**2. Family-centered networks**

Networks' composition and the kind of ties linking those who are part of support systems to the focal person are central to approach the role of the family within social networks. When the network type is based on the kinship topography, the reference to the family bonds happens either by their dominant position and proportion in the network size, when compared to other relationship' ties, on the one hand, or by their absence or weakness position, by another hand (**Table 1**).

Studies with the general population present kin-dominant and nonkindominant networks [24] or minimal family and family networks [18] within typologies. The family reference in interpersonal relations is so strong that *asfamily* relationships are also mentioned in the literature [25]. Studies with older populations tend to defend this familial reference in social networks. Besides composition or other structural network properties, community participation, and social activities, especially of a religious nature, also emerged as a defining feature of several network types in research with older adults [3]. However, these are not the subject of this chapter.

Although not all typologies use nomenclatures that obviously express the central role of family ties in their definition, description, or presentation of the followed methodology, the family's role appears almost always as an essential reference. This happens in the cases of typologies presented by Mugford and Kendig [26] or Auslander [27]. The first typology highlights the family and the marital status, crossing the network size with the ties' multiplexity level, grouping attenuated, intense, diffuse, complex, and balanced networks [26]. The second one is based on supportiveness and the loss of support members, aggregating supportive, replacement, and traditional networks [27], corresponding the last type to a network where family ties and parenting are dominant features.

*Intergenerational Support Networks and Wellbeing in Old Age DOI: http://dx.doi.org/10.5772/intechopen.96302*


**Table 1.**

*Social network' types regarding the family system.*

Other typologies highlight specific family subsystems, household composition, parenting, conjugality, and marital status. Some examples are the marital networks [28], unmarried and married or coresidence networks [12], couple-focused type [28], child-focused [29]. Multigenerational households' networks [21], and widowhood networks [23].

The reviewed studies that propose network typologies were developed in different parts of the world and different cultures [3]. Although some cultures value the family more than others, as is the case in the southern European countries (e.g., [30]) concerning the rest of the continent, in the personal social networks of older people, the family seems to be a reference in the definition and characterization of networks regardless of culture. Alongside the family, friendship relations are also fundamental in defining the types of network, as are neighbors and other community relations [3], but these relational fields are not as strongly addressed and emphasized as family relations.

*Family network, family-focused network,* and *kin network* emerge as a social network type in eight studies with the older population conducted in diverse contexts and populations, such as the United States of America, China, Germany, and Israel.

Howard Litwin [6, 10, 11, 15] has proposed several typologies of older people's personal social networks, studying Israeli elders and older Soviet Jews who emigrated to Israel, highlighting the family role in social support and social relations. A more recent study with older European demonstrates that a greater family closeness brings beneficial changes to the social network in old age [17].

*Family networks* are addressed in the Litwin [6] typology developed based on a sample consisting of 2.079 older Jews in Israel, alongside *diverse network, friends network, neighbors network,* and *restricted network*. *Family networks* are defined as having frequent contacts with children living nearby and minimal contacts with neighbors and friends, being the diverse networks (composed of relatives, friends, and neighbors) the most frequent in the sample. The same type emerges in another study with older Americans [7], confirming the typologies and each type's main features. In Litwin's early studies [10, 11], of which proposals for typologies have

been presented, the kin network was the most frequent and tended to be larger than the average network size. The network included members of the extended family, with affective proximity and long-lasting ties. Another type was the *family intensive network*, which was, in turn, focused on the nuclear family and cohabitants, being the smallest type, dense and intimate, with long-lasting ties and frequent contacts. Although the *family intensive network* presented the highest social support levels in the mentioned research [10, 11], another study [14] that pointed out that the typology (*composed of kin, family intensive, friend focused and diffuse ties networks*) proved to be a good predictor of social support, the *kin network* was associated with a higher level of support and the family intensive network with the lowest level.

The *narrow family-focused network* and the traditional extended family network, alongside with other four network types (*diversified support network, friends-andneighbors-based support network, attenuated network,* and *religious family-focused network*), emerged in an investigation with the participation of Israeli elders [15]. The diversified support network was associated with marriage and parenting and was the most common type. The *narrow family-focused network* had frequent contacts with children and siblings living nearby, with support from close relatives, but limited community participation, being frequent among married individuals. Also associated with married people with a large offspring, the traditional extended family network comprises supportive children and other relatives living nearby [15]. Closeness and nearness are crucial in both network types.

In a different cultural context, such as China, the same closeness trait typifies the social networks of the older people in two studies [9, 14]. Cheng et al. [14] identified *family-focused* and *distant family networks*, as well as *diverse, friend-focused,* and *restricted networks*, highlighting the relevance of extended family support among an older Chinese population, especially in the absence of children and close relatives. The research undertaken by Li and Zhang [9] asserts that family networks are centered on kinship with close children.

