Overall, emotional and social loneliness and their associated factors: a large-scale cross-sectional study of young and middle-aged adults in Germany.

Overall, emotional and social loneliness and their associated factors: a large-scale cross-sectional study of young and middle-aged adults in Germany

Background

Loneliness is increasingly recognized as a major public health concern, yet little is known about the distinct patterns and associated factors of emotional and social loneliness during young and middle adulthood. This study examined age-related differences in overall, emotional, and social loneliness and associated sociodemographic, social, health-related, and personality factors among adults in Germany.

Methods

We analysed nationally representative data from the German Family Demography Panel Study (FReDA), including 14,443 adults aged 18 to 51 years. Loneliness was assessed using the validated De Jong Gierveld Loneliness Short Scale, distinguishing overall, emotional, and social loneliness. Multivariable regression models examined associations with sociodemographic, social, health-related, and personality factors. Analyses were stratified by age and gender.

Results

High levels of loneliness were common, with 24.3% of participants reporting intense emotional loneliness, 29.6% intense overall loneliness, and 35.8% intense social loneliness. Overall loneliness remained relatively stable across age groups. In contrast, emotional loneliness was lower in older age groups, while social loneliness was higher. Social relationships showed some of the strongest associations with loneliness, particularly social loneliness: discussing personal matters within one’s social network (β = -0.32, 95% CI -0.37 to -0.28) and having a partner (β = -0.21, 95% CI -0.26 to -0.16) were associated with lower social loneliness, while first-generation migration background was associated with higher social loneliness (β = 0.31, 95% CI 0.24 to 0.38). Mental health burdens were consistently associated with higher loneliness, especially emotional loneliness, with depressed mood showing the strongest association (β = 0.22, 95% CI 0.17 to 0.27). Among personality traits, agreeableness was associated with lower social loneliness (β = -0.19, 95% CI -0.22 to -0.17), whereas neuroticism was associated with higher emotional loneliness (β = 0.17, 95% CI 0.14 to 0.20). Gender differences were generally small, as were associations with physical health conditions.

Conclusions

Loneliness in young and middle adulthood is multidimensional, with emotional and social loneliness showing distinct age-related patterns and associated factors. Social relationships showed some of the strongest associations with lower loneliness, while mental health burdens and first-generation migration background were associated with higher loneliness. Public health strategies may benefit from considering different loneliness dimensions and psychosocial contexts.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12889-026-30122-4.

Background

Loneliness - defined as the subjective perception of insufficient quality or quantity of social relationships [1, 2] - is increasingly recognised as a public health concern [3, 4]. It is associated with adverse outcomes including poorer physical and mental health and premature mortality [5–10].

Most empirical studies have treated loneliness as a unidimensional construct, typically focusing on overall loneliness as the primary outcome [5, 11–13]. Within this unidimensional literature, findings on age-related patterns are inconsistent, ranging from U-shaped and other non-linear trajectories to linear declines or stable levels [13–19], potentially reflecting differences in measurement, study design, and national or cultural context [12, 13, 20–22]. Compared with research on older adults, young and middle adulthood has received less attention in loneliness research, although this life phase (often described as the “rush hour of life” [23]) is marked by career development, family formation, and changes in social roles and relationships, which may shape experiences of loneliness [22, 24]. A recent German panel underscores the relevance of loneliness in this population, with approximately one-third of young and middle-aged adults reporting overall loneliness [25].

However, the predominantly unidimensional approach to loneliness may provide only a partial picture. A multidimensional approach may therefore offer a more comprehensive understanding of loneliness by distinguishing between emotional and social loneliness [26–28]. This distinction builds on Weiss’s relational perspective, which states that emotional loneliness reflects unmet needs for intimacy, attachment, and emotional support, while social loneliness reflects unmet needs for broader social integration, companionship, and network support [27]. Importantly, these dimensions may differ in their correlates and consequences [28].

From a theoretical perspective, the distinction between emotional and social loneliness can be further informed by Granovetter’s theory of the “strength of weak ties” [29]. Granovetter distinguishes between strong ties, such as close friends, partners, or family members, which are characterised by emotional intimacy and frequent contact, and weak ties, such as acquaintances, colleagues, or more peripheral social contacts, which connect individuals to broader social networks and social structures. Applied to loneliness research, deficits in strong ties may be particularly relevant to emotional loneliness, whereas deficits in weak ties may contribute to social loneliness through reduced integration into wider social networks. This perspective provides a theoretical rationale for expecting different associations between relational characteristics and emotional and social loneliness. At the same time, Cacioppo’s conceptualisation of loneliness [2, 30] emphasises that loneliness depends on individuals’ subjective evaluations of their relationship quality rather than on objective network characteristics alone. Together, these perspectives suggest that both the structure and perceived quality of social relationships may contribute differently to emotional and social loneliness.

Previous research has linked loneliness to a broad range of factors, including sociodemographic characteristics, social relationships, physical and mental health, and personality traits [7, 8]. However, most evidence concerns overall loneliness, and considerably less is known about whether these factors are differentially associated with emotional and social loneliness.

Sociodemographic characteristics may be relevant to loneliness because they reflect differences in social, economic, and life-course circumstances that may shape opportunities for social participation, relationship formation and access to social resources. However, evidence for dimension-specific associations remains heterogeneous. Education, for example, has shown inconsistent associations, ranging from no association with either dimension [31], to lower social loneliness among highly educated younger adults [32], and modest associations in older adults [33]. Recent evidence from young and middle-aged adults in Germany further suggests that socioeconomic resources may be relevant to both dimensions, with income poverty associated with higher and income wealth with lower emotional and social loneliness [34]. Migration-related experiences may be relevant to loneliness, as migration can disrupt established social networks and involve barriers to social integration [35–37]. These processes may affect both emotional and social loneliness, although existing evidence regarding dimension-specific associations remains inconsistent [32, 35]. Importantly, migration background encompasses heterogeneous experiences, with recent longitudinal evidence from Germany suggesting variation by gender and age at migration [38].

Social relationship factors may be particularly relevant to distinguishing emotional and social loneliness because different relationships fulfil different relational needs. Consistent with the distinction between strong and weak ties [29], close relationships may be particularly relevant to emotional loneliness, whereas broader social connections may be more relevant to social loneliness. Empirically, partnership has particularly been associated with lower emotional loneliness [31–33], whereas having children has not been associated with either dimension [32]. Recent evidence links overall loneliness to perceived support from friends, companionship within personal networks, and perceived social support. Longitudinal studies further indicate that the relevance of different social network ties may change throughout adulthood [39–41]. However, research into how broader social networks specifically relate to emotional and social loneliness remains limited.

Health-related factors may also be associated with loneliness because mental and physical health can shape individuals’ experiences of social relationships and opportunities for social participation. In a representative German population sample, overall loneliness was associated with a broad range of mental health indicators, including depressive and anxiety symptoms, perceived stress, and lower life satisfaction [16]. Recent evidence further implies that the relationship between loneliness and mental health may be bidirectional [10, 42]. Dimension-specific evidence suggests that poor mental health is associated with both emotional and social loneliness, although associations may be stronger for emotional loneliness [32, 43]. Physical health may be particularly relevant when functional limitations restrict social participation. Recent German research has shown associations with both dimensions, with particularly pronounced differences in social loneliness among middle-aged adults [44]. Together, these findings suggest that both mental and physical health should be considered when examining the correlates of different loneliness dimensions.

Personality traits may likewise be relevant as they are related to how individuals establish, maintain, and perceive social relationships. Recent evidence supports robust associations between loneliness and the Big Five personality traits (extraversion, agreeableness, conscientiousness, neuroticism, and openness), particularly higher levels of neuroticism and lower levels of extraversion [45–47]. Further, longitudinal studies have found that there are links between personality traits and subsequent loneliness [47]. However, evidence distinguishing emotional from social loneliness remains limited. Among Norwegian students, neuroticism was positively associated with both emotional and social loneliness, whereas other traits were negatively associated [48]. It remains unclear whether these dimension-specific associations also apply across young and middle adulthood.

Building on the conceptual distinction between emotional and social loneliness and the evidence for potentially dimension-specific correlates outlined above, we registered two interrelated research questions and six corresponding hypotheses [49]. Using large-scale, nationally representative data from German adults aged 18 to 51 years, we first examine the prevalence of overall, emotional, and social loneliness in young and middle adulthood, hypothesising that the three forms of loneliness vary by age (H1) and gender (H2). Second, we hypothesise that sociodemographic characteristics (H3), social relationships (H4), health-related factors (H5), and personality traits (H6) are differentially associated with overall, emotional and social loneliness. In addition, we examine whether these associations vary by age and gender.

Methods

Dataset and sample

We use data from the German Family Demography Panel Study (FReDA) [50, 51], a large-scale representative survey of the German population aged 18–51. For our study, we focus on wave 1B (autumn/winter 2021/22), which includes 20,216 respondents. Wave 1B includes most of our variables of interest. However, information on educational background is included in wave 1R (the recruitment wave). Depressed mood and partnership status are available in wave 1 A (spring/summer 2021), and Big Five personality traits are included in wave 2B (autumn/winter 2022/23). We have therefore transferred these variables to wave 1B. This is straightforward for variables obtained from waves 1R and 1 A, as all FReDA subwaves of wave 1 contribute jointly to wave 1 [52]. The Big Five personality traits, however, were measured after the loneliness outcomes in wave 2B. Their inclusion is supported by evidence indicating that personality traits remain largely stable over periods of up to four years [53, 54], which suggests that the later measurement approximates respondents’ dispositions at the time of the loneliness assessment. This does not establish temporal ordering. We therefore interpret all associations as cross-sectional and do not make claims about the direction of effects. We excluded respondents who identified their gender as diverse (N = 46) due to the insufficient sample size, which did not allow for separate analyses. Analyses were conducted on complete cases, resulting in the exclusion of respondents with missing values for any of the variables of interest (N = 5,727). This results in a final sample size of N = 14,443. Around two-thirds of the missing information was concentrated in the Big Five personality variables, which consist of 15 items and are therefore more prone to partial nonresponse. To assess the robustness of our findings, we additionally performed multiple imputation for the Big Five variables (see Supplement Tables A4-A6). Because the main analyses rely on complete cases, we also compared included and excluded participants. Excluded participants reported higher levels of social and emotional loneliness, were more likely to have a low level of education and to be first-generation migrants. Further, they reported lower satisfaction with their household income. The standardised differences were small to modest (|Cohen’s d| ≤ 0.23). The differences on most other characteristics, including the Big Five traits, depressed mood and anxiety disorders, were negligible (see Supplement Table A20).

Variables

Loneliness: Overall loneliness, emotional loneliness and social loneliness are our dependent variables. The measurement is based on the De Jong Gierveld Loneliness Short Scale (DJGLS), which is validated for several age groups and countries (26). Emotional loneliness is measured with the questionnaire items “(1) I experience a general sense of emptiness”; “(2) I miss having people around”; “(3) I often feel rejected”. The items “(4) There are plenty of people I can rely on when I have problems”; “(5) There are many people I can trust completely” and “(6) There are enough people I feel close to” are used to measure social loneliness (see FReDA questionnaire and scale overview [50]. The answer categories range from one to three, covering (“yes”, “more or less” and “no”). Processing the scale we follow the validated operationalisation by De Jong Gierveld [26, 55]. This entails counting the neutral (“more or less”) and positive (“yes”) answers on items 1, 2 and 3 for the emotional loneliness score. The neutral (“more or less”) and negative (“no”) answers on items 4, 5, and 6 are counted for the social loneliness score. Hence, for both dimensions the score ranges from zero (“not lonely”) to three (“intensely lonely”). The internal consistency, assessed using McDonald’s Omega, was acceptable for both dimensions (social loneliness: ω = 0.85; emotional loneliness: ω = 0.71). The overall loneliness score was calculated by summing the emotional and social loneliness scores, following the approach by De Jong Gierveld [26, 55], resulting in a total score ranging from 0 to 6. To facilitate graphical comparison of trends in emotional and social loneliness (each ranging from 0 to 3), with the trend in overall loneliness (ranging from 0 to 6), we divide the overall loneliness score by two. This rescaling results in a range from 0 ("not lonely") to 3 ("intensely lonely") across all three loneliness dimensions studied in this paper. The distribution of respondents across each value is robust for both the original overall loneliness scale (ranging from 0 to 6) and the adjusted overall loneliness scale (ranging from 0 to 3) (see Supplement Table A1). To enable comparability across the three dimensions of loneliness, we conducted an analysis using standardised (z‑score) values. The results remain robust (see Supplement Tables A12 - A14).

