1. Introduction
The development of age-friendly cities and communities, adapted to older people’s needs, has become an important area of work in the fields of health, ageing, and public policy. This is the result of several trends, including the complexity of demographic change, the policy objective of supporting the maintenance of people in their homes for as long as possible [
1] and the recognition of the role of the environment in active and healthy ageing [
2].
In this sense, the “Age-Friendly Cities and Communities” initiative launched by the WHO in 2007 has been extended to more than a thousand cities and communities worldwide [
3], highlighting the importance of the systematic and inclusive approach in generating enabling environments [
4]. In the Basque Country, the age-friendliness movement began in 2009 with the adhesion of the capitals to the initiative launched by the WHO. In 2012 the Basque Government launched a project at territorial level (Euskadi Lagunkoia-Age-Friendly Basque Country), which currently includes more than 60 municipalities [
5]. Since then, age-friendliness has become one of its political strategies on ageing [
6].
An age-friendly community is defined as a place where older people are actively involved, valued, and supported with infrastructure and services that are effectively tailored to their needs [
7].
Making cities and communities age-friendly ensures that they are inclusive and equitable places, leaving no one behind, especially the most vulnerable older people [
4]. Age-friendliness is primarily characterized by the adaptation of mutually reinforcing social and physical environments, by a participatory model of collaborative governance and, above all, by inclusion [
8]. This emphasizes an underlying assumption that is now widely shared by policymakers: An enabling social environment is as important as the material conditions in determining the well-being of older people’s lives [
9].
Research has shown that age-friendly environments are associated with higher levels of well-being and quality of life in older people [
10,
11,
12,
13]. It has even been concluded that older people who perceive their environments as age-friendly are almost four times more likely to report a better quality of life than those who report lower levels of age-friendliness [
11]. Its positive association to well-being is not surprising, as the criteria for an age-friendly environment align almost perfectly with both concepts [
13]. However, much existing environmental gerontological research has focused on the indoor settings, mainly homes and overlooked wider contexts such as neighborhoods and communities [
14,
15]. More recently, some studies have extensively examined the multidimensional aspects of the environment [
15,
16]. Van Dijk [
15] explored the relationship of social characteristics of the neighborhood, such as social capital, social cohesion, and social support with well-being. Park and Lee [
14] found that after controlling for demographic covariates, physical and social environment features are significantly related to the life satisfaction of older people in Korea. This finding is an empirical basis for identifying those aspects of the environment that can serve as modifiable resources and improve the psychological well-being of older people, in this case, particularly the most vulnerable ones.
Despite these conclusions, there is still insufficient understanding of the actual holistic effects of interventions on the physical and social environments [
9,
17]. More evidence needs to be generated about how improvements in both settings affect the health and well-being of older people [
8]. One of the main challenges facing all community initiatives–either related to ageing or not-is how to assess their impact on individuals and groups [
18]. The term “age-friendly” is used when considering how various aspects of a community promote or reduce the health and well-being of individuals as they age [
19]. However, scientific research continues to investigate why some places are more age-friendly than others and how age-friendliness relates to the well-being of older people [
8].
Focusing on well-being, there is a growing interest in the variables that influence the subjective well-being of older people. Some of the individual factors related to well-being in old age have been highlighted in previous research. Some of these studies conclude that well-being was strongly related to socio-demographic [
20]; socio-economic [
21,
22]; health [
23,
24]; lifestyle [
25,
26]; psychological characteristics [
27,
28]; as well as social relations [
22].
Gender is one of the most relevant socio-demographic characteristic in old age. Growing old is not the same for men and women [
29]. Sex and gender are important determinants of health and wellbeing [
30]. Gender is the characteristic with enough evidence to suggest that there is different wellbeing or survival relationships for older men and women [
31]. To explain social phenomena, the Gender Perspective takes into account the differentiated or egalitarian situation, depending on belonging to one sex or the other [
29]. Gender is an explanatory social category, also constructed, that helps to understand what lies behind biological sex. Gender, as a social construction, reveals differences in thinking, roles, health, economics, politics, and labor and in old age, these differences are even bigger. These inequalities do not appear in the last stages of life, but are nourished throughout the life cycle and are therefore continuous [
32].
