as frequent interaction, emotional closeness, and support exchange (Fors & Lennartsson, 2008; Liang & Zhang, 2017; Offer & Fischer, 2018; Ward et al., 2014).

Although intergenerational support has been identified a protective factor against depression among older adults (Schwarzbach et al., 2014; Wu et al., 2018), its impact varies based on the type of support as well as on diverse cultural backgrounds involved. Therefore, research on the impact of adult children’s outmigration on parental mental health has yielded mixed results. Studies have either reported a negative effect (Antman, 2010, 2013, 2016; Guo et al., 2009; Li, 2020; Lu, 2012; Mosca & Barrett, 2016; Muhammad et al., 2022; Scheffel & Zhang, 2019; Torres et al., 2018), or a positive one (Abas et al., 2009, 2013; Yi et al., 2019), or no significant effect (Böhme et al., 2015; Ghimire et al., 2018; Gibson et al., 2011; Waidler et al., 2017). Studies conducted by Abas et al. (2009, 2013) showed positive mental health effects for parents in rural Thailand because they had received financial support, which led to improved psychological outcomes through better living conditions. Migration has been viewed as an opportunity to alleviate poverty in families, especially in low-resource household or less industrialized rural areas.

Despite relevant literature on CEE countries (Bó et al., 2020; Botev, 2012), most studies have not investigated mental health outcomes of left behind parents (Conkova & King, 2019). Studies that have investigated the intergenerational care between outmigrated children and left behind parents focused on Poland, Lithuania, and Romania, thus creating a knowledge gap on the regional level (Bó et al., 2020; Gedvilaite-Kordušienė, 2015; Krzyżowski & Mucha, 2014; Schröder-Butterfill & Schonheinz, 2019; Zimmer et al., 2014). Therefore, the aim of the current research was to explore whether adult children’s migration significantly impacts mental health, operationalized as depression, of left behind older parents in all CEE countries where data were available. To the best of our knowledge, ours is the first study to investigate this issue in all CEE countries.

We hypothesized that older adults with migrant adult children would experience greater depression than those with no migrant adult children. Secondly, we assessed how intergenerational support, operationalized as financial support given and received, contact frequency, and emotional closeness impacted on older adults’ depression.

Materials and methods

We used data from the eighth wave of the Survey of Health, Ageing and Retirement in Europe (SHARE) conducted in 2019/2020 (Börsch-Supan, 2022; Börsch-Supan et al., 2013). SHARE is a multidisciplinary longitudinal panel survey of people aged 50 or over in 28 European countries and Israel, using nationally representative probability sampling from population registries or multistage sampling methods. Due to the Covid-19 pandemic, about 70% of all interviews in the 8^th wave were conducted using the face-to-face computer-assisted personal interviewing (CAPI) technique, combined with computer-assisted telephone interviewing (CATI) technique, for the health protection of study participants (Scherpenzeel et al., 2020). The SHARE study remains under constant scrutiny through its ethical review process. For Waves 1 to 4, the Ethics Committee of the University of Mannheim provided its review and approval. Subsequently, Wave 4 and the project’s ongoing phases received review and approval from the Ethics Council of the Max Planck Society. This study, conducted as a secondary analysis, had received ethical approval from the University of Pécs.

As per the OECD statistical definition (OECD, 2022), CEE countries include Bulgaria, Czech Republic, Croatia, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, and Slovenia. The overall sample from the 8^th wave comprised 46500 respondents. Because our study focused on 11 CEE countries, another 19 European countries, which were not part of the CEE cluster in the SHARE data, were excluded (n=28893). Out of the remaining 17607 respondents, we excluded those who were younger than 50 (n=120) because spouses of respondents aged 50 or more were interviewed and some of them were younger than 50. To address our research questions, we also excluded those older adults who had no child (n=5655), i.e. those without at least one living biological, adopted, or stepchild.