Associated Factors: To examine factors associated with loneliness, we analysed four distinct sets of variables, which capture socio-demographic characteristics, social relationships, health-related variables, and personality traits (Big Five). Table 1 provides an overview of these sets, their corresponding dimensions, and operationalisation. The selection of variables was guided by prior research on factors associated with overall loneliness and the existing research gaps regarding emotional and social loneliness. Our aim was to test whether these factors, well documented for overall loneliness, apply differently to emotional and social loneliness. We also checked for multicollinearity among the variables; correlations ranged from -0.35 to 0.28 (see Supplement Table A2). The mean variance inflation factor across all predictors is 1.92, suggesting minimal risk of bias due to shared variance among the predictors.

We started with the base model that includes age (in categories 18–25, 26–30, 31–35, 36–40, 41–45, 46–51) and binary gender, as both have consistently been discussed in the literature as relevant correlates of loneliness [13, 15, 17–19, 56]. We categorized age into groups rather than modelling it continuously, as this allows for a flexible representation of the age-loneliness relationship without imposing a linear functional form, given the mixed evidence on the shape of age-related loneliness patterns in prior research.

The first extended set covered socio-demographic variables, with higher socio-economic status consistently associated with lower overall loneliness [13, 17, 34, 57]. Household financial security (“Making ends meet”) was assessed using respondents’ subjective evaluation on a six-point scale, ranging from [1] “great difficulties” to [6] “very good”. Educational background was classified according to the International Standard Classification of Education (ISCED) 2011 and summarised on a three-point scale: low (ISCED 0–2), middle (ISCED 3–4), and high (ISCED 5–8). Migration was also considered, given mixed evidence on its role in loneliness across different age groups and loneliness dimensions [31–33, 35]. We distinguish between respondents with no migration background and those who are first- or second-generation migrants (see Table 1). Respondents of the first generation have immigrated to Germany themselves. Respondents of the second generation were born in Germany, but at least one parent immigrated to Germany.

The second set captured social relationship variables, repeatedly associated with lower overall loneliness [13, 16, 17, 21, 57]. These include partnership status (single, living-apart-together, cohabiting), as evidence highlights this variable as one of the factors most strongly associated with lower overall loneliness [13, 16, 17, 21, 57], while the association has been mainly described with emotional loneliness [31–33]. The role of having children remains unclear across all dimensions of loneliness [32] in the current literature and was incorporated as a yes/no variable. Furthermore, the role of respondents’ social networks (measured by whether personal matters are discussed with friends, acquaintances, neighbors, or colleagues; yes/no) is insufficiently researched. Current literature supports an association between friend-based social support and lower loneliness [39–41]; however, its association with emotional and social loneliness remains unclear. Following Granovetter’s distinction between strong and weak ties, partnership status and broader social network indicators were conceptually mapped onto different types of social relationships.

The third set included health-related variables, as both mental and physical conditions have consistently been associated with overall loneliness [7, 13, 16] and partly with emotional and social loneliness [32]. Mental health was captured through anxiety disorder (established medical diagnoses, self-reported) and depressed mood, assessed by asking about past-week experiences ([1] “never” to [4] “most/all of the time.”), which was recoded into a binary variable ([1] and [2] = [0] “no”; [3] and [4] = [1] “yes”). Overall life satisfaction was measured with the validated 11-point Likert scale [58], given that higher life satisfaction has been associated with lower loneliness [16]. However, evidence regarding its association with emotional and social loneliness is scarce [59]. Physical health was assessed based on common diseases in the target population according to the Global Burden of Disease Study [60, 61], including type 2 diabetes, hypertension, cancer, and asthma (all self-reported medical diagnoses). These indicators were included to examine associations between physical health and loneliness in young and middle adulthood, given heterogeneous and partly inconclusive evidence in prior studies [7, 13, 32], particularly with respect to diagnosis-based indicators and emotional and social loneliness dimensions. Obesity was classified using the World Health Organization’s (WHO) Body Mass Index (BMI) threshold of ≥ 30 [62]. In addition, two variables addressed perceived burden and positive experiences during the COVID-19 pandemic, as data collection occurred during the later phase of the pandemic and potential effects could not be excluded [63–67].

The fourth variable set comprised the Big Five personality traits, each of which has been associated with loneliness in previous research [45, 48]. We applied the validated Big Five Inventory – 2 Extra Short Form (BFI-2-XS) [68]. Instead of the 60-item BFI-2 scale, the BFI-2 Extra Short Form assesses each of the five personality domains (extraversion, agreeableness, conscientiousness, neuroticism and openness to experience) with three items - one per facet - resulting in a total of 15 items. Although its brevity is known to reduce internal consistency compared to the full scale (McDonald’s Omega ranging from ω 0.43 to ω 0.68 across domains), the BFI-2-XS has been validated for domain-level assessment, and our large sample size provides sufficient statistical power to detect meaningful associations [69]. Response options for all items ranged from [1] “not at all” to [5] “absolutely” (see Table 1). The full item battery can be found in the supplement (Supplement Table A3).

Table 1Overview of variable sets, dimensions and variables

Variable Sets Dimension Operationalisation

Variable Values

Base Model Age Age categories

(wave 1B)

18–25

26–30

31–35

36–40

41–45

46–51

Gender Gender (binary)

(wave 1B)

(1) men

(2) female

Socio-Demographic Variables Income Subjective evaluation of household financial security (Making ends meet)

(wave 1B)

(1) great difficulties

(6) very good

Educational background ISCED Categories

(wave 1R)

(1) low: ISCED 0–2

(2) middle: ISCED 3–4

(3) high: ISCED 5–8

Migration Background Migration Background

(wave 1B)

(0) no migration background

(1) 1st generation migrant

(2) 2nd generation migrant

Social Relationships Children Respondent has children

(wave 1B)

(0) no

(1) yes

Partnership Partnership Status

(wave 1 A)

(1) single

(2) Living Apart Together

(3) cohabitation

Social Network Discuss personal matters with friends, acquaintances, neighbours, and colleagues

(wave 1B)

(0) no

(1) yes

Health-Related Variables Mental Health Doctor has diagnosed anxiety disorder

(wave 1B)

(0) no

(1) yes

Depressed Mood

(wave 1 A)

(0) no

(1) yes

Life Satisfaction

(wave 1B)

(0) not at all satisfied

(10) completely satisfied

Physical Health Doctor has diagnosed type 2 diabetes

(wave 1B)

(0) no

(1) yes

Doctor has diagnosed hypertension

(wave 1B)

(0) no

(1) yes

Doctor has diagnosed cancer

(wave 1B)

(0) no

(1) yes

Doctor has diagnosed asthma

(wave 1B)

(0) no

(1) yes

Obesity (self-reported BMI)

(wave 1B)

(0) BMI < 30

(1) BMI > = 30

COVID-19 Pandemic Personally heavily burdened

(wave 1B)

(1) strongly disagrees

(5) strongly agree

Can also see a positive aspect

(wave 1B)

(1) strongly disagrees

(5) strongly agree

Personality Big Five Inventory Extraversion

(wave 2B)

(1) not at all

(5) absolutely

Agreeableness

(wave 2B)

(1) not at all

(5) absolutely

Conscientiousness

(wave 2B)

(1) not at all

(5) absolutely

Neuroticism

(wave 2B)

(1) not at all

(5) absolutely

Openness to new experiences

(wave 2B)

(1) not at all

(5) absolutely

Source: FReDA Panel

Analyses

The study’s research questions, hypotheses (H1- H6), and analysis plan were registered on the Open Science Framework (OSF) [49]. Analyses not specified in the registration were considered exploratory.

We employed multiple ordinary least squares (OLS) regressions to examine the occurrence of three dimensions of loneliness – overall, emotional, and social – across age groups and gender, as well as their associations with different sets of variables. As a robustness check, given the ordinal and polarised nature of the loneliness measures, we re-estimated our main models using ordered logit and computed the expected value of the outcome scale. The resulting predictions were virtually identical to those from the OLS models (Supplement Table A21). Separate regression models were estimated for each loneliness dimension. For each dimension, the variable sets were included stepwise in the following order: (1) age and gender (base model) (2), socio-demographic variables (3), social relationship variables (4), health-related variables, and (5) personality traits. Variables were entered stepwise to assess the contribution of the different variable sets. This stepwise approach allowed us to assess the unique contribution of each variable set while accounting for potential overlaps and interdependencies.

To illustrate predicted levels of loneliness by age and gender, we calculated marginal means, estimating values for each age category and sex while holding all other variables at their sample means. These estimates were used to produce the graphical trends shown in the results.

Statistics and reproducibility

The analyses were conducted using STATA 18. The code used for the analyses will be made publicly available on the OSF upon publication.

Results

Characteristics of participants

The final sample comprised 14,443 participants (mean age of 34.8 years) with a near-equal gender distribution (49.8% male, 50.2% female). 29.6% of respondents report intense overall loneliness, representing the highest score on the 0–3 loneliness scale. The prevalence of intense loneliness is higher for social loneliness (35.8%). While both overall and social loneliness are fairly evenly distributed across genders, females report intense emotional loneliness slightly more often than males, with a prevalence in the full sample of 24.3% (see Table 2).

Table 2Prevalence and Gender Composition of Individuals Experiencing Intense Loneliness (overall, emotional and social loneliness)

Full Sample Intense overall loneliness Intense emotional loneliness Intense social loneliness

29.57% 24.28% 35.80%

Gender

Female 14.13% 13.3% 17.72%

Male 15.44% 10.98% 18.08%

Percentages under gender indicate the share of individuals reporting intense loneliness in each category (overall, social, emotional). Intense loneliness corresponds to a score of 3 on a scale ranging from 0 to 3

Source: FReDA Panel, wave 1B (autumn/winter 2021/22). Weighted prevalence. Authors’ own calculations

These prevalence rates reflect extreme loneliness scores, while mean values capture the full distribution (see Table 3). Respondents’ ratings of overall loneliness (mean = 1.52) and emotional loneliness (mean = 1.49) are fairly evenly distributed across the 0 to 3 scale. Social loneliness, however, is more polarised (mean = 1.56), with many scoring 0 (31.0%) or 3 (35.8%). The frequency and mean values for variables relating to sociodemographic characteristics, social relationships, health and personality traits are listed in Table 3.

