Background

Population aging is poised to become one of the most significant social transformations of the 21st century, with implications for every sector of society, including the labor and financial markets, the demand for goods and services such as housing, transportation, social protection, family structures and intergenerational ties [1]. It is estimated that by 2050, there will be 1.5 billion people aged 65 and older worldwide, more than doubling the number of individuals in this age group in the year 2020 [2]. The percentage of older people in the global population is expected to increase from 9.3% in 2020 to 16.0% in 2050, indicating that by the middle of the 21st century, one in six people worldwide will be 65 years of age or older [2].

The World Health Organization [3] conceptualizes elderly people based on age criteria for research purposes. Based on this criterion, an elderly person is one who is aged 60 or over who lives in developing countries, and one who is aged 65 or over who lives in developed countries. Additionally, it is recognized that aging is a continuous, multidimensional and multidirectional process of changes dictated by the concurrent action of the genetic-biological and socio-cultural determinants of the life cycle [4, 5].

As people age, they experience a gradual decrease in physical and mental capacity, and a growing risk of disease and death which, at a biological level, results from the accumulation of molecular and cellular damage over the course of the lifetime [6]. These health conditions can be defined as the circumstances in the health of individuals that require responses from health systems professionals and users [7]. It can generate a disabling process and significantly compromise the quality of life of the elderly.

Beyond biological changes, the current context is geared towards producing a more favorable social and cultural environment for healthy aging and it is the role of public policies to help more people reach old age in the best possible state of health [8]. In order to reach that, aging is often associated with other life transitions, such as retirement. The aspects that determine retirement are interconnected to the individual’s life story, permeated by the combination of identity, family, friendship, work relationships, and professional career [9,10,11,12]. Nevertheless, to achieve this, it is necessary to be part of a social security system responsible for managing the granting and payment of pensions.

In all sorts of retirement, the economic situation of the state and the availability of similar social/welfare benefits can influence its meaning and consequences, since retirement must be thought about and sought after from a young age [13]. Several types of Welfare State regimes represent different responsibilities assumed by the market, the state and the family in the management of social risks and social security [14]. Previous research shows that countries with the most comprehensive Welfare State, such as Denmark, Sweden, and Norway, have better population health outcomes when compared to Neoliberal States such as the United States and the United Kingdom [15, 16].

The discussion about social security policies can be located between the fields of health of the elderly and workers’ health, considering that the experience of this period does not occur in isolation, but is interconnected, among other factors, to their professional trajectory and to the different stages that make up the life cycle. Researchers have continued to show a strong link between older workers, health, planned retirement age [17,18,19], current retirement behaviors [20,21,22], and adjustment and satisfaction with post-retirement life [23,24,25,26]. In this paper, we aimed to capture current evidence in a systematic review to understand how health conditions in the aging process are related to social security reforms.

Methods

Search strategy and selection criteria

The search procedures for the studies took place between September 2021 and March 2022, with the last search being carried out on March 3, 2022. This systematic review aligns with the PRISMA checklist [27, 28] and methods are outlined in detail in a protocol registered a priori on PROSPERO (CRD42021225820). Likewise, a protocol article was published in a peer-reviewed journal [29].

Eligibility was based on the Population, Intervention, Comparison, Outcomes (PICO) framework, with studies included if they met the following criteria: (1) participants who are in the process of transition to retirement or retired; (2) examined retirement guarantees as intervention/exposure which could be pension benefits, health insurance, subsidized assistance and other contributory schemes; (3) outcomes measured by quantitative methods that analyze the association or influence of social security policies on any outcome related to mental or physical health, such as psychological symptoms, mental disorders, illnesses, well-being. (4) original empirical studies published in English, Spanish, French and/or Portuguese, as these were the most common languages in the research, between 1979 and 2022 that examined aging from health conditions related to social security policies. Studies that identified any results associated with mental health and/or physical health, such as psychological symptoms, mental disorders, illnesses, well-being were included. The choice of 1979 to begin the search is due to the change in policies adopted by countries from a Welfare State to a neoliberal structure, marked by the election of Margaret Thatcher in the United Kingdom in May 1979.