Although the Framework for Action on Active Ageing highlighted gender as a cross-cutting determinant, research on ageing from a gender perspective is still limited [
29]. The gap in women’s representation in human studies has been well documented [
33]. Gender-based analysis is designed to identify the sources and consequences of inequalities between women and men and to develop strategies to address them [
30].
In this context, the general objective of this study is to analyze the interaction between age-friendliness, subjective well-being, and gender, specifically regarding the perception of age-friendliness of the physical and social environment, among men and women aged 55 and over in the Basque Country. Variables of perception of the eight areas of age-friendliness, ranging from public spaces, housing, transport, social participation and employment, civic participation, respect and social inclusion, communication and information, and social and health services, are included.
2. Materials and Methods
Although the WHO has made several documents available to facilitate the measurement of age-friendliness [
34,
35], there is no official quantitative tool yet. One of the challenges in evaluating age-friendly city initiatives is to identify an evidence-based approach [
36,
37,
38]. There is a great need for monitoring, evaluating, measuring and assessing the age-friendliness of cities and communities [
38]. Despite these challenges, in recent years different researchers have developed a number of tools to measure and evaluate these initiatives in a quantitative way [
38,
39,
40,
41,
42,
43,
44].
In this study, however, the tool for measuring friendliness is based on a survey of the living conditions of people aged 55 and over carried out periodically in the Basque Country. This means it does not count on a tool designed ad hoc, but it includes a series of relevant indicators related to the areas of age-friendliness indicated by the WHO [
45].
The methodology of this study was based on a survey of a representative sample of community-dwelling residents aged 55 and over in Basque Country (739,231 people aged 55 and over, 33.8% of the total population). Structured interviews were conducted through face to face survey based on a questionnaire assisted by computer. Sampling selection was made through stratified random sampling considering geographic area and age group (55–64 years-old, 65–69, and 80 and over) as main criteria of stratification. Sample distribution followed a proportional method for territory strata and quotas according to age group (55–64, 65–79, and 80 and over) and gender were applied. Households in each stratum were chosen by random selection of those with one person aged 55 and over, only interviewing one person per household. Sample size was determined by required level of disaggregation. Statistics on ageing generally categorize older people as being above a certain age threshold. Indeed, the United Nations (UN) defined older people as those aged 60 years or more in World Population Ageing 2013, while the WHO states that older people in developed world economies are commonly defined as those aged 65 years or more. The WHO also uses an alternative definition, whereby an older person is defined as someone who has passed the median life expectancy at birth [
46]. The age selected for analysis, 55 and over, is one of the most widely used cohorts for the study of the ageing population from research and public policy (for example: The Active Ageing Index). Although, there is still much debate about the redefinition of the threshold for the onset of old age because of its implications for the design of public policies and for the social perception of old age [
47].
The sample was composed of 2469 individuals (1177 men and 1319 women). Their ages ranged from 55 to 102 years old, with 69.36 (SD = 9.9) being the average age. Regarding marital status, 61% of the participants were married, 9% were single, 6% were divorced, and 24% were widowed (
Table 1).
Anonymity and confidentiality of the answers were guaranteed and participation in the study was voluntary. All subjects gave their informed consent for inclusion before they participated in the study and the protocol was approved by the Ethics Committee of University of Jaen (MAY.13/3.TES).
The questionnaire includes items and scales that explore perceptions in relation to areas of age-friendliness (age-friendliness components). The WHO’s subjective well-being scale (WHO5 Well Being Index -1998) is also included for this analysis.
To find out the predictive power of the components of age-friendliness on the subjective well-being of men and women aged 55 and over in the Basque Country, two multiple linear regression models have been carried out, using the WHO5 Well Being Index as the dependent variable [
48], and both the predictors of well-being according to the von Humboldt and Leal [
49] categories (16 variables) and the selected components of age-friendliness (34 variables) as independent variables
Subjective well-being is a general term used to describe the level of well-being experience of people according to subjective assessments of their lives [
49]. It was measured using the WHO-5 Well Being Index -1998 version, a five-item self-administered scale, Likert-type scale with response options of 0 to 5. This scale exclusively measures the positive aspects of the psychology of well-being in short non-invasive questions. It is one of the most widely used questionnaires to assess subjective psychological well-being [
50].
