Table 3. Characteristics of included studies on social network.

Reference	Sample/Setting [Country]	Study Type	Follow-up Times	Social Network [Measure]	MH Outcome [Measure]	Main Findings for Social Network
Baek et al., 2021 [78]	n = 291 married couples [South Korea] Age: ≥60	Korean Social Life, Health, and Aging Project (KSHAP)	5 waves: 2011, 2012, 2014–15, 2015–16, 2018–19	Fours questions in total about supportive relations and negative relations	Depressive symptoms [CES-D]	The association between husbands’ and wives’ depressive symptoms was stronger for couples that reported a low level of supportive marital relations, but only for the wife and those that reported a high level of negative marital relations for both the husband and wife.
Chang et al., 2016 [79]	n = 21728 elderly women [USA] Age: ≥65	Prospective cohort study: Nurses’ Health Study (NHS)	5 timepoints, Baseline and 4 biennial follow-up questionnaire cycles over 10-year period	Berkman-Syme Social Network Index	Depression [MHI-5 subscale of the SF-36, CESD-10, GDS-15]	Social factors (lower social network; lower subjective social status; high caregiving burden to disabled/ill relatives) were associated with higher incident late-life depression risk in age-adjusted models.
Domènech-Abella et al., 2021 [70]	n = 5066 older adults [Ireland] Age: ≥75	Longitudinal Aging Study Amsterdam (LASA)	4 waves over 10 years	Names of persons with whom they had regular contacts in the past year	Depressive symptoms [CES-D]	Decreasing social network size (Coef. = -0.02; p < 0.05), predicted higher levels of loneliness, which predicted an increase in depressive symptoms (Coef. = 0.17; p < 0.05) and further reduction of social network (Coef. = -0.20; p < 0.05).
Domènech-Abella et al., 2019 [69]	n = 5066 adults [Ireland] Age: ≥50	Irish Longitudinal Study on Ageing (TILDA)	2 waves of TILDA Second wave: 2012–13 Third wave: 2014–15	Berkman-Syme Social Network Index	MDD or GAD [CIDI]	Both objective social isolation (size of social network) and loneliness factors have been found to be robust risk factors for depression and anxiety independently, which acts as a warning not to underestimate the subjective aspects of social isolation.
Förster et al., 2018 [80]	n = 783 elderly people [Germany] Age: ≥ 75	Population-based cohort study: Leipzig Longitudinal Study of the Aged (LEILA)	3 timepoints T1: baseline T2: follow-up1 T3: follow-up2	PANT	Depressive symptoms [CES-D]	Persons with a restricted social network were more likely to develop depression, and risk of depression was particularly high for elderly with social loss experiences.
Noteboom et al., 2016 [29]	n = 1085 respondents from health care settings [Netherlands] Age:> 18	Longitudinal cohort study: Netherlands Study of Depression and Anxiety (NESDA)	2 timepoints T1: baseline T2: 2-year follow-up	Questions on how many relatives, friends or others over the age of 18 years they had regular and important contact	Depressive disorders [CIDI]	Structural (network size and partner status) did not predict depression at follow up. Pariticipants with a lifetime history of depression reported a smaller social network.
Van Den Brink et al., 2018 [30]	n = 1474 patients with a major depressive disorder [Netherlands] Age:18–90	Data from 2 cohort studies NESDA and Netherlands Study of Depression in Older Persons (NESDO)		Questions on social network characteristics	Presence of depression [CIDI] Depression severity [IDS-SR]	Social network characteristics, such as having a partner and number of persons in one’s household, are related to depression course.
Reynolds et al., 2020 [81]	n = 3005 elderly people [USA] Age:57–85	Panel data from National Social Life, Health, and Aging Project (NSHAP)	3 waves 2005, 2010, and 2015 with 5-year intervals between each wave	Questions on community-layer, interpersonal-layer, and partner-layer connection	Depressive symptoms [CES-D]	Results demonstrate multiple links between social connection and depression, and that the evolution of social networks in older adults is complex, with distinct mechanisms leading to positive and negative outcomes. Specifically, community involvement showed consistent benefits in reducing depression.
Santini et al., 2020 [82]				Social Disconnectedness Scale	Depressive symptoms [CES-D-ML] Anxiety [HADS-A]	Perceived isolation was positively associated with depression symptoms at T2 and T3 (β = 0·12; p<0·0001).
Santini et al., 2021 [83]	n = 38300 adults [13 European countries] Age: ≥ 50	Survey of Health, Ageing and Retirement in Europe (SHARE)	2 consecutive waves (2011, 2013) of the SHARE survey	The total number of close relations in the social network	Depressive symptoms [EURO-D Scale]	Social participation among people with relatively few close social ties was negatively associated with depression symptoms but did not seem to benefit to those with relatively many close social ties.

Checklist: CES-D = Center for Epidemiological Studies-Depression Scale; CES-D-ML = Center for Epidemiological Studies-Depression Minus Loneliness Scale; CIDI = Composite International Diagnostic Interview; GAD = Generalized Anxiety Disorder; GDS = Geriatric Depression Scale; HADS = Hospital Anxiety and Depression Scale; IDS-SR = Inventory of Depressive Symptomatology Self-Report version; MDD = Major Depressive Disorder; MHI-5 = Mental Health Index-5.