Searches using the indexed terms “social security” AND “aging” were conducted across Embase, Web of Science, Scopus, Pubmed, CINAHL, ASSIA (Proquest) and APA PsycNet. Table 1 presents the full search criteria. Two independent reviewers (LT, FU) screened titles and abstracts for eligibility and studies that met criteria on title and abstract, underwent full text review. Using an excel spreadsheet, data from all studies were then independently extracted by the two reviewers (LT, FU), characteristics of the study (year of publication, study location, author); study design (longitudinal study, cross-sectional, case-control, other); sample size; participant characteristics (age, sex, years of education, marital status); method of data collection; method of analysis; instruments (health conditions measurements and retirement measurements) and the main conclusions of the study.

Table 1 Search criteria for electronic databases
Full size table

The PRISMA flowchart in Fig. 1 shows that 8,758 records were found in the databases. 1,336 duplicates were removed by automation tool, leaving 7,422 articles for title and abstract screening. Of these, 72 articles underwent full-text assessment and 17 met eligibility criteria and were included.

Fig. 1
figure 1

PRISMA Flowchart

Full size image

Data analysis

Risk of bias and study quality

Risk of bias and study quality was assessed at study-level using the Newcastle-Ottawa Scale (NOS) for cross-sectional and observational studies [30]. The NOS scale employs a star system by means of a checklist consisting of three criteria: (a) Selection: where the representativeness of the participants is assessed by analyzing the sampling and sample formation processes; (b) Comparability: where the confounding factors adjusted for sample analysis are identified; and (c) Result: where the evaluation and analysis of the results are verified. According to the scoring system, studies are scored in a range from 0 to 10 points and classified as low (10 and 9 points), medium (7 and 8 points), or high (< 7 points) risk of bias. Higher scores represent better quality. Overall, the NOS scale demonstrates good inter-rater and test-retest reliability [31, 32].

Meta-ethnography

A narrative synthesis was performed using the meta-ethnography [33], which helps synthesizing the studies by combining the results found in an interpretive and non-aggregative way, to generate a higher level of analysis that produces a more relevant contribution than the individual findings of each investigation. Categories were created through thematic analysis of the data considering the evidence found in the selected studies.

Initial synthesis involved extraction of each paper findings, key concepts, metaphors and themes to determine how the studies are related to one another, and to develop descriptive codes. The key themes and relationships from the selected studies were tabulated. A translational process was then be undertaken to synthesize the findings using reciprocal analysis to create themes. The final findings were reported in a clear and concise manner to provide readers with a clear understanding of how we arrived at our findings. All stages were undertaken collaboratively by the research team. Data synthesis were independently undertaken by two reviewers (LT, FU); with a third author (JP) used for consensus as appropriate. The eMERGe meta-ethnography reporting guidance was followed [34].

Role of the funding source

The funders had no role in study design, data collection, analysis, interpretation, or writing. The corresponding author had full access to all data and final responsibility for the decision to submit for publication.

Results

Seventeen cohorts of adults and elderly people were analyzed from the following countries: Australia, Austria, Belgium, Canada, China, Czech Republic, Denmark, Estonia, France, Germany, Greece, Hungary, India, Ireland, Italy, Japan, Netherlands, Philippines, Poland, Portugal, Slovenia, South Korea, Spain, Sweden, Switzerland, United Kingdom, and United States. The sample range varied between 80 and 18,345 individuals with an age range of 30 to 87 years. Of the 17 studies, 10 reported cross-sectional data analysis, 1 reported cross-sectional and longitudinal analysis and 5 analyzed longitudinal data. The main characteristics and results of the studies are presented in Table 2.

Table 2 Characteristics of the studies
Full size table

A predominance of studies was carried out in the European continent (73.58%), with the largest number of studies concentrated in Sweden. In the Americas, in turn were 13.20% of the studies concentrated in the United States and Canada, followed by 11.32% in Asia and 1.88% in Oceania. However, there is a lack of research in African regions and in Latin and South American countries. Identified studies evaluated the relationship between health conditions that are more common in older adults, retirement and social security policy reforms - particularly those related to retirement - and were published between 1995 and 2021. Individual study risk of bias assessment is presented in Table 3.

