Zum Inhalt

Association of hugs with depression, anxiety, and suicidal ideation: findings based on data from the general adult population in Germany

  • Open Access
  • 25.10.2025
  • Original Article

Abstract

Aim

To investigate the association between the frequency of hugs and probable depression, probable anxiety, and suicidal ideation in the German adult population.

Subject and methods

Cross-sectional data were used from the German general adult population aged 18 to 74 years (representative in terms of sex, age, and federal state) with n = 3270 that were obtained in an online survey in January 2025. The frequency of daily hugs served as the independent variable. Probable depression and anxiety were quantified using the Patient Health Questionnaire-9 (PHQ-9) or Generalized Anxiety Disorder-7 (GAD-7), respectively. Suicidal ideation was assessed using the final item of the PHQ-9. Logistic regression was used to examine associations. Robustness checks were also conducted.

Results

Hugging others (on average one and two to three; compared to individuals not hugging others) on a daily basis is associated with lower odds of probable depression (e.g., hugging one individual: OR 0.65; 95% CI, 0.51 to 0.83), probable anxiety (e.g., hugging one individual: OR 0.73; 95% CI, 0.56 to 0.95), and suicidal ideation (e.g., hugging one individual: OR 0.66; 95% CI, 0.52 to 0.85). Hugging on average two to three individuals daily was also associated with significantly lower odds for all outcomes, whereas hugging four or more individuals was only significantly associated with lower odds of probable depression.

Conclusion

Hugging one to three individuals each day may assist in preventing poor mental health. However, future longitudinal evidence is needed to confirm this.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Introduction

Depression, anxiety, and suicidal ideation pose a tremendous challenge for societies and economies (GBD 2019 Mental Disorders Collaborators 2022). All of them occur frequently (Hajek and König 2021; Hajek et al. 2022a; Kino et al. 2022) and are associated with unfavorable factors such as unsteady employment biographies, social withdrawal, loneliness, physical illnesses, and suicide risk (Campbell et al. 2022; Celano and Huffman 2011; Hajek et al. 2025, 2024; Hakulinen et al. 2019; Khan et al. 2002). Moreover, they place a high economic burden on the healthcare system (Bock et al. 2017; Buczak-Stec et al. 2022; Hohls et al. 2018). They can also be a great encumbrance for friends and families (Radfar et al. 2014; Skundberg-Kletthagen et al. 2014). Therefore, it is important to identify the factors associated with them.
Thus far, there have been numerous studies examining the factors associated with depression, anxiety, and suicidal ideation (Gastaldon et al. 2022; Pelton et al. 2021; Yuan et al. 2022). For instance, socioeconomic factors such as financial difficulties, lifestyle-related factors such as physical activity, and health-related factors such as self-rated health have all been linked to the aforementioned mental health outcomes (Hajek et al. 2022b; Östberg and Nordin 2022; Stickley et al. 2023; Wanjau et al. 2023).
However, as yet, far less is known about the association of hugging with depression, anxiety, and suicidal ideation. A hug occurs when one person physically embraces another individual, often as a sign of greeting, affection or comfort (Forsell and Åström 2012). When individuals are sad, a hug may reduce negative emotions (Gordon et al. 2010). Regular hugs may also strengthen bonds and can express affection (Forsell and Åström 2012). As greetings, they may signify closeness and friendship (Forsell and Åström 2012). A recent qualitative study also emphasized the need to be hugged of patients with substance use disorder after hospital discharge (Incze et al. 2025). A quantitative study (Rosenberg et al. 2021) showed that kissing or hugging either a family member or a partner almost every day (compared to not at all) were both significantly associated with a lower prevalence of probable depression in adjusted regression analyses. That study used data from a generalizable sample of adults (spanning an age range from 18 to 94 years) in the United States during the beginning of the COVID-19 pandemic (Rosenberg et al. 2021). However, quantitative studies are completely lacking on the association of hugging with a broader range of mental health outcomes (in terms of depression, anxiety and suicidal ideation) in the post-pandemic period (i.e., in a time period when there are no longer any contact restrictions). Hence, the aim of this study was to investigate the association of the frequency of hugs with probable depression, probable anxiety, and suicidal ideation in the German adult population. Such knowledge is important for better characterizing individuals at risk of depression, anxiety, and suicidal ideation. This is also relevant because those mental health outcomes may contribute to chronic illnesses and mortality (Pan et al. 2011; Walker et al. 2015).

Methods

Sample

Data were collected through an online survey (January 2025). Overall, n = 3270 individuals aged 18 to 74 years and residing in Germany took part in the survey. The only inclusion criteria were that participants had to be within the specified age range and living in Germany. The survey was administered by Bilendi, which is a reputable, International Organization for Standardization (ISO)-certified market research firm. Quota sampling was employed to ensure that the sample accurately represented the German adult population in terms of sex, age, and federal state.
All participants provided informed consent before taking part in the study. Ethical approval was granted by the Psychological Ethics Committee of the University Medical Center Hamburg-Eppendorf (number: LPEK-0849). All methods were performed in accordance with the relevant guidelines and regulations.