Table 3Overview of Participant Characteristics

Variables

Proportions/ Mean (SD) Min Max

Loneliness Variables Overall Loneliness Mean: 1.52 0 3

0 (Not lonely) 11.9%

0.5 17.0%

1 14.3%

1.5 14.6%

2 12.7%

2.5 12.8%

3 (Intensely Lonely) 16.8%

Emotional Loneliness Mean: 1.49 0 3

0 (Not Lonely) 22.8%

1 29.8%

2 23.1%

3 (Intensely Lonely) 24.3%

Social Loneliness Mean:1.56 0 3

0 (Not lonely) 31.0%

1 17.8%

2 15.3%

3 (Intensely Lonely) 35.9%

Base Model Age Mean: 34.77 18 51

18–25 20.1%

26–30 14.9%

31–35 17.1%

36–40 16.8%

41–45 15.5%

46–51 15.5%

Sex 0 1

Male 49.8%

Female 50.2%

Socio-Demographic Variables Subjective Evaluation of HH Financial Security 4.43 (1.26) 1 6

Educational Background 1 3

Low 10.7%

Middle 56.3%

High 33.0%

Migration Background 0 2

No Migration Background 72.1%

1. Generation Migrant 16.2%

2. Generation Migrant 11.7%

Social Relationships Children 0 1

No Children 52.6%

Kids 47.4%

Partnership Status 1 3

Single 28.9%

Living Apart Together 12.2%

Cohabitation 58.9%

Discuss Personal Matters with Social Network 0 1

No 39.3%

Yes 60.7%

Health-Related Variables Anxiety 0 1

No 89.3%

Yes 10.7%

Depressed Mood 0 1

No 85.7%

Yes 14.3%

Life Satisfaction 7.21 (1.87) 0 10

Type 2 Diabetes 0 1

No 97.4%

Yes 2.6%

Hypertension 0 1

No 89.1%

Yes 10.9%

Cancer 0 1

No 98.2%

Yes 1.8%

Asthma 0 1

No 91.1%

Yes 8.9%

Obesity 0 1

No 82.2%

Yes 17.8%

COVID-19 Pandemic: See a Positive Side 3.50 (1.13) 1 5

COVID-19 Pandemic: Negatively Affected 3.20 (1.17) 1 5

Big Five Extraversion 3.03 (0.74) 1 5

Agreeableness 3.63 (0.73) 1 5

Conscientiousness 3.65 (0.78) 1 5

Neuroticism 2.95 (0.65) 1 5

Openness to Experience 3.30 (0.79) 1 5

N = 14,443. Source: FReDA Panel, wave 1B (autumn/winter 2021/22); educational background from wave 1R, depressed mood and partnership status from wave 1A, Big Five personality traits from wave 2B (autumn/winter 2022/23). Descriptive statistics are weighted. Authors’ own calculations

Base model estimates of overall, emotional, and social loneliness

Figure 1 presents mean levels of overall, emotional, and social loneliness across age groups from the base model, which includes age and gender only. Overall loneliness was most pronounced among 18-25-year-olds (mean: 1.50; 95% CI: 1.46, 1.53, p < .001), lower in the 26–30 age group (mean: 1.38; 95% CI: 1.34, 1.42, p < .001), and somewhat higher again in older age groups, with a mean of 1.47 (95% CI: 1.42, 1.51, p < .001) in the 46-51-year-olds. Emotional loneliness was highest in the youngest group (18–25 years; mean: 1.65; 95% CI: 1.61, 1.69, p < .001) and lower in the 26–30 group, with successively lower values in subsequent age groups, reaching the lowest levels in the 46–51 years age group (mean: 1.33; 95% CI: 1.28, 1.38, p < .001). Levels were relatively stable thereafter. In contrast, social loneliness was lowest among 18-25-year-olds (mean: 1.35; 95% CI: 1.30, 1.39, p < .001) and 26-30-year-olds (mean: 1.28; 95% CI: 1.23, 1.34, p < .001), and higher in older groups, with the highest mean observed in the 46–51 age group (mean: 1.60; 95% CI: 1.55, 1.66, p < .001; see Supplement Table A7).

Factors associated with overall, emotional, and social loneliness

In the next step, the variable sets (1) socio-demographic variables (2), social relationships (3), health-related variables, and (4) personality traits were added sequentially to the base model. The stepwise regression results for all included variables are provided in Supplement Tables A8 – A10. Table 4 represents the regression coefficients and 95% CI for all variables included in the fully adjusted model, grouped by covariate domain.

Social relationships showed the most consistent negative associations with loneliness across all dimensions, with the strongest associations observed for social loneliness (e.g., social network: β: -0.32; 95% CI: -0.37, -0.28, p < .001, R² = 0.15; partnership [living apart together]: β: -0.23; 95% CI: -0.29, -0.16, p < .001, R² = 0.15; partnership [cohabitation]: β: -0.21; 95% CI: -0.26, -0.16, p < .001, R² = 0.15). Life satisfaction showed a stable, though less pronounced, negative association with all loneliness dimensions (β: -0.12; 95% CI: -0.13, -0.11, p < .001, R² = 0.20 for overall loneliness, R² = 0.19 for emotional loneliness, R² = 0.15 for social loneliness). Older age (from approximately 36 years onward) was associated with lower levels of emotional loneliness (β ranging from − 0.14 to -0.12, p < .001). Among the Big Five personality traits, agreeableness showed the most pronounced association with lower levels of social (β: -0.19; 95% CI: -0.22, -0.17, p < .001, R² = 0.15) and overall loneliness (β: -0.12; 95% CI: -0.14, -0.10, p < .001, R² = 0.20). A better subjective evaluation of household financial security was associated with lower levels of loneliness (β ranging from − 0.07 to -0.06, p < .001). Additionally, perceiving positive aspects during the COVID-19 pandemic was negatively associated with loneliness, although effect sizes were small (β ranging from − 0.04 to -0.02, p < .01).

A first-generation migration background was positively associated with higher levels of social (β: 0.31; 95% CI: 0.24, 0.38, p < .001, R² = 0.15) and overall loneliness (β: 0.21; 95% CI: 0.15, 0.26, p < .001, R² = 0.20). In contrast, second-generation migrants showed substantially smaller associations, as exemplified by social loneliness (β: 0.08; 95% CI: 0.01, 0.14, p = .014, R² = 0.15). Indicators of mental health burden (depressed mood and anxiety disorders) were consistently associated with higher levels of loneliness across all dimensions, particularly in emotional loneliness [(depressed mood: β: 0.22; 95% CI: 0.17, 0.27, p < .001, R² = 0.19); (anxiety disorder: β: 0.11; 95% CI: 0.06, 0.17, p < .001, R² = 0.19)]. Age showed a positive association with social loneliness, with stronger associations observed in older age groups (β ranging from 0.18 to 0.32, p < .001). Among the Big Five, neuroticism was positively associated with all loneliness dimensions, with the comparatively highest association with emotional loneliness (β: 0.17; 95% CI: 0.14, 0.20, p < .001, R² = 0.19). In addition, personal burden experienced during the COVID-19 pandemic was positively associated with emotional loneliness (β: 0.13; 95% CI: 0.11, 0.14, p < .001, R² = 0.19) and, to a lesser extent, overall loneliness (β: 0.07; 95% CI: 0.06, 0.08, p < .001, R² = 0.20). Finally, female gender was positively associated with emotional (β: 0.09; 95% CI: 0.05, 0.12, p < .001, R² = 0.19) and overall loneliness (β: 0.04; 95% CI: 0.01, 0.07, p < .01, R² = 0.20).

Associations of educational attainment, parental status, and most physical health indicators with any dimension of loneliness were small, with estimates generally close to zero and CIs including the null value (Table 4, Supplement Tables A8 – A10). With regard to explained variance (R²), the health status variable set contributed notably to the variance in overall and emotional loneliness. Further details are presented in Supplement Tables A8 – A10.

Overall, the most consistent and comparatively pronounced associations across all three loneliness dimensions were observed for partnership status, social network, depressed mood, and life satisfaction, although their magnitude varied across the dimensions. Several associations were more dimension-specific: Social network, partnership status and first-generation migration background were particularly strongly associated with social loneliness, whereas depressed mood, partnership status and life satisfaction were more strongly associated with emotional loneliness. Across the fully adjusted models, the comparatively large associations were in the range of approximately 0.2 to 0.3 points on the loneliness scale ranging from 0 to 3, while many other statistically detectable associations were below 0.1 scale points and were therefore small in substantive terms.

Table 4Factors Associated with Overall, Emotional, and Social Loneliness (n = 14,443)

Variable Sets Predictor Overall loneliness β [95% CI] Emotional loneliness β [95% CI] Social loneliness β [95% CI]

Base Model Age 18-25 Ref. Ref. Ref.

26-30 0.00 [-0.05; 0.05] -0.03 [-0.09; 0.03] 0.04 [-0.03; 0.10]

31-35 0.06 [0.01; 0.12]* -0.06 [-0.12; 0.00] 0.18 [0.11; 0.25]**

36-40 0.06 [0.00; 0.12]* -0.12 [-0.19; -0.06]** 0.24 [0.17; 0.32]**

41-45 0.08 [0.02; 0.14]* -0.14 [-0.21; -0.08]** 0.29 [0.21; 0.37]**

46-51 0.09 [0.03; 0.15]* -0.14 [-0.21; -0.07]** 0.32 [0.24; 0.40]**

Gender men Ref. Ref. Ref.

female 0.04 [0.01; 0.07]* 0.09 [0.05; 0.12]** -0.00 [-0.04; 0.04]

Socio-Demographic Variables Income 1-6, higher = greater satisfaction -0.06 [-0.08; -0.05]** -0.06 [-0.07; -0.04]** -0.07 [-0.09; -0.05]**

Educational background low: ISCED 0-2 Ref. Ref. Ref.

middle: ISCED 3-4 0.00 [-0.09; 0.09] 0.05 [-0.05; 0.15] -0.04 [-0.16; 0.07]

high: ISCED 5-8 -0.08 [-0.17; 0.01] -0.03 [-0.13; 0.07] -0.12 [-0.24; -0.01]*

Migration Background no migration background Ref. Ref. Ref.

1st generation migrant 0.21 [0.15; 0.26]** 0.10 [0.05; 0.16]** 0.31 [0.24; 0.38]**

2nd generation migrant 0.05 [0.00; 0.09]* 0.02 [-0.03; 0.07]* 0.08 [0.01; 0.14]*

Social Relationships Children no Ref. Ref. Ref.

yes 0.01 [-0.03; 0.05] -0.01 [-0.05; 0.04] 0.02 [-0.03; 0.07]

Partnership single Ref. Ref. Ref.

Living Apart Together -0.18 [-0.23; -0.13]** -0.14 [-0.19; -0.08]** -0.23 [-0.29; -0.16]**

cohabitation -0.18 [-0.23; -0.14]** -0.16 [-0.21; -0.12]** -0.21 [-0.26; -0.15]**

Social Network no Ref. Ref. Ref.

yes -0.20 [-0.23; -0.16]** -0.07 [-0.10; -0.03]** -0.32 [-0.37; -0.28]**

Health-Related Variables Mental Health Anxiety: no Ref. Ref. Ref.

Anxiety: yes 0.11 [0.06; 0.16]** 0.11 [0.06; 0.17]** 0.10 [0.04; 0.17]*

Depressed mood: no Ref. Ref. Ref.

Depressed mood: yes 0.19 [0.14; 0.24]** 0.22 [0.17; 0.27]** 0.16 [0.10; 0.22]**

0-10, higher = greater satisfaction -0.12 [-0.13; -0.11]** -0.12 [-0.13; -0.11]** -0.12 [-0.13; -0.11]**

Physical Health Diabetes type 2: no Ref. Ref. Ref.

Diabetes type 2: yes -0.05 [-0.15; 0.04] -0.06 [-0.16; 0.05] -0.05 [-0.18; 0.07]

Hypertension: no Ref. Ref. Ref.

Hypertension: yes 0.02 [-0.04; 0.07] -0.01 [-0.07; 0.05] 0.04 [-0.02; 0.11]

Cancer: no Ref. Ref. Ref.