Table 3 Risk of Bias Assessment
Full size table

Health as a way to promote an active working life for the elderly

The perception that individuals have about their health condition and their permanence in the labor market is related. Four studies brought results suggesting this relationship [35,36,37,38]. In all studies, a good perception of health in general scope was found to be a determining factor for remaining in the labor market. Although retirement is an expected event, many older people would consider staying in the labor market for longer if there were better working conditions, such as additional senior citizen days, longer vacations, flexible work hours, and if the work was less physically demanding [38]. Also, unionized workers reported that favoring of prolonging work is not out of sheer necessity, but rather, because the expression of this desire comes from work attachment and professional identification [35]. Retirees who were in excellent health retired from their career jobs, were more likely to take bridge jobs, that bridge the gap between full-time employment and complete withdrawal from the labor force [36]. Workers who reported fair or poor physical health were less likely to remain employed after the ages of 62 and 65, moreover, there was a gradual decline in self-reported health and worsen health conditions over time [37]. According to the data found, a good self-reported health status is a factor that promotes the extension of elderly individuals in the labor market, despite meeting the legal eligibility criteria for retirement.

Health as an indicator for reforms in social security policies

Health conditions were associated with changes in countries’ laws about the eligibility criteria for receiving social security benefits. Four studies explored how health conditions could work as indicators for social security policy reforms [39,40,41,42]. The studies considered the following health conditions: subjective well-being, life satisfaction, and health status and related them to changes in social security of the countries subject to their analysis.

An increase in pension insecurity is associated with a reduction in life satisfaction, and it is a negative and significant relationship. The individuals most affected by pension insecurity are those who are further away from their retirement, have lower incomes, rate their life expectancy as low, have higher cognitive abilities, and do not expect private pension payments. However, while younger cohorts have more time to adapt to new pension systems or accumulate other types of savings, individuals that will retire in the foreseeable future are at risk of needing to work longer or receive lower pensions [40]. In a long term, increasing the age of formal retirement is relatively neutral with regard to subjective well-being, and suggests that later formal retirement simply delays the benefits to be enjoyed at retirement [39]. Employment rates increased in the 50–59 age group with welfare reform, but only among healthy individuals, with the odds ratio for receiving temporary benefits or not being eligible for benefits increasing for people with moderate to severe health problems [41]. Companies that aim to extend working time, where the social environment is more advantageous to their continuation after achieving the legal retirement age, and/or those who do not have experience with age discrimination, adjust more easily to the increase in retirement age. Likewise, employees with poor health have more difficulty adjusting to this augmentation, and better health status is related to fewer negative emotions and thoughts about prolonged employment, but also to increase behavior to facilitate a longer working life [42].

These results indicate that health conditions may be associated with the enhancement in the legal retirement age criterion. A good health condition can help individuals to adapt to the changes generated by the reforms. Also, there is a significant cost to people with poor health and to those who are farthest from retirement, despite presenting a certain neutrality with regard to the positive health of those who are near to retirement when a reform is sanctioned. Thus, when amending criteria to extend time in the labor force to solve fiscal problems, policymakers should analyze the impact on the health of individuals who are forced to postpone retirement, which corroborates its use as an indicator for social security policies, according to the demands of its population.

Retirement planning as a strategic element for coping with post-retirement life

Well-being in retirement is directly related to the attitudes of workers throughout their lives. Four studies looked at the relationship between individuals’ retirement planning during the aging process for their benefit receipt and their health conditions in old age [43,44,45,46].

Social and financial perceptions of post-retirement life were identified as factors that significantly influence retirement planning. On social perceptions, the major components that influence retirement planning detected were depression, role clarity of retired people and social involvement. About financial perceptions, the components identified were financial obligations, government support during retirement, uncertainty from financial perceptions and preparation for post-retirement life [46]. Therefore, contentment and security with participants’ financial situation exert an important factor for retirement preparedness [43, 46]. In this sense, people who actively planned for retirement were much more likely to have a high net worth, personal savings or investment, or a defined contribution plan as their primary source of retirement income, and much less likely to have a low net worth. People who actively planned for retirement were less likely to have the government insurance plan as their primary source. Nevertheless, there was no significant difference between people who actively planned for retirement and people who did not in the percentage of poor health. Most respondents identified their health as excellent or good, except for individuals with no retirement and a low level of wealth whose showed a considerable decrease in QoL compared to individuals with retirement and a low level of wealth [44, 45].