Dependent variables

The well-known Patient Health Questionnaire-9 (PHQ-9) (Kroenke et al. 2001) was used to quantify probable depression. A final score was generated by summing all nine items. This score varies from 0 to 27, whereby higher values denote more depressive symptoms. The established cutoff score of ≥ 10 was used to indicate probable depression (Kroenke et al. 2001). In our study, Cronbach’s alpha was 0.90 (McDonald’s omega = 0.90).
The widely used Generalized Anxiety Disorder-7 (GAD-7) scale (Spitzer et al. 2006) was employed to measure probable anxiety. Summing up all seven items, the final score thus ranges from 0 to 21, whereby higher values denote more anxiety symptoms. As suggested by the scale’s developers, a cutoff score of ≥ 10 was used to indicate probable anxiety (Spitzer et al. 2006). In this study, Cronbach’s alpha and McDonald’s omega were both 0.92.
The last item (item 9) of the aforementioned Patient Health Questionnaire (PHQ-9) (Kroenke et al. 2001) was used to quantify the frequency of suicidal ideation in the previous 2 weeks. The four response options for this item were as follows: “not at all,” “several days,” “more than half the days,” or “nearly every day.” As is common (Huarcaya-Victoria et al. 2025; Tassone et al. 2024), responses were dichotomized (0 = “not at all” or 1 = at least “several days”). This single item has been used extensively to quantify suicidal ideation in previous research (for an overview see Kim et al. 2021). It is also worth noting that this form of assessment predicts higher suicide risk across several groups (Kim et al. 2021; Louzon et al. 2016; Rossom et al. 2017; Simon et al. 2013).

Key independent variable: frequency of hugs

The average daily number of individuals hugged served as a measure of hugging frequency. Aligning with previous research (Hajek and König 2024), participants could indicate a range from 0 to 99 individuals. Four categories were established (see Hajek and König 2024): the first category comprised individuals who reported zero hugs (indicating they did not engage in hugging on average each day), while the second category included those who hugged on average one individual per day. The third category encompassed individuals who hugged on average two or three individuals daily, and the fourth category consisted of individuals who hugged on average four or more individuals per day.

Covariates

The selection of covariates was made based on previous research and grounded in theoretical considerations (Gastaldon et al. 2022; Hajek et al. 2022b; Yuan et al. 2022). Sociodemographic covariates included age, sex (male, female, diverse), federal state (separating the 16 federal states in Germany), marital status (single, divorced, widowed, or cohabiting—either married or in a partnership, living separately but married, or partnered), level of education (ISCED-97 framework: low, medium or high; UNESCO 2006), employment status (full-time, retired, other), and religious affiliation (no religion, Christianity, Islam, or other faiths). Lifestyle factors considered were the frequency of physical activity (no physical activity, less than 1 h per week, 1–2 h per week, 2–4 h per week, more than 4 h per week), alcohol intake (daily, several times a week, once a week, 1–3 times monthly, less often, or never), smoking history (never smoked, former smoker, occasional smoker, or daily smoker), and health-conscious diet (extent to which attention is paid to having a health-conscious diet, with four categories: not at all, a little, strongly, very strongly). Health-related covariates included self-rated health quantified by a single item (from 1 = very poor to 5 = very good), and a count of chronic conditions (based on 15 conditions such as stroke, diabetes, or heart disease).

Statistical analysis

First, the sample characteristics were calculated among the total sample and also stratified by the three mental health outcomes. The p-values for the group differences are based on independent t-tests or chi-square tests, as appropriate. Following this, both unadjusted and adjusted logistic regression analyses were undertaken. In the adjusted models, the sociodemographic variables were entered as the initial covariates. Thereafter, the lifestyle-related variables were added as additional covariates, followed by the inclusion of the health-related covariates.
In a robustness check for probable depression, the ninth item of the PHQ-9 was removed to avoid overlapping mental health outcomes, creating the PHQ-8 scale (ranging from 0 to 24, with higher scores reflecting more depressive symptoms). For the PHQ-8, a cut-point of ≥ 10 is also recommended given that suicidal thoughts are fairly uncommon in previous general adult population samples, and the effect of this item’s removal on scoring is often minor (Kroenke et al. 2009). We thus used this cut-point in our analysis. In a second robustness check, the number of children living in one’s own household (ranging from 0 to 9) and the living situation (living alone in a private household, living with a partner in a private household, living with relatives in a private household, living with friends/acquaintances in a private household, living in assisted living, living in a nursing home) were added as covariates to the main model (to take account of hugs within the household/family in particular).
To assess internal consistency, McDonald’s omega was calculated (using the “omegacoef” command in Stata 18). The significance threshold was set at p < 0.05. All analyses were carried out using Stata Now 19.5, MP-Parallel Edition (StataCorp, College Station, TX, USA).