Additionally, we excluded those with their child younger than 18 years (n=89), and those reporting a change of child location in Wave 8 (n=1270) to avoid potential bias and error in measuring duration of exposure, as well as individuals with missing data in the key variable of adult children’s migration status (n=1340). In this study, left-behind parents were defined as older adults who had one or more adult children living at a distance from their household for at least the past two years. The analytical sample size was 9133 respondents.

Depression

To assess the mental health of older adults in CEE countries, the EURO-D instrument was used in the SHARE study, developed for comparing the prevalence of major depressive symptoms in later life across different European countries (Mehrbrodt et al., 2021). The instrument’s internal validity and cross-cultural validity have been established in prior studies (Castro-Costa et al., 2008; Prince et al., 1999). Respondents were asked whether they experienced pessimism, depressive mood, suicidality, guilt, sleep difficulty, interest, irritability, fatigue, concentration, enjoyment, appetite, and tearfulness during the previous month in the local language. The answers of respondents were coded as 0=’no’ or 1=’yes’ and a sum score was calculated based on answers to each item. Depressive symptoms in older adults were measured using a composite score based on 12 items, with a total score ranging from 0 to 12. A higher score suggests a more severe degree of depression. Cronbach’s alpha score in this study was 0.72.

Adult child out-migration status

Adult child out-migration status was the main measure of this study. To align with previous studies (Abas et al., 2013, 2009), we recategorized the original responses into four new categories: 1) no migrant child (all children in the same household as the parents), 2) all children within 500 km of the parental residence, 3) some children over 500 km away from the parental residence, 4) all children over 500 km away from the parental residence. Using the furthest measured proximity, we operationalized migration over 500 km as long-distance migration.

Intergenerational support

Several intergenerational support variables were used to examine the impact of adult child outmigration on older parents’ depression, including contact frequency with children, emotional closeness with children, financial support received from children, and financial support given to children adopted from previous studies (Abas et al., 2009; Aichberger et al., 2010; Lu, 2012; Thapa et al., 2018). Contact frequency with children was assessed with the categories of 0=’never’, 1=’less than once a month’, 2=’once a month’, 3=’every 2 week’, 4=’once a week’, 5=’several times a week’, and 6=’daily contact’. A mean score of contact frequency with children was calculated based on answers to each item on each child. Emotional closeness with children was measured by answering the question about how close they feel to their children. Response categories were 1=’don’t know’, 2=’somewhat close’, 3=’very close’, and 4=’extremely close’. A mean score of emotional closeness with children was calculated based on answers to each item on each child. To assess financial support received from children and given to children, respondents were asked whether they had received/given any financial or material gift or support amounting to €250 (e.g. receiving/giving money, covering specific types of costs such as those for medical care, insurance, or down payment for a home) from/to their children during the last 12 months. Response categories were 0=’no’ and 1=’yes’ and a mean score for the financial support received from children and given to children respectively were calculated based on answers to individual items.

Covariates

Sociodemographic, physical and cognitive health, functional disability, and social network variables were included as covariates, as they were known to affect depressive symptoms in older adults (Abas et al., 2013; Aichberger et al., 2010; Ghimire et al., 2018; Lu, 2012; Thapa et al., 2018).

Sociodemographic variables

Sociodemographic variables included in this study were age, gender, education, relationship status, employment status, economic status, residence area, and number of children. The ages of both parents and children were recorded by subtracting their birth year from the year 2020. Each parent’s gender was coded as 1=’male’ or 2=’female’. Education level was classified into 1=’lower’ (levels 0–3) or 2=’higher’ (level 4–6), based on the International Standard Classification of Education (UNESCO, 2006) (ISCED) version 97, which ranges from pre-primary education (level 0) to the second stage of tertiary education (level 6). Relationship status was categorized as either 1=’no partner’ or 2=’with partner’. Employment status was self-reported and classified as either 1=’not working’ or 2=’working’. Economic status was assessed by a question related to the total monthly income of the household. Respondents were asked if their household was able to make ends meet, and responses were recoded as either 1=’easy’ or 2=’difficult’. These categories were reclassified from the original four options: ‘easy—fairly easy—with some difficulty—with great difficulty’. Residence area was derived from the interviewer module and recategorized as 1=’rural’ (small town, rural area, or village) or 2=’urban’ (big city, suburbs, or outskirts of a big city and large town). Number of living children was recorded up to 20.