To have a retirement planning during life, and consequently the coverage by a pension plan, can help positively in the post-retirement life, especially in the individual’s perceptions, whether they are social, health or financial. Such help is mainly due to the psychological perceptions of financial issues that may influence how the individual will experience his or her old age. Then, social security planning can work as a strategy for coping with post-retirement life, since it not only prepares workers to meet their needs, but also supports them in the face of concerns about the losses of this phase of life.

The relationship between social security policies and psychological health

A total of five studies have analyzed the relationship of social security with psychological health, investigating symptoms of depression [47,48,49,50], anxiety [51] and stress [49].

Lower job control and poorer self-reported health lead to a lower retirement age, also, the risk of depressive symptoms is increased for people with a lower level of education [47, 49]. In addition, greater satisfaction of the needs for autonomy, competence, and relatedness was related to less depressive symptoms at baseline. However, satisfaction of pre-retirement needs was not a statistically significant predictor of subsequent changes in depressive symptoms throughout the transition to retirement. As for the basic psychological needs, only autonomy showed statistical significance, which demonstrated the existence of an initial short-term increase throughout the transition to retirement [50]. Besides, workers reported being in better health, less depressed, with more energy, fewer chronic conditions, and fewer limitations in their activities. Those who were retired reported feeling more bored, helpless, and hopeless [47]. Furthermore, being absent from the workforce through early retirement due to depression and other mental health disorders results in considerably less income than being in the workforce full time, as well as less wealth than those who have no mental health condition [48]. Regarding anxiety, a cross-country study suggests that the development of a social security system where the individual holds coverage for living expenses after retirement and health care decreases people’s concern about the future [51].

The results indicate that there is a link between psychological health and social security policies established when individuals decide to take early retirement, as a result of symptoms such as depression and stress, which generate a labor disability, and the need to activate the social security protection system due to a forced exit from the labor market. As well, the opposite logic can be seen where the existence of a robust social security system that provides coverage for life’s adversities, such as illness and old age, reduces symptoms such as anxiety.

Discussion

This meta-ethnography identified 17 eligible studies that examined the relationship between health conditions associated with aging and social security policies among people nearing retirement or retired. Most of the studies included in this systematic review involved cohorts aged 40 years or older and investigated associations between social security policies for people of retirement age and perceptions of, or behaviors related to, general health, psychological health or physical functioning. The synthesis of the evidence suggests that health can operate as a way to promote the working life for the elderly and as an indicator for social security policy reforms, that retirement planning is a strategic element for coping with post-retirement life, and that there is a relationship between social security policies and psychological symptoms.

About health as a way to promote the working life, four studies have found that changes related to sociodemographic dynamics point out that the phase between the ages of 50 and 70 has emerged as a type of second part of working life, which can be supported by a good self-assessment of the subject’s general health status when perceiving the possibility of staying in the labor market, albeit in an adapted way, such as by adopting bridge jobs [35,36,37,38]. The evidence suggests that if people can experience their old age in good health, they can be productive, still work and contribute to society, in a slightly different way from that of a younger person, promoting independence and increasing a healthy life for the elderly.

When it comes to health as indicator for social security reforms, of all the studies included in the synthesis, four studies allowed us to identify that a good health status can help individuals adapt to the changes generated by the reforms of the legal age criterion in the social security models and that people in poor health are the ones who suffer most from the crisis caused by unexpected changes in the welfare system [39,40,41,42]. This result is consistent with the literature reviewed, which has observed a variation in the health behavior of workers and in the health conditions of the samples researched that approaches social security reforms [52,53,54]. The results indicate that to ensure a healthy aging population, when reforming social security systems, policymakers have to enhance positive impact on health, since social protection aims to provide income security, health care and support at every stage of life, with particular attention to the most marginalized. However, the underlying mechanisms by which social security reforms appear to have this effect on health have not been evidenced, which may reflect an empirical evidence gap that is possibly developing.

Moreover, four studies included in the review enabled to indicate that there are actions in the life course that can help to obtain a satisfactory health after leaving the labor market, such as retirement planning; which according to the results found can reduce worry about retirement, keep anxiety under control, improve income and quality of life in the realization of this life event [43,44,45,46]. Retirement planning is defined as a goal-oriented behavior in which individuals devote efforts to prepare for their withdrawal from the labor market [25]; that could function as a strategic element for coping with post-retirement life.