Results

Sample characteristics

The sample characteristics are presented in Table 1 (with n = 3270 individuals). The mean age of the sample was 47 years (SD = 15.3 years, varying from 18 to 74 years), and slightly more than 50% were female.
Table 1
Characteristics of the sample (total sample and also stratified by the outcomes, n = 3270)
Variables
Total
Absence of probable depression
Presence of probable depression
p-value
Absence of probable anxiety
Presence of probable anxiety
p-value
Absence of suicidal ideation
Presence of suicidal ideation
 
n (%)
3270 (100.0)
2343 (71.7)
927 (28.3)
 
2553 (78.1)
717 (21.9)
 
2458 (75.2)
812 (24.8)
 
Sex: N (%)
   
0.01
  
0.07
  
 < 0.01
 Men
1614 (49.4)
1195 (51.0)
419 (45.2)
 
1286 (50.4)
328 (45.7)
 
1174 (47.8)
440 (54.2)
 
 Women
1647 (50.4)
1142 (48.7)
505 (54.5)
 
1261 (49.4)
386 (53.8)
 
1278 (52.0)
369 (45.4)
 
 Diverse
9 (0.3)
6 (0.3)
3 (0.3)
 
6 (0.2)
3 (0.4)
 
6 (0.2)
3 (0.4)
 
Age: Mean (SD)
47.0 (15.3)
49.6 (14.8)
40.3 (14.4)
 < 0.001
49.0 (14.9)
39.7 (14.2)
 < 0.001
49.1 (14.9)
40.4 (14.6)
 < 0.001
Marital situation: N (%)
   
 < 0.001
  
 < 0.001
  
 < 0.001
 Single
908 (27.8)
579 (24.7)
329 (35.5)
 
656 (25.7)
252 (35.1)
 
623 (25.3)
285 (35.1)
 
 Divorced
277 (8.5)
208 (8.9)
69 (7.4)
 
227 (8.9)
50 (7.0)
 
213 (8.7)
64 (7.9)
 
 Widowed
103 (3.1)
76 (3.2)
27 (2.9)
 
89 (3.5)
14 (2.0)
 
84 (3.4)
19 (2.3)
 
 Married, cohabiting with spouse
1867 (57.1)
1406 (60.0)
461 (49.7)
 
1500 (58.8)
367 (51.2)
 
1451 (59.0)
416 (51.2)
 
 Married, not cohabiting with spouse
115 (3.5)
74 (3.2)
41 (4.4)
 
81 (3.2)
34 (4.7)
 
87 (3.5)
28 (3.4)
 
Educational level: N (%)
   
 < 0.001
  
 < 0.001
  
 < 0.001
 Low
337 (10.3)
205 (8.7)
132 (14.2)
 
238 (9.3)
99 (13.8)
 
206 (8.4)
131 (16.1)
 
 Medium
1552 (47.5)
1115 (47.6)
437 (47.1)
 
1205 (47.2)
347 (48.4)
 
1187 (48.3)
365 (45.0)
 
 High
1381 (42.2)
1023 (43.7)
358 (38.6)
 
1110 (43.5)
271 (37.8)
 
1065 (43.3)
316 (38.9)
 
Employment status: N (%)
   
 < 0.001
  
 < 0.001
  
 < 0.001
 Full-time employed
1629 (49.8)
1184 (50.5)
445 (48.0)
 
1269 (49.7)
360 (50.2)
 
1213 (49.3)
416 (51.2)
 
 Retired
649 (19.8)
522 (22.3)
127 (13.7)
 
566 (22.2)
83 (11.6)
 
543 (22.1)
106 (13.1)
 
 Other
992 (30.3)
637 (27.2)
355 (38.3)
 
718 (28.1)
274 (38.2)
 
702 (28.6)
290 (35.7)
 
Religious affiliation: N (%)
   
 < 0.001
  
 < 0.001
  
 < 0.001
 No religious affiliation
1459 (44.6)
1093 (46.6)
366 (39.5)
 
1188 (46.5)
271 (37.8)
 
1148 (46.7)
311 (38.3)
 
 Christianity
1659 (50.7)
1177 (50.2)
482 (52.0)
 
1276 (50.0)
383 (53.4)
 
1223 (49.8)
436 (53.7)
 
 Islam
90 (2.8)
41 (1.7)
49 (5.3)
 
51 (2.0)
39 (5.4)
 
51 (2.1)
39 (4.8)
 
 Other
62 (1.9)
32 (1.4)
30 (3.2)
 
38 (1.5)
24 (3.3)
 
36 (1.5)
26 (3.2)
 
Smoking status: N (%)
   
 < 0.001
  
 < 0.001
  
 < 0.001
 Yes, daily
737 (22.5)
517 (22.1)
220 (23.7)
 
556 (21.8)
181 (25.2)
 
536 (21.8)
201 (24.8)
 
 Yes, sometimes
347 (10.6)
187 (8.0)
160 (17.3)
 
230 (9.0)
117 (16.3)
 
189 (7.7)
158 (19.5)
 
 No, not anymore
891 (27.2)
677 (28.9)
214 (23.1)
 
