Modelling age-friendly environment for social connectedness: a cross-sectional study

Introduction

Malaysia is expected to become an aged nation by 2030 where 15.3%, approximately 5.8 million of its population, would comprise of individuals aged 60 and above according to the Department of Statistics, Malaysia.¹ Malaysia is categorised as one of the nations that is ageing at the fastest pace because Malaysia would transform into an aged nation within 20 years, as individuals aged 60 and above only comprised of 7.6% of its population in year 2010.² The main problems faced by ageing adults in Malaysia are illness, deteriorating physical functioning, existing building and facilities not being age-friendly designed, as well as less social interactions and feeling less connected to others.³

According to The World Factbook, the majority of Malaysians (77.7%) live in urban areas. Urbanisation and population ageing have become issues for healthcare-related personnel, researchers, and town planners in developing healthy and liveable environments for the elderly.⁴ The threats found in urban areas are harmful to social well-being such as environmental pollution, tobacco and alcohol abuse, malnutrition, lack of physical activities, inaccessible transportation, poor infrastructure, and inadequate healthcare services.⁵ Furthermore, new buildings and most of the old buildings in Malaysia were built without meeting the statutory guidelines for disabled persons.⁶ The development of an age-friendly environment is essential because the current environment is built without taking into considerations the needs of people of all ages.⁷

Having the opportunity to deepen existing relationships and social connectedness with others is important for those who are ageing.⁸ Relationships become gradually more important when an individual ages; through an age-friendly environment initiative, one can improve one’s social connections and receive adequate social support, which is beneficial for maintaining health and well-being.⁹ In order to improve the well-being of the ageing population, an age-friendly environment should be developed to support social connectedness, participation, and integration among the elderly.¹⁰

In acknowledging the growth of the ageing population worldwide, the World Health Organisation (WHO) introduced two policy frameworks, namely, active ageing¹¹ and global age-friendly cities.¹² WHO (2007, p. 1)¹² defined an age-friendly city as one that “encourages active ageing by optimizing opportunities for health, participation, and security in order to enhance quality of life” of elderly individuals. An age-friendly community helps the elderly “to maintain social connectedness while deepening existing relationships” (Emlet & Moceri, 2012, p. 2).¹⁰

Empirical studies of the age-friendly concept are few.¹³ Despite Malaysia being predicted to become an aged nation by 2030, studies conducted in relation to the ageing of the population are relatively limited compared to developed countries. There is an urgency to restructure policies for the ageing population that includes local factors that address the forthcoming health and social needs.¹⁴ The ageing of the population in Malaysia has created unprecedented challenges to policy makers, scholars, and business organisations as Malaysia will become an aged nation by 2030.

In an initiative to make environments better for the elderly and to accommodate their needs, the WHO Global Network of Age-friendly Cities and Communities was formed to facilitate mutual communication, discussions, and learning among countries, cities, and communities globally. Many developed countries, cities, and communities joined the network and adopted the global age-friendly cities guide.¹² To date, the network consists of 847 cities and communities across 41 countries, which include more than 230 million people, and the network is expanding. Taiping city in Perak, Malaysia was chosen for the two-year age-friendly city pilot project in August 2019 and is on track to achieve age-friendly city status.

It should be pointed out that Malaysia has not adopted the WHO (2007) global age-friendly cities guide.¹² Hence it is important to identify the constructs of age-friendly environments and subsequently examine the relationships between age-friendly environments and social connectedness from the perspectives of middle aged and older adults caring for themselves in Malaysia. This study empirically modelled an age-friendly environment and social connectedness model through structural equation modelling (SEM). SEM provides higher accuracy in predicting the research model as validity of the model is statistically evidenced through the assessment of goodness-of-fit indices.¹⁵

This study aims to empirically identify the significant constructs of age-friendly environments by adapting the WHO (2007) global age-friendly cities¹² guide as well as examining the relationship between age-friendly environments and social connectedness by bridging the literature gap as suggested by Menec et al. (2011)¹⁶ and Emlet and Moceri (2012).¹⁰ This is among the first few attempts to model an age-friendly environment by identifying its significant constructs and subsequently linking to social connectedness from the perspective of middle-age and older adults caring for themselves in a developing country. The proposed research model is important and adds empirical evidence on the readiness and preparation of Malaysia for becoming an aged nation by 2030.