Cancer: yes 0.03 [-0.08; 0.14] 0.04 [-0.08; 0.16] 0.02 [-0.12; 0.16]

Asthma: no Ref. Ref. Ref.

Asthma: yes 0.05 [-0.00; 0.10] 0.04 [-0.01; 0.10] 0.06 [-0.01; 0.12]

Obesity (BMI<30): no Ref. Ref. Ref.

Obesity (BMI≥30): yes 0.00 [-0.04; 0.05] -0.02 [-0.06; 0.03] 0.02 [-0.03; 0.08]

COVID-19 Pandemic Positive Aspects: 1-5, higher = greater disagreement -0.03 [-0.04; -0.02]** -0.02 [-0.04; -0.01]* -0.04 [-0.05; -0.02]*

Personal Burden: 1-5, higher = greater disagreement 0.07 [0.06; 0.08]** 0.13 [0.11; 0.14]** 0.01 [-0.01; 0.03]

Personality Big Five Inventory 1-5, higher = greater neuroticism 0.12 [0.10; 0.15]** 0.17 [0.14; 0.20]** 0.07 [0.04; 0.10]**

1-5, higher = greater openness -0.02 [-0.04; 0.00] 0.01 [-0.01; 0.03] -0.04 [-0.07; -0.02]*

1-5, higher = greater agreeableness -0.12 [-0.14; -0.10]** -0.05 [-0.07; -0.02]** -0.19 [-0.22; -0.17]**

1-5, higher = greater conscientiousness -0.05 [-0.07; -0.03]** -0.07 [-0.09; -0.04]** -0.04 [-0.07; -0.02]**

1-5, higher = greater extraversion -0.05 [-0.07; -0.03]** -0.02 [-0.05; 0.00] -0.08 [-0.11; -0.05]**

Socio-demographic variables Social relationships Health status Personality (Big Five)

Coefficients and 95% CI (confidence intervals) are based on Model 4, which includes socio-demographic variables, social relationships, health status, and personality traits

Source: FReDA Panel, wave 1B (autumn/winter 2021/22); educational background from wave 1R, depressed mood and partnership status from wave 1A, Big Five personality traits from wave 2B (autumn/winter 2022/23). Authors’ own calculations

* p<0.05, ** p<0.001

Full model estimates of overall, emotional, and social loneliness

Next, we stratified each loneliness dimension again by age in the fully adjusted model to analyse whether the inclusion of the variables was associated with differences in loneliness across age groups. The results show that the previously observed higher mean of overall loneliness among individuals aged 18–25 years in the base model was less pronounced in the fully adjusted model, resulting in only minor age-related differences (ranging from 1.39 to 1.48, p < .001; Fig. 2, Supplement Table A11). However, differences across age groups in emotional and social loneliness remained, though at a reduced magnitude. Emotional loneliness was highest among 18-25-year-olds (mean: 1.51; 95% CI: 1.47–1.55, p < .001) and showed lower means in older age groups, with the lowest mean observed in the 46-51-year age group (mean: 1.37; 95% CI: 1.32–1.42, p < .001). Conversely, social loneliness was lowest in the youngest age group (mean: 1.28; 95% CI: 1.23–1.33, p < .001) and was higher in older groups, with the highest mean observed in the 46-51-year group (mean: 1.60; 95% CI: 1.54–1.65, p < .001).

Loneliness by gender: overall, emotional, and social loneliness

Gender-stratified associations were generally small, and estimates were largely similar between women and men (Fig. 3, Supplement Table A15). For emotional loneliness there is a tendency for women to report higher levels than men, with significant differences in the 31–35 and 36–40 age groups. In gender-specific models of associated predictors, associations were generally similar between men and women, with the clearest gender difference observed for ‘openness to experience’. Openness showed a small negative association with social loneliness in men (β: -0.09; 95% CI: -0.12, -0.05, p < .001), whereas the estimate was close to zero in women (β: -0.01; 95% CI: -0.05, 0.03, p = .64). Overall, gender-specific differences in predictor associations were limited, with openness to experience representing the clearest isolated difference. Detailed results are available in Supplement Tables A16-17.

Loneliness by age: overall, emotional, and social loneliness

For each loneliness dimension, we additionally tested age-group differences in the three most relevant predictors identified in the full model: depressed mood, partnership status, and life satisfaction for emotional loneliness, and migration background, partnership status, and social network for social loneliness. Across these predictors, heterogeneity by age group emerged only for partnership status: For emotional loneliness, being in a partnership (LAT and cohabitation) was associated with significantly lower loneliness only among respondents in middle adulthood (age groups 41–45 and 46–51), with no significant difference between partnered and single individuals in the younger age groups. For social loneliness, by contrast, partnered respondents showed significantly lower loneliness not only in the oldest age group [46–51] but also among 26-30- and 31-35-year-olds. Overall, age-related heterogeneity in predictor associations was limited and emerged primarily for partnership status, whereas the remaining associations were broadly consistent across age groups. Detailed results are available in Supplement Tables A18-A19.

Discussion

This study provides evidence on the age- and gender-related distribution of overall, emotional and social loneliness and their associated variables among adults aged 18 to 51 in Germany. Three key findings emerge. First, intense loneliness was common with prevalence rates ranging from 24.3% for emotional loneliness to 29.6% for overall loneliness and 35.8% for social loneliness. Second, emotional and social loneliness showed opposing age-related patterns: emotional loneliness was lower in older age groups, whereas social loneliness was higher. In contrast, overall loneliness varied little across age groups. Third, social relationships and mental health-related factors showed the most consistent associations across loneliness dimensions, although the magnitude of these associations varied. Social networks and first-generation migration background showed particularly pronounced associations with social loneliness, whereas depressed mood was most strongly associated with emotional loneliness. These findings underscore the conceptual distinction between loneliness dimensions and the importance of considering their correlates. Given the large sample size, we emphasise the magnitude and precision of estimated associations, as reflected by regression coefficients (β) and their 95% CIs, rather than statistical significance alone. Small coefficients are interpreted cautiously even when their CIs exclude the null value [70, 71].

Approximately one-quarter to one-third of young and middle-aged adults in our study reported high levels of emotional, social or overall loneliness, assessed using the validated DJGLS [26, 55]. Although comparisons with previous research are limited by differences in measures and study designs, longitudinal evidence from Germany indicates that loneliness increased during the COVID-19 pandemic, particularly among adolescents and young adults. Although levels have since declined, they appear to remain above pre-pandemic levels [72]. As our data were collected during the COVID-19 pandemic, the observed prevalence may partly reflect these exceptional circumstances. Nevertheless, previous research suggests that associations between loneliness and sociodemographic, social and psychological factors are also observed beyond the pandemic context [7].

Regarding age-related differences, overall loneliness varied little across the age groups examined, consistent with several large-scale studies reporting relatively small differences in loneliness among middle-aged adults [13, 14, 21, 73]. However, other studies have reported higher loneliness in young adulthood than in older age groups [13, 14, 73]. In our analyses, age-related differences in overall loneliness substantially attenuated after adjustment for social relationship variables (Model 2). The relatively small differences observed across age groups are broadly aligned with the meta-analysis by Mund et al. [21], which synthesised 75 longitudinal studies across the adult life course. Importantly, these authors identified the lack of differentiation between emotional and social loneliness as a key limitation of the existing literature. By distinguishing between these dimensions, our study revealed opposing age-related patterns that were not apparent when examining overall loneliness alone. Emotional loneliness was lower in older age groups, whereas social loneliness was higher. Early adulthood is often characterised by transitions such as leaving home, starting work or education, and forming new social ties, alongside processes of identity development and increasing autonomy. These circumstances may coincide with differences in social connectedness, perceived exclusion and unmet relational needs [19, 74], although our cross-sectional data do not allow us to examine such processes directly. Previous studies differentiating loneliness dimensions during young and middle adulthood are limited. Existing evidence generally indicates lower emotional loneliness in older age groups [31, 32, 75, 76], consistent with our findings. In contrast, evidence on social loneliness is more heterogeneous, with studies reporting higher levels in older groups [32, 76], while others suggest relative stability [75] or lower levels [31]. These findings illustrate how analyses of overall loneliness alone may obscure distinct age-related patterns in emotional and social loneliness.

Gender differences in overall loneliness were small, with women reporting slightly higher levels than men. This aligns with prior research indicating that gender differences, when observed, tend to be small and inconsistent in direction [13, 16, 77–79], potentially varying by measurement instrument used (e.g. single-item vs. multi-item scales) [11]. A meta-analysis by Maes et al. [78], including 751 effect sizes from nearly 400,000 individuals, reported negligible gender differences (Hedges’ g = 0.07). In our dimension-specific analyses, emotional loneliness was slightly higher among women aged 31–40; however, the magnitude of this difference was small. Previous findings on gender differences in emotional and social loneliness are similarly mixed, with some studies reporting higher emotional loneliness among women [31, 33, 35], while others report higher social loneliness [32] and some identifying no consistent differences [48].

Social relationships emerged as particularly important correlates of loneliness. Social networks and partnership status were among the factors most consistently associated with lower loneliness across all three dimensions. This is in line with previous large-scale research highlighting the relevance of social and relational resources throughout adulthood [16, 17, 32, 40, 41]. However, the magnitude of these associations differed across dimensions. Social networks showed the strongest inverse association with social loneliness, a more modest negative association with overall loneliness and a relatively small association with emotional loneliness. Similarly, partnership status was more strongly associated with social loneliness than with overall or emotional loneliness. This contrasts with studies reporting stronger inverse associations between partnership and emotional loneliness [31–33, 35], indicating that associations between partnership and loneliness dimensions may differ across populations and contexts. Having children, by contrast, was not associated with lower loneliness, consistent with other studies reporting no systematic associations between parenthood and loneliness [17].

Migration background was another characteristic showing pronounced dimension-specific associations. Consistent with previous studies reporting higher loneliness among individuals with a migration background [35–37, 80], first-generation migration background was associated with higher social and overall loneliness, whereas its association with emotional loneliness was weaker. A Dutch study similarly identified a positive association between first-generation migration background and overall loneliness [35]. Recent longitudinal evidence from Germany further suggests heterogeneity in associations between migration background, loneliness and mental health, particularly by age at migration. For example, pronounced patterns have been reported among women who migrated to Germany after the age of 18 [38]. Possible contextual explanations discussed in the literature include the loss of close familial and cultural ties to one’s country of origin [81], experiences of exclusion, racism or discrimination [82] and barriers to establishing or maintaining broader social networks in the receiving country [38]. However, these mechanisms could not be examined in our study. Furthermore, our distinction between first- and second-generation migration background does not capture differences in age at migration, duration of residence, country of origin, language proficiency or cultural and social context. These characteristics should be considered when interpreting the observed associations. Notably, associations with second-generation migration background were substantially smaller, with estimates close to zero, especially for women.

Socioeconomic characteristics showed comparatively modest associations with loneliness. Previous studies linked higher educational attainment [13, 32, 33, 57] and higher income [13, 16, 34] to lower loneliness. In our study, however, associations between educational attainment and loneliness were generally small, with estimates close to zero across most dimensions. Household financial security showed a consistent but modest inverse association with all three loneliness dimensions.