734 (28.8)
157 (21.9)
 
700 (28.5)
191 (23.5)
 
 Never smoker
1295 (39.6)
962 (41.1)
333 (35.9)
 
1033 (40.5)
262 (36.5)
 
1033 (42.0)
262 (32.3)
 
Alcohol consumption: N (%)
   
0.68
  
0.04
  
 < 0.001
 Daily
185 (5.7)
133 (5.7)
52 (5.6)
 
139 (5.4)
46 (6.4)
 
127 (5.2)
58 (7.1)
 
 Several times a week
560 (17.1)
394 (16.8)
166 (17.9)
 
423 (16.6)
137 (19.1)
 
383 (15.6)
177 (21.8)
 
 Once a week
578 (17.7)
409 (17.5)
169 (18.2)
 
450 (17.6)
128 (17.9)
 
426 (17.3)
152 (18.7)
 
 1–3 times a month
558 (17.1)
416 (17.8)
142 (15.3)
 
462 (18.1)
96 (13.4)
 
446 (18.1)
112 (13.8)
 
 Less often
782 (23.9)
560 (23.9)
222 (23.9)
 
616 (24.1)
166 (23.2)
 
613 (24.9)
169 (20.8)
 
 Never
607 (18.6)
431 (18.4)
176 (19.0)
 
463 (18.1)
144 (20.1)
 
463 (18.8)
144 (17.7)
 
Physical activity: N (%)
   
 < 0.001
  
 < 0.01
  
0.16
 No physical activity
843 (25.8)
603 (25.7)
240 (25.9)
 
652 (25.5)
191 (26.6)
 
640 (26.0)
203 (25.0)
 
 Less than 1 h per week
605 (18.5)
397 (16.9)
208 (22.4)
 
451 (17.7)
154 (21.5)
 
445 (18.1)
160 (19.7)
 
 Regularly, 1–2 h per week
813 (24.9)
574 (24.5)
239 (25.8)
 
624 (24.4)
189 (26.4)
 
594 (24.2)
219 (27.0)
 
 Regularly, 2–4 h per week
550 (16.8)
415 (17.7)
135 (14.6)
 
449 (17.6)
101 (14.1)
 
417 (17.0)
133 (16.4)
 
 Regularly, more than 4 h per week
459 (14.0)
354 (15.1)
105 (11.3)
 
377 (14.8)
82 (11.4)
 
362 (14.7)
97 (11.9)
 
Health-conscious diet: N (%)
   
0.02
  
 < 0.01
  
 < 0.01
 Not at all
351 (10.7)
239 (10.2)
112 (12.1)
 
255 (10.0)
96 (13.4)
 
240 (9.8)
111 (13.7)
 
 A little
1172 (35.8)
854 (36.4)
318 (34.3)
 
931 (36.5)
241 (33.6)
 
884 (36.0)
288 (35.5)
 
 Strongly
1505 (46.0)
1094 (46.7)
411 (44.3)
 
1191 (46.7)
314 (43.8)
 
1162 (47.3)
343 (42.2)
 
 Very strongly
242 (7.4)
156 (6.7)
86 (9.3)
 
176 (6.9)
66 (9.2)
 
172 (7.0)
70 (8.6)
 
Self-rated health: Mean (SD)
3.6 (0.8)
3.7 (0.7)
3.2 (0.9)
 < 0.001
3.7 (0.8)
3.2 (1.0)
 < 0.001
3.6 (0.8)
3.3 (0.9)
 < 0.001
Number of chronic conditions: Mean (SD)
1.7 (1.8)
1.5 (1.6)
2.3 (2.1)
 < 0.001
1.6 (1.7)
2.3 (2.1)
 < 0.001
1.6 (1.7)
2.1 (2.0)
 < 0.001
Hugging others (average per day): N (%)
   
 < 0.01
  
0.06
  
 < 0.01
 0
814 (24.9)
541 (23.1)
273 (29.4)
 
612 (24.0)
202 (28.2)
 
581 (23.6)
233 (28.7)
 
 1
1251 (38.3)
915 (39.1)
336 (36.2)
 
994 (38.9)
257 (35.8)
 
972 (39.5)
279 (34.4)
 
 2–3
998 (30.5)
739 (31.5)
259 (27.9)
 
792 (31.0)
206 (28.7)
 
763 (31.0)
235 (28.9)
 
 4+
207 (6.3)
148 (6.3)
59 (6.4)
 
155 (6.1)
52 (7.3)
 
142 (5.8)
65 (8.0)
 
Notes: p-values are based on independent t-tests or chi-square tests, as appropriate
Overall, 24.9% of respondents reported no daily hugs, 38.3% hugged one individual, while 30.5% hugged two or three individuals per day. Moreover, 6.3% hugged four or more people each day. In total, 28.3% of the respondents had probable depression, 21.9% had probable anxiety, and 24.8% reported having suicidal thoughts for at least several days in the past 2 weeks. In bivariate analyses there was a significant association of hugging others with probable depression (Cramer’s V = 0.07, p < 0.01) and suicidal ideation (Cramer’s V = 0.07, p < 0.01). The association between hugging others and probable anxiety was of borderline statistical significance: Cramer’s V = 0.05, p = 0.056). Additional details are shown in Table 1.