Somatic comorbidities, functional disability, and physical inactivity

The number of chronic illnesses (e.g. heart disease, stroke, hypertension, diabetes, high blood sugar, cancer, and lung disease) was recorded from 0 to 14. Functional disability was assessed using the modified version that counted the number of limitations in activities of daily living (ADL) and complex instrumental activities of daily living (IADL) (Steel et al., 2003). For the modified ADL, these assessments encompassed six daily activities—including tasks such as bathing, dressing, walking, eating, transitioning to/from bed, and using the toilet. These activities were coded from 0 to 6 as a continuous variable. Similarly, the modified IADL comprised nine institutional activities, such as taking medications, grocery shopping, preparing meals, using the telephone, using maps, household tasks, managing finances, leaving the house, and laundry. These activities were coded from 0 to 9 as a continuous variable. In both measures, higher scores corresponded to increased challenges with these activities (Mehrbrodt et al., 2021). Physical inactivity was assessed by asking respondents how often they engaged in activities that required moderate levels of energy, such as gardening, cleaning the car, or walking. Responses were coded as 0 for active (more than once a week, once a week, and one to three times a month) and 1 for inactive (hardly ever or never) (Malter & Börsch-Supan, 2013).

Cognitive function

The ten words recall test and the semantic verbal fluency test were employed to assess objective cognitive performance. In the ten words recall test, respondents were asked to remember and retrieve any word from a list of 10 words that had been read out approximately 5 minutes earlier. The score ranged from 0 to 10. In the semantic verbal fluency test, respondents were given 60 seconds to generate as many animal names as possible. The total count of accurate words was recorded (Mehrbrodt et al., 2021).

Social network size and satisfaction

Social network size was assessed based on respondents’ report on the number of individuals in their network with a scale ranging from 0 to 7 (Malter & Börsch-Supan, 2013). They were also asked to rate their satisfaction with their network on a scale of 0 (completely dissatisfied) to 10 (completely satisfied).

Statistical analysis

Descriptive statistics including frequencies, proportions, and means (±SD) were done for main measures. To minimize the effect of selective nonresponse and panel attrition, we applied calibrated, cross-sectional individual level weights provided by SHARE. Analysis of covariance (ANCOVA) was applied to assess differences in depression across all four child migration categories. Socio-demographic variables such as age, gender, education, relationship, employment, economic status, residence area as well as somatic comorbidities, functional disability, physical inactivity, social network size and satisfaction, were included as covariates. Post hoc differences across child migration categories were assessed by using the Least Significant Difference (LSD) post-hoc test to compare differences between pairs of groups in the child migration status variable. To predict depression, hierarchical linear regression was used. Outliers, by means of Studentized residuals, were removed from the regression analysis. Respondents falling outside of the <2 and >−2 range were deselected. To understand and compare the unique contribution of non-sociodemographic predictors, we also developed four regression models separately for each child migration group. Statistical analyses were performed using SPSS Statistics, Windows version 25. Tests used for hypothesis testing were one-tailed, with a significance level of 0.05.

Table 1. Unweighted sample characteristics.

Variable	Total sample (n = 9133)
Age, mean (SD)	71.00 (9.71)
Gender, female, %	69.37
Education, higher, %	24.40
Employment, working, %	26.25
Relationship, no partner, %	43.80
Ability to make ends meet, with difficulty, %	53.38
Residence, rural, %	63.08
Number of children, mean (SD)	2.16 (1.09)
Number of chronic diseases, mean (SD)	2.09 (1.70)
Verbal fluency, mean (SD)	19.48 (7.71)
Delayed recall, mean (SD)	3.59 (2.11)
Limited daily activities (ADL), mean (SD)	0.32 (0.99)
Disability in instrumental daily activities (IADL), mean (SD)	0.69 (1.72)
Physical inactivity, inactive, %	15.56
Social network size, mean (SD)	2.48 (1.44)
Social network satisfaction, mean (SD)	9.03 (1.42)
Financial support received from children, mean (SD)	0.05 (0.19)
Financial support given to children, mean (SD)	0.09 (0.26)
Contact frequency, mean (SD)	1.73 (0.90)
Emotional closeness, mean (SD)	3.48 (0.59)
Euro-D, mean (SD)	2.78 (2.42)
Child migration status, No migrant child, %	11.30
All in locality, %	71.98
Some children over 500 km away, %	12.20
All children over 500 km away, %	4.52