Emphasizing the important role of spouse and adult children in Asian cultures, Park et al. [12] also found a *family-focused network* type, following their study with older Korean immigrants in the United States of America. However, the conjugality or marital status defined three of the six types of networks that the authors agglomerated*: married/coresidence network; unmarried/diverse network; unmarried/restricted network*. The study also presents *diverse* and *restricted networks* that will be discussed below. The *married/coresidence network* was one of the most frequent types in the sample. The unmarried types included those who were not married and likely to live alone. Another typology resulting from a study with South Korean seniors [28] also presents a *couple-focused network* type comprising young and educated married individuals with children and a distant large family, presenting low community participation. This study shows the prevalence of *restricted* and *couple-focused networks* representing limited social relations and the lack of a conventional *family-centered network* type [28].

Another research with the migrant population highlights intergenerational cohabitation, addressing the cultural and ethnic diversity in the social networks, carried out by Burholt and Dobbs [21]. Two types of *multigenerational households* emerge, mainly composed of relatives and centered in family relationships, involving married or widowed individuals: one with older integrated networks, another with younger family networks. The other types are the *family and friends integrated networks* and the *non-kin restricted networks*. The last type integrates non-relatives that are more isolated, typical of childless older persons living alone or only with their spouses, with formal support. Multigenerational living arrangements are common in certain familistic cultures, impacting the network types [21].

The contrast between *family* and *nonfamily* networks emerge in the study of Fiori, Antonucci, and Cortina [8] developed with older NorthAmerican, although they were less frequent types than others such as the *diverse networks*. The family network was focused on relations with children, and the *nonfamily network* belongs to unmarried or childless individuals, presenting the most limited social ties in the typology.

The family also emerges as central in elderly network typologies in studies developed in European countries. The precursor studies on types of senior networks from Clare Wenger [19, 22] in the United Kingdom establish a classification where family ties, social support, and community integration are relevant. The *family dependent support network* is centered on close family ties, having a small size, is usually held by older widows, with low community involvement.

A study with older adults in Germany [13] reported six different network types, in which the *family-focused network* represents those that belong to married people with frequent contact with family ties. In opposition, a *nonfamily restricted network* is characterized by infrequent family contact. As previously emphasized, the relationship between parents and children also plays a central role, defining the *family-intensive network* found in a study developed in Finland [16]. It pointed out the role of extra-family interaction for defining other network types. The *family network* type was found as the most frequent in a study with older adults in Portugal [2] with a very high proportion of family ties in the networks' size, attesting to the older persons' familistic nature networks in that context.

Confidant networks among older adults, composed of trust and especially meaningful persons with whom we share important issues, decisions or problems, also present the same trends found in network typologies in a broader sense, regarding the relevance of family ties in their definition. In a study with a large sample of several European countries, Litwin [31] defined a typology, thus distinguishes between *proximal family-based confidant networks* and *distal familybased confidant networks*. Closeness, proximity, and intimacy were the features that define the network types. The majority of participants were embedded in *family-based networks*, all very close emotionally and some with high proximity and frequent contact (including spouse and children, children, spouse, and other family members), in contrast with *nonfamily-oriented network* types (friend network and "the other" network). Another study on confidant or trust-relations networks in Portugal [20] addresses familial and nonfamily networks, based on size and composition (*small/big predominantly familial networks; small/big predominantly nonfamily networks*). *Small familial networks* are the most frequent. Proximity is also defining for the predominantly familial networks.

Particularly in later life, the family is considered the core of social networks [1]. For this reason, all typologies mention network types composed of family ties, dominated by family relations, or that recognize the importance of this kind of ties by underlining their absence or their scarce presence in the older persons' lives, as mention before. However, it is mainly in familistic cultural contexts that stronger kinship ties are emphasized [32].

Beyond closeness, nearness, and intimacy, other features like cohesiveness, accessibility, and stability are significant to understand family relations within social networks.

Stability over time refers to relational durability and the projection of the relational commitment in the future regarding intergenerationally. Longer relationships are associated with predominantly familial networks [2, 11]. A recent study evidenced that older Europeans [17] experienced diverse network transition, although, over time, close-family-based networks tended to prevail. Those in close-family networks had better life satisfaction and fewer depressive symptoms than the elderly with other network types, which points to greater family closeness as beneficial.

High cohesion or density levels between members are also expected in networks dominated by kinship [2, 20, 33]. This feature has positive and negative sides. Positively, it increases the ability to recognize the support needs and mobilize support quickly and effectively. This is a relevant trait since accessible family ties are strongly correlated with wellbeing in the older population [31]. Negatively, highdensity levels are also associated with centripetal functioning and control [33], inward oriented through cohesion forces, which can promote closure and isolation around family relationships, making possible changes difficult.