Regression analysis

Findings from the unadjusted and adjusted regression analyses are presented in Table 2. Hereafter, we mostly focus on describing the fully-adjusted models (i.e., adjusting for sociodemographic, lifestyle-related, and health-related covariates). However, it is worth noting that adding covariates to the regression models attenuated the associations of interest only slightly.
Table 2
Link of hugging frequency with probable depression, probable anxiety, and suicidal ideation. Results based on logistic regression analyses
Independent variables
Probable depression
Probable anxiety
Suicidal ideation
Hugging others (average per day): 1 (Reference category: 0)
0.73**
0.67***
0.65***
0.65***
0.78*
0.71**
0.71**
0.73*
0.72**
0.69**
0.65***
0.66**
 
(0.60–0.88)
(0.54–0.83)
(0.52–0.81)
(0.51–0.83)
(0.63–0.97)
(0.56–0.90)
(0.56–0.90)
(0.56–0.95)
(0.58–0.88)
(0.55–0.87)
(0.51–0.82)
(0.52–0.85)
2–3
0.69***
0.57***
0.53***
0.61***
0.79*
0.63***
0.60***
0.70*
0.77*
0.65***
0.56***
0.63***
 
(0.57–0.85)
(0.45–0.73)
(0.41–0.68)
(0.47–0.80)
(0.63–0.98)
(0.49–0.82)
(0.46–0.78)
(0.52–0.93)
(0.62–0.95)
(0.51–0.84)
(0.43–0.72)
(0.48–0.83)
4+
0.79
0.62*
0.55**
0.58*
1.02
0.77
0.71+
0.78
1.14
0.97
0.77
0.85
 
(0.57–1.10)
(0.43–0.90)
(0.37–0.81)
(0.38–0.89)
(0.71–1.45)
(0.52–1.14)
(0.47–1.06)
(0.51–1.20)
(0.82–1.59)
(0.67–1.40)
(0.52–1.13)
(0.57–1.28)
Sociodemographic covariates
 
 
 
Lifestyle-related covariates
  
  
  
Health-related covariates
   
   
   
R2
3270
3270
3270
3270
3270
3270
3270
3270
3270
3270
3270
3270
Observations
0.004
0.10
0.12
0.24
0.002
0.09
0.11
0.19
0.004
0.08
0.11
0.17
*** p < 0.001, ** p < 0.01, * p < 0.05, +p < 0.10, odds ratios are shown (95% CI are depicted in parentheses); sociodemographic covariates include sex, age, federal state, educational level, marital status, employment situation, and religious affiliation; lifestyle-related covariates include smoking status, physical activity, alcohol consumption, and health-conscious diet; health-related covariates cover self-rated health, and the number of chronic illnesses. Probable depression was based on the cut-point of 10 (PHQ-9), probable anxiety was based on the cut-point of 10 (GAD-7), and suicidal ideation was indicated by responding with “several days” or more to the ninth item of the PHQ-9
Hugging others (on average one to three persons; compared to individuals not hugging anyone) on a daily basis was associated with lower odds of probable depression (e.g., hugging one individual: OR 0.65; 95% CI, 0.51 to 0.83), probable anxiety (e.g., hugging one individual: OR: 0.73; 95% CI, 0.56 to 0.95), and suicidal ideation (e.g., hugging one individual: OR 0.66; 95% CI, 0.52 to 0.85). It is of note that the largest significant differences were found between individuals who hugged two to three individuals on average daily and those who did not hug anyone. It is also worth noting that hugging on average 4 or more individuals daily (compared to individuals not hugging others) was significantly associated with lower odds of probable depression (OR 0.58; 95% CI, 0.38 to 0.89), but not with lower odds of probable anxiety or suicidal ideation.
In a robustness check (described in detail in the statistical analysis section), the PHQ-8 was used rather than the PHQ-9 to quantify probable depression (to avoid an overlap in the mental health outcomes). The results of this analysis are presented in Table 3. The results remained nearly the same as those from the main analysis shown in Table 2. In a second robustness check, the main model was extended by adding the covariates on the number of children in the household and living arrangements (see Table 4). The findings remained virtually the same compared to the main model.
Table 3
Link of hugging frequency with probable depression (based on the PHQ-8 rather than the PHQ-9). Results based on logistic regression analyses
Independent variables
Probable depression
Hugging others (average per day): 1 (Reference category: 0)
0.71***
0.64***
0.62***
0.62***
 
(0.58–0.86)
(0.51–0.80)
(0.50–0.78)
(0.48–0.79)
2–3
0.69***
0.56***
0.53***
0.61***
 