Literature review

An age-friendly environment offers a comprehensive “age-integrated as well as age-targeted” initiative that helps the ageing adults to improve their social connectedness and self-worth¹² (WHO, 2007, p. 72). The concept of social connectivity is an indispensable advantage of age-friendly environments given that an age-friendly environment forms a mutual connection between the ageing adults and their environment.¹⁶ The literature suggests that local authorities and practitioners should focus on neighbourhood design and establish community-based interventions to improve social connectedness and social participation among the elderly.^17,18

Lawton’s ecological theory of ageing (Nahemow & Lawton, 1973)¹⁹ and Bronfenbrenner’s ecological theory (1977)²⁰ are widely used to address the adaptation between environment and humans, whereby there should be a balance between both.^19,20

The environment can bring certain pressures or hassle to daily living as an individual ages, which would create inconveniences for them.²¹ According to the ecological theory of Lawton (1974),²¹ there are two main causes for the decrease in capacity as one ages, namely perceived pressure from the environment and weakening of bio-physical functioning.

Bronfenbrenner (1977)²⁰ explained that the ecological environment can be visualised as a set of nested environmental structures, which are arranged at each inclusion of another and interact bi-directionally. The structures are the microsystem, mesosystem, exosystem and macrosystem. Microsystem refers to interactions between an individual and their immediate environment such as family, home, friends, and workplace. Mesosystem refers to the interconnectivity of an individual with the interaction of multiple immediate environments. According to the Bronfenbrenner’s ecological theory (1977),²⁰ individuals interact with the different environment structures in daily living. The ecological frameworks focus on the interconnectivity between humans and varied social and physical environments.²² Both the ecological frameworks emphasise the vitality of the social aspect of an environment.^20,23

Lui et al. (2009)²⁴ and Menec et al. (2011)¹⁶ indicated that elderly people would like to remain active and be involved in community activities in their silver days. An age-friendly environment supports and enhances social connectivity among the ageing population.

In the application of ecological frameworks to conceptualise an age-friendly environment, the environment should be dynamic to accommodate the changes in physical and cognitive functioning as well as social capacity as people grow older.¹⁰ Furthermore, social inclusion is a significant component in addressing age-friendly environments and could be improved by removing the obstacles brought by elements of the physical and social environments.²⁵

Feldman and Oberlink (2003)²⁶ are amongst the pioneers of conceptualisation of age-friendly environments. They conducted focus group interviews involving adults aged 35 and above and community leaders from the United States. The study found four indicators of age-friendly environments, namely identifying basic necessities, promoting social and civic engagement, maximising independence for frail and disabled elderly and optimising physical and mental functioning and well-being. The study concluded that financial security, health and healthcare, social connections, housing and supportive services, transportation and safety were important factors for the elderly to age-in-place.

Subsequently, WHO (2007)¹² conducted focus group interviews to examine age-friendly city features from a wider perspective involving 1,485 elderly people aged 60 and above, 250 caregivers, and 515 service providers across 33 cities across continents worldwide. They found eight features that made a city age-friendly, which were outdoor spaces and buildings, transportation, housing, social participation, respect and social inclusion, civic participation and employment, community support and health services, as well as communication and information. On the other hand, Alley et al. (2007)²⁷ studied elder-friendly communities using qualitative methods. They found that the significant criteria of elder-friendly communities were accessible and inexpensive transportation, safe residences, healthcare, safety, and being involved in the community.

Richard et al. (2009)¹⁷ examined the associations between neighbourhood features and social participation among 282 urban elderly people in Canada through a survey. The results found that social participation had positive implications for their health. Hence, they suggested that the urban and local community authorities build a supportive neighbourhood that promoted social interaction and participation among the elderly.