The family is seen as an intergenerational exchange system, where the family norm of reciprocity is established. When there is autonomy in old age, older people tend to have high social participation levels, having a strong helping paper in family life caring for younger generations [34]. Intergenerational solidarity becomes especially relevant when there are losses in independence and autonomy in the elderly. Generally, a dependence trajectory requires a care trajectory, being very relevant the informal care with a clear family anchorage [35]. Both sides of intergenerational care and exchange are important in network relations, promoting horizontal reciprocity.

Family relationships have become more complex and less dependent on marriage, blood ties, or cultural bonds, with the diversification of bonds beyond the household, linked by friendship, vicinity, dissolved marriages, step-parenting, or other ties [25, 36]. A new pattern of close relational commitments emerges, as well as care arrangements, involving close kin, extended kin, non-kin, and ex-kin, bringing enormous diversification to relational patterns [25, 36].

The levels and the ways of interaction between the older persons and their families, friends, neighbors, and community groups set the ties binding individuals to each other. Diversification is pointed as key to wellbeing, as we will discuss further.

### **3. Diverse and restricted networks, wellbeing and psychosocial risk**

Diverse and restricted social, personal networks are two frequent types that emerge in typologies when studying structural and functional dimensions of social relationships in old age (**Table 2**). The interaction of these types of networks with wellbeing and psychosocial risk tends to function in opposite directions.

The diverse networks underline the relational diversity in the network's composition, integrating family, friendship, neighborhood, and community bonds. However, they also highlight their broad distribution, not focusing specifically on one of the available relational fields and the associated functional diversity or their broad size. In the opposite sense, the restricted networks are associated with a confined composition, with few members and scarce social participation.

Several studies clearly point out the opposition between the diverse and restricted networks in the typologies they present based on samples with an older population [6, 7, 9, 12, 14, 28].

Litwin [6] found that *diverse networks* were the most frequent type in a study in Israel, composed of relatives, friends, and neighbors, with a high contact rate. Usually, this type was associated with married men, younger and with adult children living nearby. *Friends networks* had similar features of *diverse networks*, but with minimal contact with neighbors. Restricted networks were more frequent among widowed older persons, having the most limited social ties, narrow contacts with adult children, and no contact with friends or neighbors. In another research with older Americans [7], the authors confirmed those features, reporting hat

*Intergenerational Support Networks and Wellbeing in Old Age DOI: http://dx.doi.org/10.5772/intechopen.96302*


#### **Table 2.**

*Diverse and restricted network' types.*

restricted network was associated with lower social capital and greater social risk, while the diverse networks were associated with higher social capital and wellbeing.

*Diverse networks* were also the most frequent type in the study by Cheng et al. [14], focusing on social networks and older adults' subjective wellbeing in China. The least frequent were *family-focused* and *restricted networks*. *Diverse networks* tend to be those with a higher number of members and a frequency of contacts. The conclusions report the same direction of the above-mentioned study, that *diverse* and *friend focused networks* are associated with higher levels of wellbeing and the best indicators of support, while restricted networks are associated with lower levels [14]. Also, with the Chinese population, Li and Zhang [9] asserts that the most balanced social resources were found in the *diverse network*, which is associated with marriage, social engagement, and urban settings. *Restricted networks* were the most common in the sample and were associated with older individuals and rural contexts. The *diverse networks* yield the most beneficial health results, and the r*estricted* had the most negative outcomes in subjective health and psychological wellbeing. The authors consider it essential to facilitate older people's social participation with physical limitations and poor psychological conditions, enhancing their interaction with diversified groups [9].

Confirming the same trend in findings, with some specifications, Park, Smith, and Dunkle [28] identified *restricted* and *diverse networks* within the typology that resulted from the study with South Korean older adults. The *restricted* type is characterized by closeness to adult children and limited contact with friends, comprising older urban men with social participation. The *diverse* type presents the greatest extent and diversity of ties, being held by healthy and active women. *Restricted networks* were prevalent in the sample and reported lower wellbeing and higher depressive symptomatology than other types. The *diverse networks* presented low levels of social support, despite its most favorable characteristics.

The last reviewed study that presents *diverse* and *restricted network* types [12] indicates the *diverse* type as one of the most frequent types. Individuals with *diverse networks* were married and lived with others, having frequent contact with close family and friends. They also maintained community participation, including religious activities. The *restricted* were less represented in the sample, referring to individuals with limited contacts with family and friends and minimal involvement in religion. The study also proposes another restrictive type of network: the *unmarried/restricted* type, which included unmarried people and likely to live alone, with

closer relationships with friends than family and minimal religious participation. Similar to previous studies, the *diverse network* is associated with better health and lower depressive symptoms, whereas the *unmarried/restricted* had the opposite outcomes [12].