Mental health-related factors were also consistently associated with loneliness, particularly its emotional dimension. Previous research has identified better physical and mental health [13, 16, 17, 32, 35, 57, 79] as well as higher life satisfaction and subjective well-being [12, 16, 57] as correlates of lower loneliness. However, relatively few studies have examined these associations separately for emotional and social loneliness while incorporating self-reported clinical diagnoses [32, 35]. In our analyses, depressed mood and anxiety disorders were meaningfully associated with higher loneliness across all dimensions, with the strongest associations observed for emotional loneliness, consistent with previous findings on psychological distress [32]. Life satisfaction was also consistently inversely associated with all loneliness dimensions. In contrast, individual physical health conditions, including hypertension, type 2 diabetes, asthma and obesity, showed generally small associations, with estimates close to zero and CIs frequently including the null value. This aligns with previous research reporting small or inconsistent associations between most physical health conditions and loneliness, particularly among younger populations [35]. For example, Hofman et al. (2022) reported weak associations of poor perceived health with both emotional and social loneliness and of higher body mass index with emotional loneliness, whereas associations with chronic diseases and functional limitations were small, inconsistent or non-significant [32]. Taken together, these findings suggest that physical health conditions may be less strongly associated with loneliness than psychosocial and mental health-related factors in young and middle adulthood. Stronger associations at later ages remain possible, particularly given the relatively low prevalence of severe physical illnesses in our comparatively young panel sample [60] and previous evidence on loneliness across the life course [9].

Further, our findings align with previous research on personality traits (Big Five), confirming that extraversion, agreeableness, conscientiousness, and (partially) openness are associated with lower levels of loneliness, while neuroticism remains consistently positively associated with all loneliness dimensions [45, 48]. Notably, in contrast to earlier studies, agreeableness showed the strongest negative association with loneliness in the fully adjusted models, particularly for social loneliness.

The observed differences between loneliness dimensions can be considered in light of Granovetter’s theory of strong and weak ties [29], which distinguishes emotionally close relationships associated with attachment needs from broader social connections associated with social integration and belonging. Consistent with this framework, social network ties showed the strongest association with social loneliness in our study. Migration background further supports a weak-tie interpretation: Having a first-generation migration background was more strongly associated with social than emotional loneliness. This pattern may be consistent with greater barriers to establishing or maintaining broader weak-tie networks in the host society, although the underlying mechanisms could not be examined in the present study. Individual characteristics also align with this pattern, as agreeableness was strongly associated with lower social loneliness but only weakly with emotional loneliness, suggesting that cooperative interpersonal styles may be conducive to maintaining broader weak-tie networks. In contrast, emotional loneliness was less clearly associated with strong-tie mechanisms. Partnership status was not more associated with emotional loneliness than with social loneliness, indicating that the presence of a partner alone may not be sufficient to offset emotional disconnection. Instead, emotional loneliness showed its strongest associations with psychological factors; depressed mood was particularly strongly associated with this dimension. Together, these findings suggest that emotional and social loneliness are associated with partially distinct relational and psychological factors, reinforcing the importance of a multidimensional, theory-informed approach.

As data collection took place during the COVID-19 pandemic (autumn/winter 2021/22), we additionally considered participants’ perceived pandemic-related experiences. Perceiving the pandemic as a personal burden was associated with higher emotional and overall loneliness, whereas reporting positive aspects was weakly associated with lower loneliness across all dimensions. These findings are broadly consistent with previous research linking pandemic-related stress to higher loneliness and resilience-related factors to lower loneliness [66]. However, the observed associations were considerably smaller than those of the main social and psychological correlates identified in our study. Thus, while pandemic-related experiences may be relevant to interpreting loneliness during the study period, social relationships and mental health-related factors remained the more pronounced correlates in our analyses.

Our findings have implications for future loneliness research and public health strategies. Loneliness is increasingly recognised as a public health concern by institutions such as the WHO [4], the UK government [83], and Germany’s national strategy [72, 84]. Nevertheless, evidence on the effectiveness of interventions remains limited. Psychological and social interventions have shown promising findings, particularly among older adults [73, 85–88], while evidence on digital approaches is mixed [87, 89, 90]. The methodological quality of existing intervention studies is often limited, and relatively few approaches explicitly distinguish between loneliness dimensions or are tailored to different age groups [73, 88]. In this context, the dimension-specific associations observed in our study suggest that distinguishing emotional and social loneliness may be relevant when developing and evaluating future interventions. Strengthening individuals’ “social convoy” through meaningful relationships, as discussed in theoretical approaches by De Jong Gierveld and Fokkema, has been proposed as a potential approach to supporting resilience [91].

Taken together, our findings underscore the multifactorial and context-sensitive nature of loneliness in young and mid-adulthood. Individuals without a partner or with limited social networks, those reporting mental health burdens, and people with a first-generation migration background may warrant particular attention in future research. The observed distinct age-related patterns of emotional and social loneliness also suggest that research may benefit from considering loneliness dimensions alongside age, social context, relationship status and psychological resources. This is particularly relevant given established associations between loneliness and adverse health outcomes, including depression, cardiovascular disease and all-cause mortality. Importantly, our observational findings do not establish whether interventions targeting the characteristics identified in this study would reduce loneliness. The effectiveness of dimension-specific or targeted approaches therefore requires investigation in appropriately designed intervention studies.

A major strength of our study is the multidimensional assessment of loneliness, distinguishing between emotional and social aspects, which is particularly relevant for young and middle adulthood. Drawing on large, representative panel data, we incorporated a broad set of validated measures (e.g., loneliness, Big Five personality traits, life satisfaction), along with detailed sociodemographic variables, health conditions, and migration background. However, some limitations need to be acknowledged. First, the cross-sectional design of our study limits causal interpretation and precludes differentiation between age and cohort effects, restricting conclusions about developmental trajectories over time [92]. Temporal ordering cannot be established, and bidirectional relationships are plausible, particularly for social relationships and mental health, as loneliness may itself be associated with social engagement, perceived relationship quality, and mental health. Additionally, the measures were not all collected concurrently. In particular, depressed mood and partnership status were assessed in Wave 1 A and the Big Five personality traits in Wave 2B, whereas the loneliness outcomes were assessed in Wave 1B. Although personality traits are relatively stable, they are not fixed, and loneliness may itself be related to later changes in self-reported personality. Depressed mood or partnership status may also fluctuate more strongly between subwaves. Nevertheless, identifying factors consistently associated with different loneliness dimensions can help characterise groups reporting higher or lower loneliness and inform hypotheses for future longitudinal and intervention research. In addition, the large, age-diverse sample (spanning nearly four decades) enables a comprehensive analysis of a wide range of relevant factors. Second, while many associations reached statistical significance, their magnitude varied substantially, and several estimates were small or close to zero. Explained variances (R²) were modest, which is common in population-based studies of psychosocial phenomena [93]. Third, additional limitations include the use of self-reported measures for income and BMI, which may not fully reflect objective experiences. In this vein, due to the ultra-short nature of the BFI-2-XS, internal consistency for the Big Five domains was modest. However, this is an expected trade-off between brevity and reliability [69]. Fourth, as data were collected during the COVID-19 pandemic, prevalence may reflect exceptional circumstances, which should be considered when interpreting generalisability. Additionally, as data collection for Wave 1 took place in spring/summer (subwave A) and autumn/winter (subwave B), and as these two subwaves were combined into one wave, different levels of pandemic severity at a macro level may have made the salience of the pandemic different across subwaves during data collection. Fifth, our analyses are based on complete cases and therefore exclude a significant proportion of the initial sample (N = 5,727 out of 20,216), primarily due to missing data on the Big Five items. Although multiple imputation analyses for the Big Five variables point to similar results, the analytic sample is not a random subset of the initial sample. Respondents who were excluded reported somewhat higher levels of loneliness, lower levels of education and lower satisfaction with their household income, and were more often first-generation migrants. While the differences were small to modest, they suggest that selection bias in the analytical sample cannot be ruled out. Consequently, mean levels of loneliness in our analytical sample may be slightly underestimated. Since the variance of the loneliness measures is almost identical in both groups, we consider it unlikely that the range will be substantially restricted, but we cannot exclude the possibility that some associations will differ in the excluded population. Lastly, as emotional and social loneliness items differ in wording direction (negative vs. positive), the results may be partly influenced by different response styles or mood activation, although we expect such bias to be small given the prior validation of the scale.

Conclusion

In conclusion, this study advances the understanding of loneliness in young and middle adulthood by highlighting its multidimensional nature. Distinguishing between emotional and social loneliness revealed opposing age-related patterns that were not apparent when considering overall loneliness. Social relationships and mental health-related characteristics were consistently associated with loneliness, although the magnitude of associations differed across dimensions. Higher loneliness was particularly evident among individuals with limited social networks, those reporting mental health conditions, and those with a first-generation migration background. These findings highlight the relevance of considering loneliness dimensions alongside their psychosocial context in future research and the development and evaluation of public health strategies. Given the cross-sectional design, longitudinal and intervention studies are needed to clarify temporal relationships and evaluate potential approaches to addressing loneliness.

Supplementary Information

References

1.Perlman D Peplau L Toward a social psychology of loneliness Personal relationships disorder 1981 3 31 43 Perlman D, Peplau L. Toward a social psychology of loneliness. Personal relationships disorder. 1981;3:31–43.

2.Cacioppo JT Patrick W Loneliness Human nature and the need for social connection 2009 New York, LondonW. W. NortonCacioppo JT, Patrick W, Loneliness. Human nature and the need for social connection. New York, London: W. W. Norton; 2009.

3.eClinicalMedicineThe epidemic of loneliness eClinicalMedicine 2023 66 102395 38192594 10.1016/j.eclinm.2023.102395 eClinicalMedicine. The epidemic of loneliness. eClinicalMedicine. 2023;66:102395.10.1016/j.eclinm.2023.102395 PMC10772224 38192594

4.World Health Organization. WHO launches commission to foster social connection. Geneva. 2023. Available from: URL: https://www.who.int/news/item/15-11-2023-who-launches-commission-to-foster-social-connection. Cited 14 Feb 2025.

5.Surkalim DL Luo M Eres R Gebel K van Buskirk J Bauman A The prevalence of loneliness across 113 countries: systematic review and meta-analysis BMJ (Clinical Res ed 2022 376 e067068 1 17 10.1136/bmj-2021-067068 PMC8826180 35140066 Surkalim DL, Luo M, Eres R, Gebel K, van Buskirk J, Bauman A, et al. The prevalence of loneliness across 113 countries: systematic review and meta-analysis. BMJ (Clinical Res ed. 2022;376(e067068):1–17.10.1136/bmj-2021-067068 PMC8826180 35140066

6.Wang J Mann F Lloyd-Evans B Ma R Johnson S Associations between loneliness and perceived social support and outcomes of mental health problems: a systematic review BMC Psychiatry 2018 18 156 10.1186/s12888-018-1736-5 29843662 Wang J, Mann F, Lloyd-Evans B, Ma R, Johnson S. Associations between loneliness and perceived social support and outcomes of mental health problems: a systematic review. BMC Psychiatry. 2018;18:156.10.1186/s12888-018-1736-5 PMC5975705 29843662

7.Barjaková M Garnero A d’Hombres B Risk factors for loneliness: A literature review Soc Sci Med 2023 334 116163 37625251 10.1016/j.socscimed.2023.116163 Barjaková M, Garnero A, d’Hombres B. Risk factors for loneliness: A literature review. Soc Sci Med. 2023;334:116163.10.1016/j.socscimed.2023.116163 PMC10523154 37625251

8.Ernst M Loneliness as a transdiagnostic risk factor for mental and physical health in adulthood — state of the evidence, research gaps, and ways forward Curr Opin Behav Sci 2025 66 101614 Ernst M. Loneliness as a transdiagnostic risk factor for mental and physical health in adulthood — state of the evidence, research gaps, and ways forward. Curr Opin Behav Sci. 2025;66:101614.