Menec et al. (2011)¹⁶ conceptualised the age-friendly community based on the WHO’s 2007 age-friendly city model¹² and ecological theory. Their study proposed seven important domains, namely physical environment, housing, social environment, opportunities for participation, informal and formal community support and health services, transportation and communication and information. They also suggested that the concept of social connectivity served as a basic support to age-friendly communities. Subsequently, Menec et al. (2013)¹³ explored the characteristics of age-friendliness in Canada through a survey of 1,373 individuals. A larger percentage of the respondents scored highly on social age-friendly domains, such as social environment, participation opportunities, health-care services, and communication and information, in comparison with physical domains. Hence, they found that age-friendly domains in the social environment were easier to implement than in the physical environment.

From the Asian perspective, Kadoya (2013)²⁸ explored the barriers faced by the elderly when interacting with society through their participation in local community programmes in Japan and found that the ability to be mobile affected their social interaction. Hence, he concluded that social inclusion was an important factor for age-friendly cities. Lai et al. (2016)²⁹ studied the WHO (2007) age-friendly¹² environment features and linked them with active ageing. The study surveyed 211 informal caregivers and 402 adults who cared for themselves aged 45 and above in Malaysia through purposive sampling. They found that outdoor spaces and buildings, transportation, and housing as well as community support and health services were significant age-friendly environment features. The findings implied that the development of physical environment features such as a comprehensive public transportation system, elderly-friendly housing, outdoor spaces and buildings as well as medical and healthcare services should be carried out before social environment development.

Franke et al. (2013)¹⁸ conducted a qualitative study to investigate the main areas that promoted physical activity among the elderly who were ageing actively in Canada. These areas were resourcefulness, social connections, and built environment. The study suggested that new policies addressing social connectedness and accessibility to social and physical environments were required to enhance and preserve independence and mobility among the elderly.

Scharlach and Lehning (2013)²⁵ reviewed and described the various age-friendly efforts implemented in the United States, which helped to enhance social inclusion among the elderly from the perspective of physical and social environments. An environment that enabled the elderly to be mobile and with an accessible transportation system would help to promote social inclusion. In general, social inclusion could be improved by removing the hurdles found in the physical and social environments.

Ruza et al. (2015)³⁰ compared the perspectives of age-friendliness among professionals and local community residents via focus group interviews. Both the professionals and local groups agreed that community and health services had the greatest significance in establishing age-friendliness of a city. Meanwhile, communication and information as well as civic participation and employment were found to have the least significance.

Rogers and Mitzner (2017)³¹ conducted a study to predict the personality, needs, and wants of the future elderly of 2050 by taking into account their independence, health, well-being, and social connectedness. There was a similarity between the future and current elderly; they wished to prolong their independence. Social connectedness and social activities were essential for the general well-being of older adults. Therefore, being socially connected with family, friends, and neighbours helped ageing adults to improve their life satisfaction.

To bridge the research gap and supported by ecological theories, this study empirically examines the constructs of age-friendly environments and proposed positive relationships between an age-friendly environment and social connectedness from a developing country perspective. Figure 1 illustrates the proposed research model of age-friendly environments and social connectedness; and the two hypotheses are as follows:

Figure 1. Research model of age-friendly environment and social connectedness.

H₁: Outdoor spaces and buildings, transportation, housing, social participation, respect and social inclusion, civic participation and employment, community support and health services, as well as communication and information will be confirmed as constructs of an age-friendly environment.

H₂: An age-friendly environment is positively related to social connectedness.

This study examines eight features of age-friendly environments and social connectedness. These constructs are reflective models. As a result, the eight features form a latent variable of age-friendly environments. The latent construct of age-friendly environments is identified by a set of indicators. In short, the eight constructs of age-friendly environments are the indicators of age-friendly environments. Hence, an age-friendly environment is a formative model in which the valid constructs are confirmed through second-order confirmatory factor analysis (CFA).

Methods

Results

A total of 402 middle-aged and older Malaysian adults caring for themselves participated in the study. The sample had a greater proportion of female participants than male; 274 (68.2%) female and 128 (31.8%) male.⁵⁵ The sample was evenly divided between 212 (52.7%) middle-aged adults and 190 (47.3%) older adults. The sample consists of a large number of Chinese respondents (240, 59.7%) and married respondents (260, 64.7%). The majority of the respondents (278, 69.2%) were not highly educated and their highest level of education was either primary level, secondary school level, or had attained Sijil Pelajaran Malaysia (SPM). An individual who has completed their secondary school education is awarded with Sijil Pelajaran Malaysia (SPM).⁴⁸ Results of demographic profiles are presented in Table 1.