The validity of construct of the WHO-5 was analyzed with the item response theory model formulated by Rasch (2012) in older people [
51], which confirmed that the 5 items constitute a one-dimensional scale, where each item adds unique information on the level of well-being [
50]. A total score can be obtained by adding up all items, with a range of scores from 0 to 100 covering from a total absence of well-being to the highest conceivable level of well-being [
50].
In the field of later life, the WHO-5 scale has been validated in different studies for the detection of depression [
52,
53,
54], apathy [
55], or suicide [
52]. An internal and external validation of the scale has also been carried out in the general older population [
56], concluding that the WHO-5 is a useful instrument for the identification of people with reduced subjective quality of life.
This study analyzed the psychometric nature of the scale used to ensure the quality of the measurement. This was done through the analysis of internal consistency as well as through the performance of exploratory factor analyses of main components with Varimax rotation.
Regarding linear regression analyses, in order to control the predictive values of the age-friendliness components by the factors that has been identified by literature and evidence as predictors of wellbeing [
49], some independent variables were included such as age, educational level, habitat, marital status, type of household, or health status, among others. These are drawn from several categories that are divided into: Social support; socio-demographic factors; health status; and psychological factors. Moreover, a total of 34 variables related to the 8 age-friendliness areas described by WHO have been included as age-friendliness components (
Table 2). The number of indicators available in the survey related to each domain is different, which means that there are domains with 10 indicators and others where only one indicator has been found. The variables in this study encompassed within the physical environment are those included in the areas of outdoor spaces, housing, transport, communication, and health services (Areas 1, 2, 3, 7, and 8). The variables considered in the social environment are those in the areas of social participation, respect, and social inclusion, and citizen participation and employment (Areas 4, 5 and 6). The indicators corresponding to the age-friendliness components proposed by the WHO were included in the linear regression model to explore their association to wellbeing.
In a first phase, a descriptive analysis of the variables separated by men and women was carried out, following the recommendations of Calvente, Rodrigo, and Morante [
57], to observe the gender gaps. For the analysis of the quantitative variables, the basic statistics (mean and standard deviation) were used. For categorical variables, the relative frequency distribution with 95% confidence intervals was used.
In order to know the relationship of subjective well-being with the rest of the variables, an analysis of variance was carried out to determine the predictors of subjective well-being, by gender. In order to know the predictive power of age-friendliness over subjective well-being, linear regression models separated by gender were constructed. Then, variables that did not contribute to explain this relationship were eliminated until the final models were obtained.
4. Discussion
Gender is a characteristic with enough evidence, suggesting that there is different well-being or survival relationships for men and women [
31]. For instance, Lennartsson and Silverstein [
59] found that solitary activities reduced mortality risk for men but not for women, while Warr, Butcher, and Robertson [
60] found that family and social relationships influenced more women’s well-being compared to men. Agahi and Parker in Adams and colleagues [
31] also found that women’s mortality risk decreased with social activities, whereas men’s risk was higher with solitary hobbies and gardening, for example. This previous evidence highlights the importance of analyzing subjective well-being in men and women separately in order to clarify what influences and predicts the well-being of both.
The final model obtained in men is composed of nine predictors, four of which are included in the components of age-friendliness, particularly social and citizen participation. The other variables included are relevant factors selected from scientific evidence where social support, health status, and psychological factors predominate. Thus, the variables predicting men’s well-being are living with other people, being satisfied with personal relationships, being in good health, being satisfied with their achievements and with how safe and protected they feel, as well as carrying out physical exercise, domestic and social leisure activities, and participating in associations of a gastronomic, educational, cultural, or sporting nature.