9.Wang F Gao Y Han Z Yu Y Long Z Jiang X A systematic review and meta-analysis of 90 cohort studies of social isolation, loneliness and mortality Nat Hum Behav 2023 7 8 1307 19 10.1038/s41562-023-01617-6 37337095 Wang F, Gao Y, Han Z, Yu Y, Long Z, Jiang X, et al. A systematic review and meta-analysis of 90 cohort studies of social isolation, loneliness and mortality. Nat Hum Behav. 2023;7(8):1307–19.10.1038/s41562-023-01617-6 37337095

10.Abaei E Martin P Unraveling the dynamics of loneliness and cognition in late life: a cross-lagged panel model Front Public Health 2024 12 1425403 39171310 10.3389/fpubh.2024.1425403 Abaei E, Martin P. Unraveling the dynamics of loneliness and cognition in late life: a cross-lagged panel model. Front Public Health. 2024;12:1425403.10.3389/fpubh.2024.1425403 PMC11335492 39171310

11.Nicolaisen M Thorsen K Who are lonely? Loneliness in different age groups (18–81 years old), using two measures of loneliness Int J Aging Hum Dev 2014 78 3 229 57 25265679 10.2190/AG.78.3.b Nicolaisen M, Thorsen K. Who are lonely? Loneliness in different age groups (18–81 years old), using two measures of loneliness. Int J Aging Hum Dev. 2014;78(3):229–57.10.2190/AG.78.3.b 25265679

12.Mund M Maes M Drewke PM Gutzeit A Jaki I Qualter P Would the Real Loneliness Please Stand Up? The Validity of Loneliness Scores and the Reliability of Single-Item Scores Assessment 2023 30 4 1226 48 35246009 10.1177/10731911221077227 PMC10149889 Mund M, Maes M, Drewke PM, Gutzeit A, Jaki I, Qualter P. Would the Real Loneliness Please Stand Up? The Validity of Loneliness Scores and the Reliability of Single-Item Scores. Assessment. 2023;30(4):1226–48.10.1177/10731911221077227 PMC10149889 35246009

13.Graham EK Beck ED Jackson K Yoneda T McGhee C Pieramici L Do We Become More Lonely With Age? A Coordinated Data Analysis of Nine Longitudinal Studies Psychol Sci 2024 35 6 579 96 38687352 10.1177/09567976241242037 Graham EK, Beck ED, Jackson K, Yoneda T, McGhee C, Pieramici L, et al. Do We Become More Lonely With Age? A Coordinated Data Analysis of Nine Longitudinal Studies. Psychol Sci. 2024;35(6):579–96.10.1177/09567976241242037 PMC11788814 38687352

14.Pyle E, Evans D. Loneliness – What characteristics and circumstances are associated with feeling lonely? Office for National Statistics, UK. Available from: URL: https://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/articles/lonelinesswhatcharacteristicsandcircumstancesareassociatedwithfeelinglonely/2018-04-10.  Cited accessed 21 May 2025.

15.Lasgaard M, Friis K, Shevlin M. ‘‘Where are all the lonely people?’’ A population-based study of high-risk groups across the life span. Social Psychiatry and Psychiatric Epidemiology 2016; (51):1373–84. Available from: URL: https://link.springer.com/article/10.1007/s00127-016-1279-3.10.1007/s00127-016-1279-3 27571769

16.Klein EM Zenger M Tibubos AN Ernst M Reiner I Schmalbach B Loneliness and its relation to mental health in the general population: Validation and norm values of a brief measure J Affect Disorders Rep 2021 4 100120 Klein EM, Zenger M, Tibubos AN, Ernst M, Reiner I, Schmalbach B, et al. Loneliness and its relation to mental health in the general population: Validation and norm values of a brief measure. J Affect Disorders Rep. 2021;4:100120.

17.Hawkley LC Buecker S Kaiser T Luhmann M Loneliness from Young Adulthood to Old Age: Explaining Age Differences in Loneliness Int J Behav Dev 2022 46 1 39 49 35001993 10.1177/0165025420971048 PMC8734589 Hawkley LC, Buecker S, Kaiser T, Luhmann M. Loneliness from Young Adulthood to Old Age: Explaining Age Differences in Loneliness. Int J Behav Dev. 2022;46(1):39–49.10.1177/0165025420971048 PMC8734589 35001993

18.Luhmann M, Hawkley L. Age differences in loneliness from late adolescence to oldest old age. Developmental Psychology. 2016; 52(6):943–59. Available from: URL: https://pubmed.ncbi.nlm.nih.gov/27148782/.10.1037/dev0000117 PMC8015413 27148782

19.Barreto M Victor C Hammond C Eccles A Richins MT Qualter P Loneliness around the world: Age, gender, and cultural differences in loneliness Pers Indiv Differ 2021 169 110066 10.1016/j.paid.2020.110066 PMC7768187 33536694 Barreto M, Victor C, Hammond C, Eccles A, Richins MT, Qualter P. Loneliness around the world: Age, gender, and cultural differences in loneliness. Pers Indiv Differ. 2021;169:110066.10.1016/j.paid.2020.110066 PMC7768187 33536694

20.van den Broek T Lam J Potente C Do middle-aged and older people underreport loneliness? experimental evidence from the Netherlands Eur J Ageing 2024 21 1 29 10.1007/s10433-024-00826-w 39367963 van den Broek T, Lam J, Potente C. Do middle-aged and older people underreport loneliness? experimental evidence from the Netherlands. Eur J Ageing. 2024;21(1):29.10.1007/s10433-024-00826-w PMC11456021 39367963

21.Mund M Freuding MM Möbius K Horn N Neyer FJ The Stability and Change of Loneliness Across the Life Span: A Meta-Analysis of Longitudinal Studies Personality social Psychol review: official J Soc Personality Social Psychol Inc 2020 24 1 24 52 10.1177/1088868319850738 PMC6943963 31179872 Mund M, Freuding MM, Möbius K, Horn N, Neyer FJ. The Stability and Change of Loneliness Across the Life Span: A Meta-Analysis of Longitudinal Studies. Personality social Psychol review: official J Soc Personality Social Psychol Inc. 2020;24(1):24–52.10.1177/1088868319850738 PMC6943963 31179872

22.Buecker S Denissen JJA Luhmann M A propensity-score matched study of changes in loneliness surrounding major life events J Personal Soc Psychol 2021 121 3 669 90 10.1037/pspp0000373 33119390 Buecker S, Denissen JJA, Luhmann M. A propensity-score matched study of changes in loneliness surrounding major life events. J Personal Soc Psychol. 2021;121(3):669–90.10.1037/pspp0000373 33119390

23.Mehta CM Mehta CM I’m Trying to Navigate these Multiple Roles’: Established Adulthood as the Rush Hour of Life Established Adulthood 2025 New YorkOxford University Press1 15 Mehta CM. I’m Trying to Navigate these Multiple Roles’: Established Adulthood as the Rush Hour of Life. In: Mehta CM, editor. Established Adulthood. New York: Oxford University Press; 2025. pp. 1–15.

24.Bittman M, Rice JM. Labour Industry: J social economic relations work. The Spectre Of Overwork: An Analysis Of Trends Between 1974 And 1997 Using Australian Time-Use Diaries. 2002;12(3):5–25.

25.Diabaté S Kleinschlömer P Kriechel L Kleinschrot L Mohr E Ludwig-Walz H Loneliness and mental health across family constellations in Germany: Evidence of a stable burden J Affect Disorders Rep 2026 24 101033 Diabaté S, Kleinschlömer P, Kriechel L, Kleinschrot L, Mohr E, Ludwig-Walz H. Loneliness and mental health across family constellations in Germany: Evidence of a stable burden. J Affect Disorders Rep. 2026;24:101033.

26.Jong Gierveld J van Tilburg T The De Jong Gierveld short scales for emotional and social loneliness: tested on data from 7 countries in the UN generations and gender surveys Eur J Ageing 2010 7 2 121 30 10.1007/s10433-010-0144-6 20730083 Jong Gierveld J, van Tilburg T. The De Jong Gierveld short scales for emotional and social loneliness: tested on data from 7 countries in the UN generations and gender surveys. Eur J Ageing. 2010;7(2):121–30.10.1007/s10433-010-0144-6 PMC2921057 20730083

27.Weiss RS. Loneliness: the experience of emotional and social isolation. MIT Press. 1973.

28.Walsh BE Rottenberg J Schlauch RC Why loneliness requires a multidimensional approach: a critical narrative review Nat Mental Health 2025 3 2 175 84 10.1038/s44220-024-00382-3 Walsh BE, Rottenberg J, Schlauch RC. Why loneliness requires a multidimensional approach: a critical narrative review. Nat Mental Health. 2025;3(2):175–84.

29.Granovetter MS The Strength of Weak Ties Am J Sociol 1973 78 6 1360 80 Granovetter MS. The Strength of Weak Ties. Am J Sociol. 1973;78(6):1360–80.

30.Cacioppo JT Hawkley LC Perceived social isolation and cognition Trends Cogn Sci 2009 13 10 447 54 19726219 10.1016/j.tics.2009.06.005 Cacioppo JT, Hawkley LC. Perceived social isolation and cognition. Trends Cogn Sci. 2009;13(10):447–54.10.1016/j.tics.2009.06.005 PMC2752489 19726219

31.Kenny L Shevlin M Cloitre M Determinants of Social and Emotional Loneliness in a Nationally Representative Sample of US Adults Eur J Behav Sci 2022 5 3 36 49 Kenny L, Shevlin M, Cloitre M. Determinants of Social and Emotional Loneliness in a Nationally Representative Sample of US Adults. Eur J Behav Sci. 2022;5(3):36–49.

32.Hofman A Overberg RI Schoenmakers EC Adriaanse MC Social and emotional loneliness in a large sample of Dutch adults aged 19–65: Associations with risk factors Psychiatry Res 2022 313 114602 35544985 10.1016/j.psychres.2022.114602 Hofman A, Overberg RI, Schoenmakers EC, Adriaanse MC. Social and emotional loneliness in a large sample of Dutch adults aged 19–65: Associations with risk factors. Psychiatry Res. 2022;313:114602.10.1016/j.psychres.2022.114602 35544985

33.Fierloos IN Tan SS Williams G Alhambra-Borrás T Koppelaar E Bilajac L Socio-demographic characteristics associated with emotional and social loneliness among older adults BMC Geriatr 2021 21 1 114 10.1186/s12877-021-02058-4 33563228 Fierloos IN, Tan SS, Williams G, Alhambra-Borrás T, Koppelaar E, Bilajac L, et al. Socio-demographic characteristics associated with emotional and social loneliness among older adults. BMC Geriatr. 2021;21(1):114.10.1186/s12877-021-02058-4 PMC7871533 33563228

34.Hajek A, König HH. Association of income poverty and wealth with loneliness among young and middle-aged adults: findings based on a nationwide, representative sample from Germany. J Public Health (Berl). 2026. 10.1007/s10389-026-02813-0 .