Cronbach’s alpha test was performed to examine the reliability of the variables. An item from the housing construct was removed to improve the coefficient. Variables were reliable given that the value of Cronbach’s alpha coefficient for each variable was larger than 0.7, as shown in Table 2.

Subsequently, EFA through the principal axis factoring method and promax method of rotation was performed on 82 responses. A clean factor structure was achieved with four constructs, namely communication and information, built environment, community support and health services, and civic participation and employment. The result of the Kaiser-Meyer-Olkin (KMO) test was 0.84 and Bartlett’s test of sphericity was significant with p = 0.00. Pattern matrix analysis showed that factor loadings all exceeded 0.5, indicating that the analysis of factor structure was valid. The final factor structure suggested that the four factors be retained as each had eigenvalues higher than one. The results of EFA are shown in Table 3.

The EFA results showed that three features of age-friendly environment, namely outdoor spaces and buildings, transportation, and housing were grouped and explained as one single factor, which the study renamed as built environment. It was possible to re-classify these three features as a single factor given that they are the “key features of physical environment” (WHO, 2007, p. 9).¹² This study named the combination of the three features as built environment because built environment comprehensively addresses the features of outdoor spaces and buildings, transportation as well as housing.²⁷

At the same time, two features, namely respect and social inclusion as well as social participation, were removed as these items had cross-loading with other variables. WHO (2007, p. 9) identified respect and social inclusion, social participation, and civic participation and employment as “social environment and of culture that affect participation and mental wellbeing”.¹² Social participation as well as civic participation and employment were re-grouped and explained as opportunities for participation because they were related to participation of the ageing adults in social activities.¹⁶ Civic participation and employment was found to be more practical in the context of Malaysia in addressing social participation among the ageing adults.

Subsequently, first-order CFA of age-friendly environment and social connectedness features was carried out with 27 items. To examine the correlation among the constructs, they were allowed to covary. Overall first-order CFA of age-friendly environment and social connectedness features allowed the evaluation of convergent validity and discriminant validity.

The initial model did not achieve satisfactory model fit. Therefore, model respecification was performed by examining modification indices as standardised factor loadings had exceeded 0.5. Items with high modification indices were removed or allowed to co-vary in the model. This is because the items that highly explain a construct should be retained, which helps to improve the accuracy of a specified model.⁴⁹ A satisfactory first order CFA of age-friendly environment and social connectedness features was achieved with a model fit of CMIN/df = 1.19, GFI = 0.96, AGFI = 0.94, RMSEA = 0.02, SRMR = 0.03, NFI = 0.98, TLI = 0.99, CFI = 0.99 and ECVI = 0.62. The results of CFA of age-friendly environment and social connectedness features as well as convergent validity are presented in Table 4.

The results showed that convergent validity was achieved with all the items, with satisfactory standardised factor loadings exceeding 0.7. The AVE indicated that all latent variables had good validity with all values greater than 0.5. For the reliability assessment, all the constructs were acceptable with composite reliability (CR) and Cronbach’s alpha coefficient exceeding 0.7.

Table 5 shows the results of discriminant validity using the method of Fornell and Larcker (1981).⁴⁷ The constructs achieved discriminant validity, as the bold values are larger compared to the other values.

Second-order CFA indicated satisfactory standardised factor loading with three items exceeding 0.5 and one item near 0.5. The model was statistically fit with CMIN/df = 1.37, p-value = 0.05, GFI = 0.97, AGFI = 0.95, RMSEA = 0.03, SRMR = 0.03, NFI = 0.98, TLI = 0.99, CFI = 0.99 and ECVI = 0.39. The second-order model indicated that among the four, built environment had the highest standardised factor loading of 0.86 and the largest causal effect towards an age-friendly environment. Community support and health services followed next with a standardised factor loading of 0.80. Next was civic participation and employment with a standardised factor loading of 0.76. Communication and information, with a standardised factor loading of 0.49, had the smallest causal effect towards an age-friendly environment.