Results

A total of 9133 older adults were eligible for the study (mean age 71.00 years; 68.37% female). Out of the total, 24.40% had a higher education level, 43.80% had no partner, 73.75% were not working or retired, 53.38% had difficulty making ends meet and 63.08% lived in a rural area. The average number of chronic diseases, limited daily activities (ADL), and limited instrumental activities of daily living (IADL) were 2.09 (SD=1.70), 0.32 (SD=0.99), and 0.69 (SD=1.72), respectively. The EURO-D mean score was 2.78 (SD=2.42). Of parents in our study, 11.30% had no migrant child, 71.98% had all children residing within 500 km, 12.20% had some children located more than 500 km away, and 4.52% had all children living more than 500 km away (Table 1).

Results of the ANCOVA analysis (Table 2) revealed significant differences in parents’ levels of depression across child migration categories (F=8.80, p<0.001) after controlling for all covariates. Pairwise comparisons showed older parents having some or all their children over 500 km away had significantly higher depressive symptoms than those with no migrant child (p=0.004) or all their children within locality (p=0.003).

Table 2. Association between child migration status and depression (n = 9133).

Child migration status	Mean $\pm$ SD	Depression F test	Pairwise comparisons
No migrant child (1)	2.79 $\pm$ 0.08	8.802***	1 < 3***, 1 < 4***
All in locality (2)	2.93 $\pm$ 0.04		2 < 3***, 2 < 4**
Some over 500 km (3)	3.27 $\pm$ 0.08		3 > 1***, 3 > 2***
All over 500 km (4)	3.38 $\pm$ 0.16		4 > 1***, 4 > 2**

Notes: age, gender, education, relationship, employment, economic hardship, residence, number of children, number of chronic diseases, ADLs, IADLs, verbal fluency, delayed recall words, physical inactivity, social network size, and satisfaction on social network quality are used as covariates. LSD post-hoc test was conducted for pairwise comparisons. Data weighted.

** p < 0.01.

*** p < 0.001.

Table 3 shows multivariate stepwise linear regression results for intergenerational support variables on depression among older parents. We presented results of a hierarchical linear regression using the entire sample of 9133 after adjusting for sociodemographic and other covariates. The full model was significant (F=14.46, p<0.001) and explained 29.4% of the variance in parental depression. When entered together in the last step, the unique contribution of migration status variables to explaining the variance in depression has been revealed. Migration variables accounted for 3% of the variance in depression. Of the intergeneration support variables, our main measures of interest, we found that financial support provided to adult children (b=0.30, p<0.05), contact frequency (b=-0.10, p<0.05), and emotional closeness (b=-0.17, p<0.05) significantly predicted depression in older adults.

Individual multivariate linear regression models (Table 4) developed for each child migration status showed the unique impact of intergenerational support within each group. All models were significant and explained 29, 33, 34 and 54 per cent of the variance in parental depression. For older parents with no migrant children, greater levels of emotional closeness with their children decreased depressive symptoms by about half a point (b=-0.44, p<0.05). However, receiving more financial support from their children increased parents’ depression by over one and a half points (b=1.64, p<0.05). Increased contact frequency, another significant predictor of depression, decreased parental depression when all children lived locally (b=-0.14, p<0.05) or were all more than 500 km away (b=-0.46, p<0.05). In contrast, greater contact frequency increased depression in older parents with some children living over 500 km away (b=0.36, p<0.01).

Table 3. Multivariate, stepwise hierarchical regression for older adults’ depression.