Regarding women, the final model consists of 11 factors, 4 belonging to the components of age-friendliness within the areas of social participation, citizen participation, and social inclusion. The other includes socio-demographic, social support, health, and psychological variables. Living in a rural environment (less than 20,000 inhabitants); having been born in the Basque Country; maintaining contact with family, friends and/or neighbor; being satisfied with their personal relationships; being in good health; not needing help for the DLA; being satisfied with their achievements; doing physical exercise; tourism or travel; being satisfied with their feeling of belonging to the community; and participating in neighborhood associations, predicts higher levels of subjective well-being in women.
The results, therefore, have shown that only for women socio-demographic factors such as habitat or origin are relevant, leaving out other important variables such as age, marital status, wealth, or educational level. These findings contradict previous studies where some demographic variables, such as age, income, work status, marital status, and educational level [
61] are associated with subjective well-being. However, these studies have also revealed that these variables only explain a partial amount of the variance in well-being [
61]. This means that the weight of other variables is very important and could explain, to a greater extent, individual differences in levels of well-being and the subjective nature of the concept [
61]. Additionally, subjective well-being as an overall measure does not seem to undergo significant changes associated with age, either in cross-sectional or in longitudinal studies [
62]. On the other hand, educational level does not seem to be relevant either when controlling for other factors, and it is possible that education may exert indirect relations to subjective well-being through a mediating role [
61]. With regard to the level of wealth or income, it has also been concluded that countries with the highest level of wealth have almost no correlation between this variable and well-being [
61,
63,
64,
65,
66].
Sociodemographic variables have been maintained as predictors of subjective well-being only in women, being origin (place of birth) and rurality (less than 20,000 inhabitants). Regarding rurality, some researchers have analyzed the differences between urban and rural communities related to well-being, but scarce evidence is found in terms of ageing population [
67]. According to Van Hoof and Kazak [
67], some studies have corroborated that rural areas approach or exceed urban areas in terms of life satisfaction or well-being. However, the higher the density of urban settlement, the greater the proximity to public services, influencing the quality of life of people and being especially relevant for older people [
67]. In the literature, findings about environment and health focus largely on urban areas, however, how the age-friendly community characteristics are related to rural environments is less known [
39]. Research of well-being in rural women has also been relatively scarce [
68]; for instance, some authors found a relationship between habitat and life satisfaction in women, in which rural women obtained higher life satisfaction than urban women, although not significantly [
68]. In this study, women who live in areas with fewer than 20,000 inhabitants were more likely to report higher levels of well-being. A possible explanation may be related to the conclusions of previous research in which satisfaction with social relations is higher among older people living in rural environments [
69], due to in less populated areas people know each other better. Neighbors are a very important type of relationships in villages, especially for women [
62]. Studies such as Bosch and Gómez [
70] show that women have a deeply rooted role and caring function, and in rural environments they find support to continue to carry out their role in “good neighbor” networks in an adequate and satisfactory way.
Origin, a socio-demographic predictor of women’s well-being, refers to having been born in the Basque Country. This variable could be understood in relation to the importance of the community and belonging to the place, another predictor variable of well-being in women. As women age, their dependence on their neighborhood increases [
71] and their residential stability is associated with attachment to the community [
71,
72].
Regarding men’s well-being, socio-demographic variables have not remained as predictors. However, social support factors emerged as predictors, for instance, living with other people. For men, living in a household with other people, i.e., living together, has a positive effect on their well-being. This type of social support is very important because of their relative poorer performance and independence in leading an autonomous life on their own, at least compared to women [
73]. The assignment of traditional gender roles leads to a men’s higher level of dependence in domestic tasks, which implies the need to live with other people in order to perform them. Several studies show how living together as a couple entail higher life satisfaction, better emotions, physical and mental health, economic resources, and social integration, support, and relations [
74]. Additionally, marriage is one of the strongest predictors of subjective well-being: Married people report a higher degree of life satisfaction than single, widowed, or divorced ones [
61,
65,
66,
67].
Another predictor of well-being, in this case for men and women, is satisfaction with personal relationships. Good relationships with family, friends, and other people in their social network provide higher indices of subjective well-being in both genders. Social support networks have been identified as important influences on the affective and cognitive components of well-being [
75]. Furthermore, protective effects of social networks on morbidity and mortality have been strongly recognized [
71,
76]. This correlation between various types of social support and well-being in older people [
19] is considered a consistent result.