(0.56–0.85)
(0.44–0.72)
(0.41–0.68)
(0.46–0.80)
4+
0.74+
0.57**
0.52**
0.55**
 
(0.53–1.05)
(0.39–0.83)
(0.35–0.78)
(0.35–0.84)
Sociodemographic covariates
 
Lifestyle-related covariates
  
Health-related covariates
   
R2
3270
3270
3270
3270
Observations
0.004
0.010
0.12
0.24
*** p < 0.001, ** p < 0.01, * p < 0.05, +p < 0.10, odds ratios are shown (95% CI are depicted in parentheses); sociodemographic covariates include sex, age, federal state, educational level, marital status, employment situation, and religious affiliation; lifestyle-related covariates include smoking status, physical activity, alcohol consumption, and health-conscious diet; health-related covariates cover self-rated health, and the number of chronic illnesses. Probable depression was based on the cut-point of 10 (PHQ-8)
Table 4
Link of hugging frequency with probable depression, probable anxiety and suicidal ideation. Results based on logistic regression analyses (with additional covariates: number of children in the household, and living arrangement)
Independent variables
Probable depression
Probable anxiety
Suicidal ideation
Hugging others (average per day): 1 (Reference category: 0)
0.73**
0.68***
0.66***
0.65***
0.78*
0.71**
0.71**
0.72*
0.72**
0.69**
0.64***
0.65***
 
(0.60–0.88)
(0.54–0.85)
(0.52–0.83)
(0.50–0.84)
(0.63–0.97)
(0.56–0.91)
(0.55–0.91)
(0.55–0.94)
(0.58–0.88)
(0.55–0.87)
(0.50–0.81)
(0.51–0.83)
2–3
0.69***
0.55***
0.51***
0.59***
0.79*
0.60***
0.57***
0.67**
0.77*
0.65***
0.55***
0.62***
 
(0.57–0.85)
(0.43–0.71)
(0.39–0.66)
(0.44–0.77)
(0.63–0.98)
(0.46–0.79)
(0.44–0.76)
(0.50–0.89)
(0.62–0.95)
(0.50–0.83)
(0.42–0.72)
(0.47–0.82)
4+
0.79
0.58**
0.52**
0.55**
1.02
0.73
0.67+
0.75
1.14
0.96
0.77
0.85
 
(0.57–1.10)
(0.39–0.85)
(0.35–0.77)
(0.35–0.85)
(0.71–1.45)
(0.49–1.09)
(0.44–1.02)
(0.48–1.16)
(0.82–1.59)
(0.66–1.40)
(0.52–1.14)
(0.56–1.29)
Sociodemographic covariates
 
 
 
Lifestyle-related covariates
  
  
  
Health-related covariates
   
   
   
R2
3270
3270
3270
3270
3270
3270
3270
3270
3270
3270
3270
3270
Observations
0.004
0.10
0.12
0.24
0.002
0.09
0.11
0.20
0.004
0.09
0.12
0.17
*** p < 0.001, ** p < 0.01, * p < 0.05, +p < 0.10, odds ratios are shown (95% CI are depicted in parentheses); sociodemographic covariates include sex, age, federal state, educational level, marital status, number of children in the household, living arrangement, employment situation, and religious affiliation; lifestyle-related covariates include smoking status, physical activity, alcohol consumption, and health-conscious diet; health-related covariates cover self-rated health, and the number of chronic illnesses. Probable depression was based on the cut-point of 10 (PHQ-9), probable anxiety was based on the cut-point of 10 (GAD-7), and suicidal ideation was indicated by responding with “several days” or more to the ninth item of the PHQ-9