Although the cut-off point for standardised factor loading was 0.5, communication and information, with a standardised factor loading of 0.49, was not removed from the model. This was because the model achieved satisfactory model fit, indicating no further model re-specification procedure was required. Moreover, a standardised factor loading of 0.49 is very close to 0.5. It is worth noting that in the CFA presented by Paúl et al. (2012)⁵⁰ and Lee and Scott (2004),⁵¹ indicators were not removed when factor loading was less than 0.5.

Given these results, H₁ can be rejected. Instead of eight, only four constructs, namely built environment, community support and health services, civic participation and employment, and communication and information were confirmed as constructs of an age-friendly environment. These four constructs are in line with the characteristics of age-friendly environments described by Feldman and Oberlink (2003)²⁶ and Alley et al. (2007).²⁷ An age-friendly environment included supportive and accessible social services as well as safe neighbourhoods, which was supported by the findings of Alley et al. (2007).²⁷ The results of second-order CFA are presented in Figure 2.

Figure 2. Second-order confirmatory factor analysis of age-friendly environment features.

ChiSq, chi-square; df, degrees of freedom; GFI, goodness-of-fit index; AGFI, adjusted GFI; RMSEA, root mean square error of approximation; NFI, normed fit index, TLI, Tucker Lewis index; CFI, comparative fit index.

Next, a measurement model was specified by allowing age-friendly environment and social connectedness features to co-vary. The initial model achieved a fairly satisfactory fit as p-value was at 0.05. Model re-specification was carried out by removal of the item “I have no sense of togetherness with my peers” from social connectedness. Removing an item from the social connectedness variable was theoretically justified as the remaining three items still had a similar underlying sense of “connectedness” (Lee & Robbins, 1995, p. 236).³⁸ Hence the final model achieved satisfactory fit with p-value = 0.08, CMIN/df = 1.22, GFI = 0.96, AGFI = 0.94, RMSEA = 0.03, SRMR = 0.05, NFI = 0.98, TLI = 0.99, CFI = 0.99.

The structural model was specified by linking age-friendly environments to social connectedness. The structural model was valid as all the items had standardised factor loadings of at least 0.5. It met satisfactory goodness of fit for all the indices with an insignificant p-value at 0.08, CMIN/df = 1.22, GFI = 0.96, AGFI = 0.94, RMSEA = 0.03, SRMR = 0.05, NFI = 0.98, TLI = 0.99 and CFI = 0.99. The model showed that age-friendly environments were found to be positively associated with social connectedness, with a standardised coefficient of 0.18.

Approximately 3% of the variation in social connectedness was explained by the variation in age-friendly environment features. Hence, H₂ was supported, indicating that age-friendly environments are positively related to social connectedness. This finding is in line with the findings of Menec et al. (2011)¹⁶ and Emlet and Moceri (2012).¹⁰ The positive relationship indicates that age-friendly environments help to promote and support social connectedness among middle-aged and older adults caring for themselves.¹⁶ Middle-aged and older adults caring for themselves who were leading an active and socially connected lifestyle were able to strengthen their emotional health, improve their general well-being, and delay their ageing process.³¹ The structural model of age-friendly environments and social connectedness is illustrated in Figure 3.

Figure 3. Structural model of age-friendly environment and social connectedness features.

ChiSq, chi-square; df, degrees of freedom; GFI, goodness-of-fit index; AGFI, adjusted GFI; RMSEA, root mean square error of approximation; NFI, normed fit index, TLI, Tucker Lewis index; CFI, comparative fit index.

Discussion

It was worth noting that not all the eight features of WHO’s (2007)¹² age-friendly cities were perceived to be significant in the Malaysia context. Built environment was found to be the most important construct of an age-friendly environment. Built environment increased protection and security for middle-aged and older adults caring for themselves when they move around and interact with the environment and community.¹⁶ As an environment becomes friendly, they can participate in social activities, exercise in the park, receive adequate healthcare, go shopping, and visit family and friends.^10,16 A senior-friendly public transportation system was also important for their mobility.^25,28 Having access to community healthcare services and recreational facilities could improve their social and physical well-being, which indirectly becomes an alternative to health promotion.¹⁸

Community support and health services were found to be strongly associated with age-friendly environments. These findings are consistent with Ruza et al. (2015)³⁰ and Lai et al. (2016).³¹ Age-friendly community support and health services enable middle-aged and older adults caring for themselves to receive necessary healthcare at the lowest cost to maintain a strong sound body and mind.^9,12 Being healthy increases the capacity of middle-aged and older adults caring for themselves to interact with and live in their existing environment.²¹ Therefore, healthcare and medical services found within local communities are beneficial as these adults do not solely dependent on the services provided by the state or federal government.