Focusing on community participation, a study in Mexico [23] points to two *diverse* and two *restricted* networks: *diverse* with and without community participation and *nonfamily-restricted*. The *diverse* type includes family and friends, with frequent contacts. Having or not community participation is the dimension that differentiates them. *Diverse networks* without community participation were the most frequent in the sample. The less frequent was the *nonfamily-restricted network*, common among childless older persons, with a reduced number of relatives and a low frequency of contacts, centered in friendship relations. The nonfriendsrestricted networks were held by older people with children and an extended family, with whom they have frequent contacts, not including friends, and having limited community participation. *Restricted* types and *widowed networks* were associated with negative self-rated health, dependency, and depression [23].

Some regularities should be underlined. Several characteristics of the restricted networks are associated with the older persons' personal, relational, and social vulnerability, being less supporting and less effective. Restricted networks with the designations represented in **Table 2** are other examples. Private restricted support networks [19, 22] are small, centered in cohabitants, and characterized by the absence of relatives and friends nearby, minimal contact with neighbors and limited community engagement, and being associated with couples or isolated individuals. Older persons without spouses have attenuated networks [15], focused on the interaction with close adult children, that is their only support source, having very low contact with other ties and low community participation. *Restricted-nonfriendsunsatisfied* [13] and *restricted-nonfamily-unsupported* [13] have similar profiles: the first one held by unmarried elderly with small networks, low social activity, and low support; the second adds infrequent family contact. Stone and Rosenthal [29] found two fragile network structures regarding vulnerable groups with fewer interpersonal resources: *small - friendship-poor and socially isolated* and *small - extendedfamily and friend-focused*, both with small network size and lack of contact with the few available network members. The *small extended-family and friend focused networks* are small and dominated by siblings and friends, belonging to lifelong single and childless women [29].

*Defective networks* [16] are small, with fewer friends and no neighbors, presenting higher levels of isolation, also related to childless elderly. *Nonfriends-restricted networks* [23] belong to older people with adult children and extended family, with frequent contacts, but exclude friends, and they have low community engagement*. Nonfamily-restricted networks* [23] are associated with childless older persons, with fewer relatives and a low frequency of contacts, centered in friendship relations. Also, the *non-kin restricted networks* [21] are typical of childless older persons living alone or only with their spouses, being more isolated. They integrate non-relatives and tend to have formal care, being the most vulnerable regarding loneliness and isolation identified by the study [21].

Although socially restricted, the *couple-focused* type among South Korean elderly presents high levels of life satisfaction and low levels of depressive symptomatology [28], but it is an exception that has cultural peculiarities.

Focusing on mental health among older adults, a study with NorthAmerican [8] found diverse, family, and friends' networks, as well as two types of restricted networks (nonfamily network and nonfriends network). Depressive symptomatology was lowest for individuals in the diverse network and highest for those in the nonfriends network [8]. The nonfamily network had the most limited social ties

#### *Intergenerational Support Networks and Wellbeing in Old Age DOI: http://dx.doi.org/10.5772/intechopen.96302*

and are associated with unmarried or childless elders, and the nonfriends network had few contacts with friends and low social engagement. The absence of family in friends is considered less detrimental for mental health than the absence of friends in the family [8]. Reinforcing these conclusions, a recent study in rural China [37] found that social network types (diverse, family, friends, restricted, and familyrestricted networks) were significantly associated with depressive symptoms and health in older adults, following the same trend. Older people with diverse or friend networks presented lower depressive symptoms, whereas those with restricted and family-restricted networks had reported higher levels of depressive symptoms and lower self-rated health levels.

As stated by Fiori, Smith, and Antonucci [13], we can affirm that individuals with *diverse-supported networks* present high levels of wellbeing, realizing the opposite when the individuals have *restricted networks*. However, the relations between network types and health or wellbeing outcomes are very complex. Emotional closeness and having more than one type of relationship within networks, such as family and friends, positively affect older persons' wellbeing [38]. Compared with family-centered networks, those composed of friends have a greater impact on wellbeing and quality of life [38].

Another literature review point that having more friends than family in their social network, its diversity, size, the frequency of contacts, having co-residency with own child, and social capital was protective regarding late-life depression; low diversity strongly contributed to explaining network structures that were predictive of depression in old age [39]. Quality of support is refereed as a key mechanism by which network types affect mental health [8].

The diversity of social ties and broad network size are consistently associated with positive social support, wellbeing, and health, indicating, on the other hand, an association between restricted networks and less effective characteristics and psychosocial risks.