Discussion

The aim of the present study was to examine the association of hugging with probable depression, probable anxiety, and suicidal ideation based on a large, representative sample (in terms of sex, age group, federal state) of the adult population in Germany. We found that hugging others daily—ideally one to three individuals—is associated with reduced odds of probable depression, anxiety, and suicidal thoughts. The strongest differences emerged between those who hug two to three individuals daily and those who do not hug anyone. Hugging four or more individuals daily was significantly associated with lower odds of probable depression, but not with lower odds of probable anxiety or suicidal ideation.
Until now there has been an almost total absence of research on the association of hugging with the aforementioned mental health outcomes, making it difficult to compare our findings with former studies. In fact, we could locate only one previous quantitative study (Rosenberg et al. 2021) that examined the association between kissing or hugging either a family member or partner almost every day (in comparison to not at all) with mental health, which found that such behaviors were associated with a lower prevalence of probable depression based on data from adults in the United States at the beginning of the COVID-19 pandemic. The current study therefore significantly advances understanding in this research field by showing that hugging is not merely linked to probable depression, but also to probable anxiety and suicidal ideation in Germany in the post-pandemic era.
In general, it is probable that the daily frequency of hugs is more likely to be driven by gestures of greeting or affection rather than for comforting purposes. Therefore, we propose that the physical intimacy of a hug may strengthen individuals' awareness of their positive and meaningful social bonds—such as those with friends, romantic partners or family members (Forsell and Åström 2012). This may have a direct and positive impact on mental health (Tunçgenç et al. 2023). Even when individuals experience negative emotions (such as anger or sadness), hugs may help them feel better (Murphy et al. 2018).
Interestingly, an average of two to three hugs per day was sufficient to produce the most benefits in the present study. We believe that this level of hugging may be a marker of the quality of one’s respective relationships. More precisely, we assume that the quality of the hugs in particular, could be important (e.g. valuable hugs with friends, family members or one’s partner) for mental health outcomes (see Berry and Worthington 2001). However, this is a speculative explanation that requires further research (e.g., based on qualitative studies).
Also remarkably, hugging four or more individuals on average per day was significantly associated with lower odds of probable depression, but not with lower odds of probable anxiety or suicidal ideation. We assume that frequent hugs may alleviate depression via social support, security, and the release of oxytocin (Dreisoerner et al. 2021; Field 2010). However, it is also possible that frequent hugs are not necessarily an expression of deep friendships and connections. Thus, individuals having many daily hugs might still feel little depth and serious support which may explain the missing link to probable anxiety and suicidal ideation. This explanation also fits with the result that a lower frequency of hugs (one to three on average daily) was associated with anxiety and suicidal thoughts in our study. It is worth emphasizing that this is only an initial, speculative explanation. Other explanations may refer to the professional environment. Certain jobs may involve frequent hugging (e.g. kindergarten teachers, social workers). Further research in this area is clearly needed. The context of hugging should be further explored. Another potential explanation refers to the limited statistical power.
Some important strengths and limitations are worth bearing in mind when interpreting the present findings. Importantly, data were taken from a quota-based sample (representative in terms of sex, age group, and federal state). Moreover, established screening tools (the PHQ-9 [and PHQ-8 in a robustness check] and GAD-7) were used to quantify probable depression and probable anxiety, respectively. The usefulness of these screening tools has been repeatedly demonstrated (Plummer et al. 2016; Shin et al. 2019; Wu et al. 2020). Further, the item used to measure suicidal ideation has clear face validity. However, future research with multi-item tools is recommended to gain additional insights into the association between the frequency of hugs and suicidal ideation (Na et al. 2018). As the present study was based on cross-sectional data it has inherent limitations regarding the directionality of the observed relationships (e.g., certain depressive symptoms such as anhedonia or anxiety may contribute to a lower likelihood of hugging). Future research based on longitudinal data is therefore recommended. It is also important to note that using data from an online survey may exclude a certain proportion of individuals (e.g., those without access to the internet or older individuals not familiar with it). However, the proportion of this group of people in the general German adult population is extremely low (Statistisches Bundesamt 2025). Moreover, individuals may give more honest responses in an online-survey (e.g., compared to being interviewed face-to-face) (Casler et al. 2013).
The findings of this study indicated an association between the frequency of hugging and mental health (including suicidal ideation). The daily hugging of one to three individuals may assist in preventing poor mental health. However, future longitudinal evidence is needed to confirm this. Studies from other countries are also recommended. Moreover, studies focusing on other forms of hugging (e.g., virtual hugging) are also warranted (Wang et al. 2025).

Declarations

Ethics approval

The Psychological Ethics Committee of the University Medical Center Hamburg-Eppendorf gave its approval for this study (LPEK-0849). The study was conducted in accordance with the Declaration of Helsinki.
Each individual gave his or her informed consent.
Not applicable.