35.Kate RLF Bilecen B Steverink N A Closer Look at Loneliness: Why Do First-Generation Migrants Feel More Lonely Than Their Native Dutch Counterparts? Gerontologist 2020 60 2 291 301 31944240 10.1093/geront/gnz192 Kate RLF, Bilecen B, Steverink N. A Closer Look at Loneliness: Why Do First-Generation Migrants Feel More Lonely Than Their Native Dutch Counterparts? Gerontologist. 2020;60(2):291–301.10.1093/geront/gnz192 PMC7039375 31944240

36.Löbel L-M Kröger H Tibubos AN How Migration Status Shapes Susceptibility of Individuals’ Loneliness to Social Isolation Int J public health 2022 67 1604576 36561278 10.3389/ijph.2022.1604576 Löbel L-M, Kröger H, Tibubos AN. How Migration Status Shapes Susceptibility of Individuals’ Loneliness to Social Isolation. Int J public health. 2022;67:1604576.10.3389/ijph.2022.1604576 PMC9763294 36561278

37.Klok J van Tilburg TG Suanet B Fokkema T Huisman M National and transnational belonging among Turkish and Moroccan older migrants in the Netherlands: protective against loneliness? Eur J Ageing 2017 14 4 341 51 10.1007/s10433-017-0420-9 29180940 Klok J, van Tilburg TG, Suanet B, Fokkema T, Huisman M. National and transnational belonging among Turkish and Moroccan older migrants in the Netherlands: protective against loneliness? Eur J Ageing. 2017;14(4):341–51.10.1007/s10433-017-0420-9 PMC5684037 29180940

38.Shi S Loi S How loneliness mediates the association between migration status and health trajectories: Longitudinal evidence from Germany J Migr Health 2026 13 100408 42006127 10.1016/j.jmh.2026.100408 Shi S, Loi S. How loneliness mediates the association between migration status and health trajectories: Longitudinal evidence from Germany. J Migr Health. 2026;13:100408.10.1016/j.jmh.2026.100408 PMC13091140 42006127

39.Aslan Y Koçak O Kaya AB Alkhulayfi AMA Gómez-Salgado J Yıldırım M Roles of loneliness and life satisfaction in the relationship between perceived friend social support, positive feelings about the future and loss of motivation Acta Psychol 2025 259 105315 10.1016/j.actpsy.2025.105315 40706181 Aslan Y, Koçak O, Kaya AB, Alkhulayfi AMA, Gómez-Salgado J, Yıldırım M. Roles of loneliness and life satisfaction in the relationship between perceived friend social support, positive feelings about the future and loss of motivation. Acta Psychol. 2025;259:105315.10.1016/j.actpsy.2025.105315 40706181

40.Schafer MH Malo JS Brown BM McGourty ML What accompanies companionship? Reassessing personal networks and loneliness among young and older adults Social Networks 2026 86 211 28 Schafer MH, Malo JS, Brown BM, McGourty ML. What accompanies companionship? Reassessing personal networks and loneliness among young and older adults. Social Networks. 2026;86:211–28.

41.Schrempft S Baysson H Dumont R Stringhini S Guessous I Nehme M Predictors of perceived social support and loneliness over time: A population-based study in Switzerland, 2021–2025 Public Health 2026 258 106413 42462558 10.1016/j.puhe.2026.106413 Schrempft S, Baysson H, Dumont R, Stringhini S, Guessous I, Nehme M. Predictors of perceived social support and loneliness over time: A population-based study in Switzerland, 2021–2025. Public Health. 2026;258:106413.10.1016/j.puhe.2026.106413 42462558

42.Hilliard DD, Wootton RE, Sallis HM, van de Weijer MP, Treur JL, Qualter P, et al. Investigating relationships between loneliness, social isolation and health. Nat Commun. 2026;17:5840. 10.1038/s41467-026-74758-7 .PMC13373220 42457663

43.Misiak B Tyburski E Samochowiec A Samochowiec J Unravelling longitudinal associations of social and emotional loneliness with social isolation and mental health outcomes: a cross-lagged panel network analysis Epidemiol Psychiatr Sci 2026 35 e1 41479370 10.1017/S2045796025100383 Misiak B, Tyburski E, Samochowiec A, Samochowiec J. Unravelling longitudinal associations of social and emotional loneliness with social isolation and mental health outcomes: a cross-lagged panel network analysis. Epidemiol Psychiatr Sci. 2026;35:e1.10.1017/S2045796025100383 PMC12816935 41479370

44.Beller J, Graßhoff J, Sperlich S. Functional limitations and loneliness in middle-aged and older adults: differentiating emotional loneliness and social loneliness. BMC Public Health. 2026;26:1758. 10.1186/s12889-026-27997-8 .PMC13224435 42219485

45.Buecker S Maes M Denissen JJA Luhmann M Loneliness and the Big Five Personality Traits: A Meta–Analysis Eur J Pers 2020 34 1 8 28 Buecker S, Maes M, Denissen JJA, Luhmann M. Loneliness and the Big Five Personality Traits: A Meta–Analysis. Eur J Pers. 2020;34(1):8–28.

46.Maher P, Dragostinov Y, Vainik U, Cooper A, Allik J, Realo A, et al. Loneliness and personality: noise- and bias-free true correlations between loneliness and the Big Five personality domains. J Pers. 2026;94:907–19. 10.1111/jopy.70048 .PMC13540311 41531026

47.Schellenberg L, Bleidorn W, Haehner P. Identifying robust longitudinal transactions between loneliness and the Big Five personality traits. J Pers. 2026. 10.1111/jopy.70084 .42318811

48.Erevik EK Vedaa Ø Pallesen S Hysing M Sivertsen B The five-factor model’s personality traits and social and emotional loneliness: Two large-scale studies among Norwegian students Pers Indiv Differ 2023 207 112115 Erevik EK, Vedaa Ø, Pallesen S, Hysing M, Sivertsen B. The five-factor model’s personality traits and social and emotional loneliness: Two large-scale studies among Norwegian students. Pers Indiv Differ. 2023;207:112115.

49.Open Science Framework. Emotional and Social Loneliness in Young and Middle-Aged Adults: Anonymized study registration. 2025. Available from: URL: https://osf.io/n4t2v/overview?view_only=5338401008fe4f3fbf1c7d9e3eb75e01

50.Bujard M, Gummer T, Hank K, Neyer FJ, Pollak R, Schneider NF et al. FReDA – The German Family Demography Panel Study. Cologne. 2025.

51.Schneider NF Bujard M Wolf C Gummer T Hank K Neyer FJ Family Research and Demographic Analysis (FReDA): Evolution, Framework, Objectives, and Design of The German Family-Demographic Panel Study Comp Popul Stud 2021 46 149 86 Schneider NF, Bujard M, Wolf C, Gummer T, Hank K, Neyer FJ. Family Research and Demographic Analysis (FReDA): Evolution, Framework, Objectives, and Design of The German Family-Demographic Panel Study. Comp Popul Stud. 2021;46:149–86.

52.Hank K, Gummer T, Bujard M, Neyer FJ, Pollak R, Spieß CK, et al. A new data infrastructure for family research and demographic analysis: the German Family Demography Panel Study (FReDA). EurSociol Rev. 2024;40:1–13. 10.1093/esr/jcae019 .

53.Roberts BW DelVecchio WF The rank-order consistency of personality traits from childhood to old age: a quantitative review of longitudinal studies Psychol Bull 2000 126 1 3 25 10668348 10.1037/0033-2909.126.1.3 Roberts BW, DelVecchio WF. The rank-order consistency of personality traits from childhood to old age: a quantitative review of longitudinal studies. Psychol Bull. 2000;126(1):3–25.10.1037/0033-2909.126.1.3 10668348

54.Cobb-Clark DA Schurer S The stability of big-five personality traits Econ Lett 2012 115 1 11 5 Cobb-Clark DA, Schurer S. The stability of big-five personality traits. Econ Lett. 2012;115(1):11–5.

55.Gierveld JDJ van Tilburg T A 6-Item Scale for Overall, Emotional, and Social Loneliness Res Aging 2006 28 5 582 98 Gierveld JDJ, van Tilburg T. A 6-Item Scale for Overall, Emotional, and Social Loneliness. Res Aging. 2006;28(5):582–98.

56.Soest T Luhmann M Hansen T Gerstorf D Development of loneliness in midlife and old age: Its nature and correlates J Personal Soc Psychol 2020 118 2 388 406 10.1037/pspp0000219 30284871 Soest T, Luhmann M, Hansen T, Gerstorf D. Development of loneliness in midlife and old age: Its nature and correlates. J Personal Soc Psychol. 2020;118(2):388–406.10.1037/pspp0000219 30284871

57.Franssen T Stijnen M Hamers F Schneider F Age differences in demographic, social and health-related factors associated with loneliness across the adult life span (19–65 years): a cross-sectional study in the Netherlands BMC Public Health 2020 20 1 1118 32758200 10.1186/s12889-020-09208-0 Franssen T, Stijnen M, Hamers F, Schneider F. Age differences in demographic, social and health-related factors associated with loneliness across the adult life span (19–65 years): a cross-sectional study in the Netherlands. BMC Public Health. 2020;20(1):1118.10.1186/s12889-020-09208-0 PMC7409622 32758200

58.Cheung F Lucas RE Assessing the validity of single-item life satisfaction measures: results from three large samples Qual life research: Int J Qual life aspects Treat care rehabilitation 2014 23 10 2809 18 10.1007/s11136-014-0726-4 PMC4221492 24890827 Cheung F, Lucas RE. Assessing the validity of single-item life satisfaction measures: results from three large samples. Qual life research: Int J Qual life aspects Treat care rehabilitation. 2014;23(10):2809–18.10.1007/s11136-014-0726-4 PMC4221492 24890827

59.Salimi A Social-Emotional Loneliness and Life Satisfaction Procedia - Social Behav Sci 2011 29 292 5 Salimi A. Social-Emotional Loneliness and Life Satisfaction. Procedia - Social Behav Sci. 2011;29:292–5.

60.Porst M Lippe E Leddin J Anton A Wengler A Breitkreuz J The Burden of Disease in Germany at the National and Regional Level Deutsches Ärzteblatt international 2022 119 46 785 92 36350160 10.3238/arztebl.m2022.0314 Porst M, Lippe E, Leddin J, Anton A, Wengler A, Breitkreuz J, et al. The Burden of Disease in Germany at the National and Regional Level. Deutsches Ärzteblatt international. 2022;119(46):785–92.10.3238/arztebl.m2022.0314 PMC9902892 36350160

61.Ward ZJ Goldie SJ Global Burden of Disease Study 2021 estimates: implications for health policy and research Lancet (London England) 2024 403 10440 1958 9 38642567 10.1016/S0140-6736(24)00812-2 Ward ZJ, Goldie SJ. Global Burden of Disease Study 2021 estimates: implications for health policy and research. Lancet (London England). 2024;403(10440):1958–9.10.1016/S0140-6736(24)00812-2 38642567

62.World Health Organization, Obesity. and overweight; 2025. Available from: URL: https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight. Cited 23 May 2025.

63.Ernst M Niederer D Werner AM Czaja SJ Mikton C Ong AD Loneliness before and during the COVID-19 pandemic: A systematic review with meta-analysis Am Psychol 2022 77 5 660 77 35533109 10.1037/amp0001005 Ernst M, Niederer D, Werner AM, Czaja SJ, Mikton C, Ong AD, et al. Loneliness before and during the COVID-19 pandemic: A systematic review with meta-analysis. Am Psychol. 2022;77(5):660–77.10.1037/amp0001005 PMC9768682 35533109

64.Rebechi A Lepinteur A Clark AE Rohde N Vögele C D’Ambrosio C Loneliness during the COVID-19 pandemic: Evidence from five European countries Econ Hum Biol 2024 55 101427 39167991 10.1016/j.ehb.2024.101427 Rebechi A, Lepinteur A, Clark AE, Rohde N, Vögele C, D’Ambrosio C. Loneliness during the COVID-19 pandemic: Evidence from five European countries. Econ Hum Biol. 2024;55:101427.10.1016/j.ehb.2024.101427 39167991

65.Köster F Lipps O How loneliness increased among different age groups during COVID-19: a longitudinal analysis Eur J Ageing 2024 21 1 2 10.1007/s10433-023-00798-3 38170323 Köster F, Lipps O. How loneliness increased among different age groups during COVID-19: a longitudinal analysis. Eur J Ageing. 2024;21(1):2.10.1007/s10433-023-00798-3 PMC10764689 38170323

66.Reinwarth AC Ernst M Gerstorf D Brähler E Wild PS Münzel T Results of a multi-perspective examination of loneliness trajectories and its determinants in German adults during the COVID-19 pandemic Sci Rep 2025 15 1 17889 10.1038/s41598-025-00678-z 40410266 Reinwarth AC, Ernst M, Gerstorf D, Brähler E, Wild PS, Münzel T, et al. Results of a multi-perspective examination of loneliness trajectories and its determinants in German adults during the COVID-19 pandemic. Sci Rep. 2025;15(1):17889.10.1038/s41598-025-00678-z PMC12102375 40410266

67.Entringer T. Epidemiologie von Einsamkeit in Deutschland: KNE Expertise 4/2022. Berlin: Kompetenznetz Einsamkeit. 2022. Available from: URL: https://kompetenznetz-einsamkeit.de/wp-content/uploads/2022/07/KNE_Expertise04_220629.pdf.