Health predominates among other predictors of subjective well-being. The relationship between health and well-being has also been found in multiple studies [
19,
22,
77,
78]. Menec and Nowicki [
39] found that perceived health had significant positive effects on the physical environment, social environment, opportunities for participation, and transport options. In our study, perceived health is the strongest predictor in the final models for men and women.
In addition, in women, the situation of dependence or autonomy remains together with health. For them, not needing help to carry out the daily life activities is also a clear predictor of higher subjective well-being. Functional limitations of older people have been found to be associated with it [
61,
78] and some studies have corroborated how this affects women more [
61]. These results may be related to the importance of living together for men, but not for women. In this sense, older women, more “autonomous” in terms of not needing other people to run their home, can find greater obstacles in functional dependence situations, decreasing their subjective well-being.
On the other hand, in men, satisfaction with feelings of safety and protection is included as one of the psychological factors predicting well-being. These results can be analyzed in coherence with the previous ones on coexistence. For their greater well-being, men need to perceive a safety network that begins with live as a couple. In later life “the family becomes the great substitute for employment as a source of sociability, identity and self-esteem or time structuring” ([
79], p. 115). When men retire, they move from a socially open relationship in a working environment, which provides them with references of identity, prestige, friendships, solidarity and so on, to a state that requires them to adjust to new developments in the world of marriage and the family [
80]. In women, however, the continuity of the role can have positive consequences, with a better psycho-social adjustment at this life stage.
Moreover, satisfaction with the achievements made in life is one of the psychological factors emerged as predictor of well-being for both men and women. In this sense, self-esteem and self-efficacy have positive relationships with the highest level of well-being [
61]. Regarding self-efficacy, Gómez et al. [
81] indicate that the more confident a person feels in achieving their goals and objectives in life, the higher level of subjective well-being experiences. According to Diener et al. [
65], the relationship between subjective well-being and goals is mediated not only by the fact that people have clear goals, but also by the progress made in achieving them [
62]. At the individual level, the continuity or replacement of roles through participation in appropriate activities and with family support can contribute to the individual’s sense of meaning or purpose and the maintenance of a sense of identity [
31]. Participation in activities involves the pursuit or achievement of personal goals, thus adding a sense of personal mastery or achievement [
31,
82].
Therefore, the activities carried out by people take on special relevance for the perception of well-being. The other large group of variables included in the model presented in this study are those age-friendliness components described above. Of the eight areas of age-friendliness, only few remain in the model for predicting well-being in men and women, mainly social and citizen participation.
Among the social participation activities predicting well-being for men and women, physical or sporting activities such as doing sport or exercise, walking, going to the mountains, etc., predominate. Several studies have already confirmed this positive relationship between the practice of physical exercise and the feeling of well-being and personal satisfaction [
61,
83,
84], and even effects have been found on physical and psychological health in older people [
85].
In men, it has also been identified that participation in domestic leisure activities or hobbies, such as gardening, handicrafts, DIY, etc., is a predictor of well-being. These activities, commonly more solitary, have obtained contradictory results in previous research [
86], although a recent study has found that “solitary-active” activity is associated with a reduction in mortality risk in men [
59].
Other types of activities that favor the subjective well-being of men are those of a social nature such as going to the bar, meeting friends, going outdoors to have lunch or dinner, etc. These type of social participation and social contacts have positive correlations with personal well-being measures, as confirmed by previous research [
86] and from which the ageing theory of activity emerges [
87]. The review conducted by Adams et al. [
31] confirms that informal social participation, such as visiting friends, has a positive outcome in relation to well-being in old age. According to these authors, strong evidence relating social activity with positive well-being, as inherent social intimacy seems, to be a very important, if not the most important, aspect of engagement that influences people’s well-being [
31].