Civic participation and employment referred to voluntary options and paid opportunities in workplaces that take into consideration the disabilities of the ageing adults. The results indicated that middle-aged and older adults caring for themselves in Malaysia preferred to contribute and participate in society through paid or voluntary opportunities after their retirement. This result echoed the findings of Feldman and Oberlink (2003)²⁶ who indicated that financial security as well as social and civic engagement were important age-friendly features.

Fourth, the communication and information construct was found to have a weak association with age-friendly environments, which was supported by Ruza et al. (2015).³⁰ The result shows that communication and information obtained through the Internet assist with and ease communication among middle-aged and older adults caring for themselves, helping them to stay in touch with people, which in turn makes them feel less isolated. A possible explanation is that middle-aged and older adults caring for themselves are afraid of losing up-to-date information, news and being isolated from society.¹²

On the other hand, social participation and respect and social inclusion were not significant constructs of an age-friendly environment. A possible reason may be that poor public transport systems and service quality in Malaysia are disabling rather than enabling and may hinder some social participation. The public transportation system in Malaysia is not well connected or friendly to older people. Poor services and inaccessible public transportation in Malaysia discouraged people from using it.⁵²

The results have implications for the Malaysian government and its related agencies to develop public policies for the welfare of its ageing population and older adults. The government must improve the living environment of the ageing population and focus on the physical environment. Furthermore, the results found that middle-aged and older adults caring for themselves were keen to participate in the labour force. Therefore, the authorities could encourage the private sectors to hire ageing workers by subsidising their training expenses or their salary. The Inland Revenue Board of Malaysia could develop schemes for income tax rebates or tax reliefs for companies that recruit older employees aged 60 years and above.

The results imply opportunities for a sustainable economy. An age-friendly workplace reduces the barriers to working opportunities for older employees.¹² Business organisations can renovate their workplace or design an ergonomic workplace to accommodate the physical considerations of ageing adults. The findings support the development of an age-friendly business industry within the age-friendly environment. An age-friendly business industry is a solution to economic growth and sustainability as there are opportunities for new businesses,¹⁶ such as building senior-friendly housing with safety features, recreation facilities, and healthcare centres in the community. Another age-friendly business opportunity is to help middle-agers plan financially for their retirement, such as offering long-term care insurance and private retirement schemes.⁵³

Conclusions

The overall findings indicate a simpler age-friendly environment in the Malaysian context, which evidenced only four age-friendly features instead of the eight recommended by WHO (2007).¹² Built environment was found to be the most important construct, implying that governments of developing countries should give priority to developing features of the built environment, such as a supportive external environment and an accessible public transportation system. In Malaysia, many of the built environment features and an age-friendly transportation system have yet to be comprehensively formed.²⁹ The poor public transportation system may affect the development of social environment features such as opportunities for civic engagement and social participation.^28,30

A non-probability sampling method was employed and most of the respondents were active adults caring for themselves, which may limit the generalisation of the findings. Therefore, attention should be paid when generalising the results to other developing countries or different settings. Future studies can examine the moderative effect of demographic variables such as gender and culture. An age-friendly environment index would enable the authorities to further identify areas of improvement for the benefit of the ageing population.

The results concluded that an age-friendly environment helps middle-aged and older adults caring for themselves to feel connected to the world and society as well as to feel related to anyone. Age-friendly environments help elderly people to stay active and participate in their community through accessible transportation systems and barrier-free environments.²⁶ Social connectedness can be promoted by eliminating the physical barriers that arise from the built environment.^25,29 Feeling connected with society, family, and friends can become a form of emotional resource.²⁹ Age-friendly environments enable elderly people to improve their social connectivity, indirectly promoting independence as well as physical and social well-being.

Data availability