Variable	b (unstandardized) weight	p value
Step 1
Intercept	3.82	< 0.001
Age	−0.01	0.32
Gender	0.73	< 0.001
Education	−0.15	0.17
Relationship	−0.22	0.01
Employment	0.27	0.01
Economic hardship	0.57	< 0.001
Rural-urban residence	−0.02	0.82
Number of children	0.04	0.24
Number of chronic diseases	0.28	< 0.001
ADLs^a	0.13	0.01
IADLs^b	0.33	< 0.001
Verbal fluency	−0.01	0.12
Delayed recall words	−0.22	< 0.001
Physical inactivity	0.11	0.26
Social network size	−0.10	< 0.001
Social network satisfaction	−0.21	< 0.001
Step 2
Financial support received from children	0.28	0.19
Financial support given to children	0.30	0.04
Contact frequency	−0.10	0.03
Emotional closeness	−0.17	0.01
Step 3
All in household (ref.)
All within 500 km	0.15	0.20
Some over 500 km	0.48	0.00
All over 500 km	0.59	0.01

Note: Dependent variable: older parent’s depression. Data weighted. ^a= Activities of Daily Living, ^b = Instrumental Activities of Daily Living. Adjusted R square: 0.294.

Table 4. Individual regression models for child migration categories predicting parent depression.

Category	No migrant child		All within 500 km		Some over 500 km		All over 500 km
Model	R^2 = 0.34, p < 0.001		R^2 = 0.29, p < 0.001		R^2 = 0.33, p < 0.001		R^2 = 0.54, p < 0.001
Variables	b weight*	p	b weight	p	b weight	p	b weight	p
Financial support received from children	1.64	0.03	0.53	0.06	0.26	0.62	−0.14	0.80
Financial support given to children	0.83	0.06	0.10	0.58	0.72	0.09	0.01	1.00
Contact frequency	−0.55	0.80	−0.14	0.01	0.36	0.00	−0.46	0.02
Emotional closeness	−0.44	0.02	−0.12	0.16	−0.20	0.23	−0.61	0.10

Note: Dependent variable: older parent depression. Data weighted.

* Unstandardized regression coefficient.

Discussion

Our study aimed to examine the effects of adult children’s out-migration on the mental health of older adults in CEE countries, and we sought to explore what role intergenerational support has in promoting positive mental health outcomes. This study is among the first to document the potential negative mental health consequences of long-distance adult children’s out-migration in CEE countries.

We found that older parents whose adult children migrated over 500 km from their households were at a higher risk for depression, compared to those without a migrant child. However, there was no significant difference in depression levels between those with all children livingwithin 500 km and those with no migrant child, suggesting that short-distance migration does not negatively impact the mental health of older parents. This could be due to the ease of maintaining regular physical contacts and providing timely support for adult children living within 500 km. Despite the free mobility policy enforced within the EU, irregular physical contacts between children and parents increased the negative impact of long-distance migration (beyond 500 km) on depression and mental health of older adults. Overall, these findings highlight the impact of outmigration has on mental health of older parents in CEE countries and underscore the importance of intergenerational support.

As for intergenerational support, our study revealed that parent-child contact frequency and emotional closeness both had a significant impact on reducing depression among older adults. On the other hand, providing financial support to adult children was found to significantly increase depression, while receiving financial support from adult children had no significant effect on their parents’ depression. This pattern aligns with prior studies highlighting the greater impact of psychological support in improving the mental health of older adults when compared to material or instrumental support (Gur-Yaish et al., 2013; Merz & Huxhold, 2010).

Out of the four regression models predicting depression by child migration status, we identified two distinct effects of financial support received from children and emotional closeness for the parents without migrant children. Within this group, emotional closeness with their children significantly reduced the risk of depression by bolstering their feelings of self-efficacy, fostering a sense of intimacy and trust with their adult children (Lin & Chen, 2018). Conversely, financial support received from their children increased the risk of depression for parents without migrant children. Although the causal link between financial support received and depression is not evident, some studies have reported that providing financial support to parents can evoke feelings of burden, leading to excessive guilt and shame among older parents (Shiraz et al., 2020; Silverstein et al., 2013).