However, for women, tourism, travel, and/or excursions are the social participation activities that predict subjective well-being. These results are corroborated by previous studies that have shown how travelling in old age affects physical, psychological, social, and spiritual dimensions such as integration, acceptance, contribution, updating, and coherence; and that this leisure activity offers opportunities for significant participation in adulthood [
88,
89]. The review done by Morgan et al. [
90] finds that tourism can improve the well-being of older people and promote a renewed sense of purpose, facilitating their transition from work to retirement. Several authors have found that tourism has a positive psychological impact on older people, on their well-being, quality of life, perceived health, and life satisfaction, regardless of the type or length of travel [
90]. However, few studies analyzed tourism in old age from a gender perspective [
88,
89]. Liechty et al. [
88] found that group travel with other older women promoted a sense of belonging, empowerment, feelings of self-determination, personal growth, positive emotions and has well-being implications.
On the other hand, in the area of age-friendliness that encompasses social inclusion, the variable feeling of belonging was included. This predicts well-being in women, but not in men. Similar results were found in the study of Tiraphat et al. [
11], including social trust as one of the predictors of age-friendly environments [
11]. Recent studies have shown the positive effects of social capital, as measured by group participation, sense of belonging, and relationship with neighbors [
91] on the perceived functional and psychological health of older people. Well-being is also closely related to a sense of community, and this acts as a predictor of social and psychological well-being [
92]. With regard to gender, Phillipson et al. [
93] found that women were more concerned about the deterioration of social capital, as they fulfil the role of “neighborhood keepers”. So, as women get older, their dependence on their neighborhood increases [
71]. The feeling of belonging seems to be related, in turn, to other predictors of well-being in women such as origin and rurality. In Young et al.’s study [
76] another variable associated with a greater sense of belonging in women was living alone, that is, those women who lived alone had a greater sense of belonging to the community arguing that they may have developed social support networks to compensate for living alone.
Finally, for both men and women, the resulting regression models had variables from the civic participation component, although the specific indicator differs. For men, it is participation in educational, cultural, sporting, professional, gastronomic, choral, or literary associations, which predict well-being. Conversely, for women the predictor is participation in neighborhood associations. Several studies have found that participation of older people in socially productive activities is associated with well-being [
94] as well as meet service needs in the community [
15].
The differences between men and women in the type of civic participation are linked to the other predictor variables and the gender differences found throughout this study. Associations of a cultural, sporting, professional, choral, or gastronomic nature are traditionally made up of men in the Basque Country, something that is reflected today in their greater participation in this type of associations [
95,
96]. Neighborhood associations played a fundamental role in the recovery of areas of sociability in the Basque Country, not only in the field of self-management, the assembly movement, or the control of municipal management, but also in others such as the revitalization or recovery of popular festivals [
95]. This type of associative participation being a predictor of women’s participation is a result that is in line with the previous ones. The neighborhood is a space conquered beyond the home, which means many benefits for women, and which is related to other variables such as rurality, origin, or sense of belonging.
Therefore, none of the variables introduced as physical environment areas are kept as predictor variables in the resulting models, in contradiction to other studies that have found that the areas of “outdoor spaces and buildings”, “transport”, “housing”, along with “community support and health services” [
15], and “security” [
97] seemed to be essential elements for older people and an important goal to prevent them from losing their social and physical well-being [
97].
Despite the findings, our study also has some inherent limitations. First, it is based on the use of a survey not specifically designed for the purpose of measuring perceptions of friendliness. For this reason, some indicators may not fully reach to catch the multidimensionality of friendliness components. Even so, this survey has a wide variety of indicators related to this concept, so this limitation can become an opportunity by optimizing the use of this tool for different purposes and to test its potential in the research of this or other fields. Additionally, it is a cross-sectional study, which prevents capturing the dynamics of friendliness and well-being and extracting causal inferences. Therefore, it is not possible to determine their direction of the association based on the findings of this study. However, the results establish a significant association, which is an important step that encourages further studies to identify directionality. On the other hand, although the survey contains a variable to measure the habitat in which the respondent resides, which would allow a multilevel analysis to be made by looking at the differences between the more rural and more urban environments of men and women, the sampling conducted does not provide a sufficient number of people who meet these conditions for a detailed analysis of habitat of less than 20,000 inhabitants. Future research could indeed be developed in order to identify area level effects regarding friendliness.