Regarding the relationship between social security policies and psychological health, four studies suggested that the presentation of symptoms such as depression and stress, may demand from the social security system, as they are capable of disabling individuals, who will have a forced exit from the labor market [47,48,49,50]. And a cross-sectional study allowed us to infer that in countries where the level of development and comprehensiveness of its security system is higher, its population presents a lower anxiety picture when participants are asked about old age [51]. This is consistent with previous literature, where better health outcomes have been found in countries with a more extensive welfare state [15, 16]. These findings support the idea that mental health should be thought about and promoted, especially in the workplace, once social environments can affect health. A public-health guideline to aging should consider approaches that reinforce rehabilitation, adaptation and psychosocial growth.

In general, a significant number of studies have employed self-reported instruments to measure health conditions when considered in their general aspect [35, 43, 47, 48], which supports the importance of self-report as a meaningful indicator of health status. The increasing validity and adaptability of self-assessment scales have enhanced their use for academic, clinical, research, and epidemiological purposes, offering adequate levels of reliability in measuring and prognosticating short- and long-term measures of health [55]. Furthermore, the results found in this review can help to create the environments and opportunities that enable people to be and do what they value throughout their lives, increasing wellbeing and participation in society and promoting a healthy aging.

About the limitations of this review, the cross-sectional analysis of most studies restricts the validity of the results, as this prevented us from examining the cause-and-effect relationship of the variables. Also, considerable methodological variation was found in the theoretical perspectives consulted, the follow-up periods, and the questionnaires used in the studies to assess health conditions and social security measures, which hampered the meta-analytic analysis. This could have improved the interpretation and generalizability of the results and thus provided greater validity of the evidence.

The difficulty in defining and measuring retirement was also noted. On a conceptual level, a variety of theoretical approaches were found that operationalized retirement through self-report, legal concept, labor force participation, and pension receipt. However, this theoretical-conceptual variation may not be problematic as these approaches are not mutually exclusive as each assesses and analyzes a particular component of what is meant by retirement.

In spite of this significant heterogeneity in results, the multifaceted nature of health and social security allowed us to find a substantial amount of research that worked on their relationship, and made it possible to conduct the meta-ethnography. 58.82% of the studies had a low assessment score, i.e., a high risk of bias, represented by the lack of representativeness of the samples, the predominant use of self-assessment scales, and low risk factor verification. Finally, the selected publications were only from 1995 on, although our search covered research published from 1979 onwards, mainly due to the low methodological quality of the studies found in this period and the scarce quantity of studies detected between 1979 and 1994, revealing an increase in academic production and its publication from the mid-1990s.

Despite the limitations, the main strength of this systematic review was to conduct an analysis of health conditions related to social security policy reforms, synthesizing the evidence reported in a substantial number of relevant studies. These studies covered diverse population-based cohorts in large samples of middle-aged and elderly individuals, demonstrating the appropriate applicability of the theoretical construct of social security policies in diverse cultural contexts and methodological advances in the development and validation of outcome measures. This reflects not only the growing interest in research on variables based on human experience, but also in the search for empirical evidence to support the contribution of multidisciplinary constructs directed at public policy. At last, the searches of studies in four languages - English, Portuguese, French and Spanish - facilitated the understanding of the relationship between health conditions and social security policy reforms in samples of middle-aged and elderly participants from different cultures.

Conclusions

The results of this review included important health domains such as general health functioning, psychological health, and work disability factors. Overall, it showed that there is a link between health and retirement, where health is a relevant factor in deciding when to exit the labor market. This may encourage future researchers and policy makers to analyze the ramifications of its relationship to advance the promotion of quality of life for the elderly population.

For future research, the need arises to study and analyze the underlying mechanisms through which social security policy reforms and health conditions are related. Likewise, their potential benefits could be assessed through interventions aimed at promoting health for older workers, preventing psychological symptomatology, and planning for retirement. At the theoretical level, the conceptual diversity of retirement could represent an opportunity to operationalize this variable as a multifaceted construct, which could improve its explanatory and interpretive capacity in the face of different health outcomes for aging.