For those with children living separately, frequency of parent-child contact emerged as the main predictor of depression. These findings align with previous studies that have highlighted the positive impacts of close contact with adult children (Buber & Engelhardt, 2008; Lawton et al., 1994). Tosi and Grundy (2019) argued that maintaining close contact with adult children can enhance the psychological security of older parents, particularly in countries experiencing societal transformations, rapid changes, and lacking robust public support system as demonstrated in their research in Bulgaria, Georgia, and Russia.

However, contact frequency, when some children resided more than 500 km away, had an opposite effect. Parents who maintained more frequent contact with their children in this category displayed higher levels of depressive symptoms. This finding has not been observed elsewhere in the literature. Future research is recommended to identify underlying explanations for this outcome. We have been looking at different variables and their influence on contact frequency in our sample, but a deeper analysis would be out of scope for this paper.

Caution needs to be exercised when interpreting results, considering that SHARE wave 8 was conducted during the Covid-19 pandemic. The crisis prompted migrant children to be more attentive to their left-behind parents, leading to an increased frequency of virtual contacts and a higher level of support compared to usual circumstances. Furthermore, the pandemic facilitated the return of numerous migrants to their countries of origin. Parents with returned migrants were not included in the present study, potentially resulting in an underestimation of our model predicting depression outcomes for parents.

To improve the mental health outcomes of left-behind parents in CEE countries, public health services may be advised to incorporate assessments that focus on the impact of adult children migration (Kureková, 2013). Migrant-sending countries and communities should allocate more resources towards these programs to meet the mental health needs of left-behind parents. Finally, future research and policy efforts should devote attention to the challenges faced by the burgeoning group of older adults being left behind by adult children’s outmigration.

Limitations

Authors acknowledge limitations in this study. Primarily, our study was based on a secondary data analysis approach and authors therefore had no control over data collection and quality. Additionally, the cross-sectional nature of the study design limited establishing a causal relationship per se between adult child migration and depressive symptoms in older adults. Studies with longitudinal design are needed to assess the impact of adult child migration on older adults’ depression and to establish what contextual factors have a long-term effect on older adults’ mental health.

While the SHARE dataset provided information on intergenerational factors, it did not include details on voluntary or involuntary nature of migration, its causes, or whether it was international or interregional. To correct for these shortcomings, we suggest future studies employ more comprehensive measures including seasonal or short-term circular migration.

Our study also relied on self-reported information from older adults, which could have been biased by cognitive impairment, disability, or respondents’ lack of awareness of their medication use.

Finally, we acknowledge that older adults are a heterogeneous population with diverse characteristics such as age, gender, education, income, and number of children, which may influence the effects of adult child migration on depression. While we employed covariates to control for heterogeneity, these statistical approaches provided only secondary techniques over appropriate sample selection.

Conclusion

Our study has revealed a significant link between adult children’s migration and mental health of older adults in CEE countries. Notably, older adults with children who migrated 500 km away were more prone to experiencing depressive symptoms compared to those without. Furthermore, we also observed effects of intergenerational support associated with child migration status. These findings underscore the critical importance and impact socioeconomic factors have on the mental health of older adults. Considering our results, future research should focus on the mental health burden faced by older adults who are left behind by adult children migration. We also suggest to further explore mechanisms that underlie the potentially protective effects of intergenerational support. Migration will continue to expand in an increasingly globalized world thus making it a significant explanatory variable. Therefore, authors argue that future research investigating causes of older adults’ depression should incorporate measures to assess the influence and unique contribution of child migration on parental mental health in migrant-sending communities.

Acknowledgements

This paper uses data from SHARE Waves 8 (DOI: 10.6103/SHARE.w8.800) see Börsch-Supan et al. (2013) for methodological details. The SHARE data collection has been funded by the European Commission, 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 U.S. 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).

Disclosure statement

The Authors declare that there is no conflict of interest.