68.Soto CJ John OP The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power J Personal Soc Psychol 2017 113 1 117 43 10.1037/pspp0000096 27055049 Soto CJ, John OP. The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power. J Personal Soc Psychol. 2017;113(1):117–43.10.1037/pspp0000096 27055049

69.Soto CJ John OP Short and extra-short forms of the Big Five Inventory–2: The BFI-2-S and BFI-2-XS J Res Pers 2017 68 69 81 Soto CJ, John OP. Short and extra-short forms of the Big Five Inventory–2: The BFI-2-S and BFI-2-XS. J Res Pers. 2017;68:69–81.

70.Funder DC Ozer DJ Evaluating Effect Size in Psychological Research: Sense and Nonsense Adv Methods Practices Psychol Sci 2019 2 2 156 68 Funder DC, Ozer DJ. Evaluating Effect Size in Psychological Research: Sense and Nonsense. Adv Methods Practices Psychol Sci. 2019;2(2):156–68.

71.Götz FM Gosling SD Rentfrow PJ Small Effects: The Indispensable Foundation for a Cumulative Psychological Science Perspect Psychol science: J Association Psychol Sci 2022 17 1 205 15 10.1177/1745691620984483 34213378 Götz FM, Gosling SD, Rentfrow PJ. Small Effects: The Indispensable Foundation for a Cumulative Psychological Science. Perspect Psychol science: J Association Psychol Sci. 2022;17(1):205–15.10.1177/1745691620984483 34213378

72.Schobin J, Arriagada C, Gibson-Kunze M, Einsamkeitsbarometer. 2024: Langzeitentwicklung von Einsamkeit in Deutschland; 2024. Available from: URL: https://www.bmfsfj.de/resource/blob/240528/5a00706c4e1d60528b4fed062e9debcc/einsamkeitsbarometer-2024-data.pdf.

73.Lasgaard M, Løvschall C, Qualter P, Laustsen LM, Lim MH, Maindal HT, et al. Are loneliness interventions effective in reducing loneliness? A meta-analytic review of 128 studies. Eur J Public Health. 2022;32(Suppl 3):ckac129.266. 10.1093/eurpub/ckac129.266 .

74.Qualter P, Vanhalst J, Harris R, van Roekel E, Lodder G, Bangee M et al. Loneliness Across the Life Span. Perspectives on Psychological Science. 2015; (10(2)):250–64. Available from: URL: https://journals.sagepub.com/doi/epub/10.1177/1745691615568999.10.1177/1745691615568999 25910393

75.Manoli A McCarthy J Ramsey R Estimating the prevalence of social and emotional loneliness across the adult lifespan Sci Rep 2022 12 1 21045 10.1038/s41598-022-24084-x 36473900 Manoli A, McCarthy J, Ramsey R. Estimating the prevalence of social and emotional loneliness across the adult lifespan. Sci Rep. 2022;12(1):21045.10.1038/s41598-022-24084-x PMC9727057 36473900

76.van Tilburg TG Residual loneliness in the Netherlands after the COVID-19 pandemic: An application of the single interrupted time series design with pre-, peri- and post-pandemic observations Public Health 2024 237 238 44 39461031 10.1016/j.puhe.2024.10.023 van Tilburg TG. Residual loneliness in the Netherlands after the COVID-19 pandemic: An application of the single interrupted time series design with pre-, peri- and post-pandemic observations. Public Health. 2024;237:238–44.10.1016/j.puhe.2024.10.023 39461031

77.Nicolaisen M Thorsen K Gender Differences in Loneliness Over Time: A 15-Year Longitudinal Study of Men and Women in the Second Part of Life Int J Aging Hum Dev 2024 98 1 103 32 37563853 10.1177/00914150231194243 Nicolaisen M, Thorsen K. Gender Differences in Loneliness Over Time: A 15-Year Longitudinal Study of Men and Women in the Second Part of Life. Int J Aging Hum Dev. 2024;98(1):103–32.10.1177/00914150231194243 37563853

78.Maes M Qualter P Vanhalst J van den Noortgate W Goossens L Gender Differences in Loneliness across the Lifespan: A Meta–Analysis Eur J Pers 2019 33 6 642 54 Maes M, Qualter P, Vanhalst J, van den Noortgate W, Goossens L. Gender Differences in Loneliness across the Lifespan: A Meta–Analysis. Eur J Pers. 2019;33(6):642–54.

79.Buecker S Mund M Chwastek S Sostmann M Luhmann M Is Loneliness in Emerging Adults Increasing Over Time? A Preregistered Cross-Temporal Meta-Analysis and Systematic Review Psychol Bull 2021 147 8 787 805 34898234 10.1037/bul0000332 Buecker S, Mund M, Chwastek S, Sostmann M, Luhmann M. Is Loneliness in Emerging Adults Increasing Over Time? A Preregistered Cross-Temporal Meta-Analysis and Systematic Review. Psychol Bull. 2021;147(8):787–805.10.1037/bul0000332 34898234

80.Hajek A König H-H Migrationshintergrund und Einsamkeit im mittleren und hohen Alter in Deutschland Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2024 67 10 1137 43 39112746 10.1007/s00103-024-03923-4 Hajek A, König H-H. Migrationshintergrund und Einsamkeit im mittleren und hohen Alter in Deutschland. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 2024;67(10):1137–43.10.1007/s00103-024-03923-4 PMC11424693 39112746

81.Castañeda H Over-foreignization or unused potential? A critical review of migrant health in Germany and responses toward unauthorized migration Soc Sci Med 2012 74 6 830 8 21788099 10.1016/j.socscimed.2011.05.024 Castañeda H. Over-foreignization or unused potential? A critical review of migrant health in Germany and responses toward unauthorized migration. Soc Sci Med. 2012;74(6):830–8.10.1016/j.socscimed.2011.05.024 21788099

82.Kajikhina K Koschollek C Bozorgmehr K Sarma N Hövener C Rassismus und Diskriminierung im Kontext gesundheitlicher Ungleichheit – ein narratives Review Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2023 66 10 1099 108 37735190 10.1007/s00103-023-03764-7 Kajikhina K, Koschollek C, Bozorgmehr K, Sarma N, Hövener C. Rassismus und Diskriminierung im Kontext gesundheitlicher Ungleichheit – ein narratives Review. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 2023;66(10):1099–108.10.1007/s00103-023-03764-7 PMC10539181 37735190

83.Pimlott N The ministry of loneliness Can family physician Medecin de famille canadien 2018 64 3 166 PMC5851382 29540374 Pimlott N. The ministry of loneliness. Can family physician Medecin de famille canadien. 2018;64(3):166.PMC5851382 29540374

84.Bundesministerium für Familie, Senioren, Frauen und Jugend. Strategie der Bundesregierung gegen Einsamkeit. Berlin. 2023. Available from: URL: https://www.bmfsfj.de/resource/blob/234584/9c0557454d1156026525fe67061e292e/2023-strategie-gegen-einsamkeit-data.pdf. Cited 23 May 2025.

85.Beckers A, Buecker S, Casabianca E, Nurminen M. Effectiveness of interventions tackling loneliness. EUR 31313 EN. Luxembourg: Publications Office of the European Union; 2022. 10.2760/277109 .

86.Blodgett JM, Tiley K, Harkness F, Musella M. What works to reduce loneliness: a rapid systematic review of 101 interventions. J Public Health Policy. 2025;46:245–68. 10.1057/s41271-025-00561-1 .PMC12119363 40050496

87.Hansen T Nes RB Hynek K Nilsen TS Reneflot A Stene-Larsen K Tackling social disconnection: an umbrella review of RCT-based interventions targeting social isolation and loneliness BMC Public Health 2024 24 1 1917 10.1186/s12889-024-19396-8 39020331 Hansen T, Nes RB, Hynek K, Nilsen TS, Reneflot A, Stene-Larsen K, et al. Tackling social disconnection: an umbrella review of RCT-based interventions targeting social isolation and loneliness. BMC Public Health. 2024;24(1):1917.10.1186/s12889-024-19396-8 PMC11256365 39020331

88.Hickin N Käll A Shafran R Sutcliffe S Manzotti G Langan D The effectiveness of psychological interventions for loneliness: A systematic review and meta-analysis Clin Psychol Rev 2021 88 102066 34339939 10.1016/j.cpr.2021.102066 Hickin N, Käll A, Shafran R, Sutcliffe S, Manzotti G, Langan D. The effectiveness of psychological interventions for loneliness: A systematic review and meta-analysis. Clin Psychol Rev. 2021;88:102066.10.1016/j.cpr.2021.102066 34339939

89.Seewer N Skoko A Käll A Andersson G Luhmann M Berger T Efficacy of an Internet-based self-help intervention with human guidance or automated messages to alleviate loneliness: a three-armed randomized controlled trial Sci Rep 2024 14 1 6569 10.1038/s41598-024-57254-0 38503870 Seewer N, Skoko A, Käll A, Andersson G, Luhmann M, Berger T, et al. Efficacy of an Internet-based self-help intervention with human guidance or automated messages to alleviate loneliness: a three-armed randomized controlled trial. Sci Rep. 2024;14(1):6569.10.1038/s41598-024-57254-0 PMC10951227 38503870

90.Jin W Liu Y Yuan S Bai R Li X Bai Z The Effectiveness of Technology-Based Interventions for Reducing Loneliness in Older Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials Front Psychol 2021 12 711030 34955948 10.3389/fpsyg.2021.711030 Jin W, Liu Y, Yuan S, Bai R, Li X, Bai Z. The Effectiveness of Technology-Based Interventions for Reducing Loneliness in Older Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Front Psychol. 2021;12:711030.10.3389/fpsyg.2021.711030 PMC8692663 34955948

91.de Jong Gierveld J Fokkema T Strategies to prevent loneliness Addressing loneliness: Coping, prevention and clinical interventions 2015 New York, LondonRoutledge Taylor & Francis Group218 30 de Jong Gierveld J, Fokkema T. Strategies to prevent loneliness. Addressing loneliness: Coping, prevention and clinical interventions. Researching social psychology. Volume 1. New York, London: Routledge Taylor & Francis Group; 2015. pp. 218–30.

92.Kraemer HC Yesavage JA Taylor JL Kupfer D How can we learn about developmental processes from cross-sectional studies, or can we? Am J Psychiatry 2000 157 2 163 71 10.1176/appi.ajp.157.2.163 10671382 Kraemer HC, Yesavage JA, Taylor JL, Kupfer D. How can we learn about developmental processes from cross-sectional studies, or can we? Am J Psychiatry. 2000;157:163–71. 10.1176/appi.ajp.157.2.163 .10671382

93.Ozili PK. The Acceptable R-Square in Empirical Modelling for Social Science Research. SSRN Journal. 2022.