Conflicts of interest

The authors have no competing interests to declare that are relevant to the content of this article.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Download
Titel
Association of hugs with depression, anxiety, and suicidal ideation: findings based on data from the general adult population in Germany
Verfasst von
André Hajek
Andrew Stickley
Larissa Zwar
Karl Peltzer
Supa Pengpid
Razak M. Gyasi
Dong Keon Yon
Hans-Helmut König
Publikationsdatum
25.10.2025
Verlag
Springer Berlin Heidelberg
Erschienen in
Journal of Public Health
Print ISSN: 2198-1833
Elektronische ISSN: 1613-2238
DOI
https://doi.org/10.1007/s10389-025-02627-6
Zurück zum Zitat Berry JW, Worthington EL Jr (2001) Forgivingness, relationship quality, stress while imagining relationship events, and physical and mental health. J Counsel Psychol 48(4):447CrossRef
Zurück zum Zitat Bock J-O, Hajek A, Weyerer S, Werle J, Wagner M, Maier W, Stark A, Kaduszkiewicz H, Wiese B, Moor L (2017) The impact of depressive symptoms on healthcare costs in late life: longitudinal findings from the AgeMooDe study. Am J Geriatr Psychiatry 25(2):131–141CrossRefPubMed
Zurück zum Zitat Buczak-Stec EW, Löbner M, Stein J, Stark A, Kaduszkiewicz H, Werle J, Heser K, Wiese B, Weyerer S, Wagner M (2022) Depressive symptoms and healthcare utilization in late life. Longitudinal evidence from the AgeMooDe study. Front Med 9:924309CrossRef
Zurück zum Zitat Campbell D, Green MJ, Davies N, Demou E, Howe LD, Harrison S, Smith DJ, Howard DM, McIntosh AM, Munafò M (2022) Effects of depression on employment and social outcomes: a Mendelian randomisation study. J Epidemiol Community Health 76(6):563–571CrossRefPubMed
Zurück zum Zitat Casler K, Bickel L, Hackett E (2013) Separate but equal? A comparison of participants and data gathered via Amazon’s MTurk, social media, and face-to-face behavioral testing. Comput Human Behav 29(6):2156–2160CrossRef
Zurück zum Zitat Celano CM, Huffman JC (2011) Depression and cardiac disease: a review. Cardiol Rev 19(3):130–142CrossRefPubMed
Zurück zum Zitat Dreisoerner A, Junker NM, Schlotz W, Heimrich J, Bloemeke S, Ditzen B, van Dick R (2021) Self-soothing touch and being hugged reduce cortisol responses to stress: a randomized controlled trial on stress, physical touch, and social identity. Compr Psychoneuroendocrinol 8:100091CrossRefPubMedPubMedCentral
Zurück zum Zitat Field T (2010) Touch for socioemotional and physical well-being: a review. Dev Rev 30(4):367–383CrossRef
Zurück zum Zitat Forsell LM, Åström JA (2012) Meanings of hugging: from greeting behavior to touching implications. Comprehensive Psychol 1 02.17. 21. CP. 01.13
Zurück zum Zitat Gastaldon C, Solmi M, Correll CU, Barbui C, Schoretsanitis G (2022) Risk factors of postpartum depression and depressive symptoms: umbrella review of current evidence from systematic reviews and meta-analyses of observational studies. Br J Psychiatry 221(4):591–602CrossRefPubMed
Zurück zum Zitat GBD 2019 Mental Disorders Collaborators (2022) Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry 9(2):137–150CrossRefPubMedCentral
Zurück zum Zitat Gordon I, Zagoory-Sharon O, Leckman JF, Feldman R (2010) Oxytocin, cortisol, and triadic family interactions. Physiol Behav 101(5):679–684CrossRefPubMed
Zurück zum Zitat Hajek A, König H-H (2021) The prevalence and correlates of probable major depressive disorder and probable generalized anxiety disorder during the COVID-19 pandemic. Results of a nationally representative survey in Germany. Int J Environ Res Public Health 18(23):12302CrossRefPubMedPubMedCentral
Zurück zum Zitat Hajek A, Neumann-Böhme S, Sabat I, Torbica A, Schreyögg J, Barros PP, Stargardt T, König H-H (2022a) Depression and anxiety in later COVID-19 waves across Europe: new evidence from the European COvid Survey (ECOS). Psychiatr Res 317:114902CrossRef
Zurück zum Zitat Hajek A, Sabat I, Neumann-Böhme S, Schreyögg J, Barros PP, Stargardt T, König H-H (2022b) Prevalence and determinants of probable depression and anxiety during the COVID-19 pandemic in seven countries: longitudinal evidence from the European COvid Survey (ECOS). J Affect Disord 299:517–524CrossRefPubMed
Zurück zum Zitat Hajek A, Teo AR, Zwar L, König HH (2024) Validation of the German version of the 25-item hikikomori questionnaire 25. Int J Methods Psychiatr Res 33(2):e2027CrossRefPubMedPubMedCentral
Zurück zum Zitat Hajek A, Gyasi RM, Kostev K, Soysal P, Veronese N, Smith L, Jacob L, Oh H, Pengpid S, Peltzer K (2025) Multimorbidity clusters and their contribution to well-being among the oldest old: results based on a nationally representative sample in Germany. Arch Gerontol Geriatr 130:105726CrossRefPubMed
Zurück zum Zitat Hajek A, König H-H (2024) Interest in anime and manga: relationship with (mental) health, social disconnectedness, social joy and subjective well-being. J Public Health 1–11
Zurück zum Zitat Hakulinen C, Musliner KL, Agerbo E (2019) Bipolar disorder and depression in early adulthood and long-term employment, income, and educational attainment: a nationwide cohort study of 2,390,127 individuals. Depress Anxiety 36(11):1080–1088CrossRefPubMed
Zurück zum Zitat Hohls JK, Koenig H-H, Raynik YI, Hajek A (2018) A systematic review of the association of anxiety with health care utilization and costs in people aged 65 years and older. J Affect Disord 232:163–176CrossRefPubMed
Zurück zum Zitat Huarcaya-Victoria J, Salvador-Carillo J, Cabrejos-Espinoza M, Ponce-Pantigoso M, Broncano-Rivera K, Loarte-Cabrera A (2025) Sociodemographic and clinical determinants of suicidal ideation: an analysis in a Peruvian consultation-liaison psychiatry unit. BMC Psychiatry 25(1):45CrossRefPubMedPubMedCentral
Zurück zum Zitat Incze MA, Anderson T, Hansen AM, Szczotka K, Stolebarger L, Tuckett S, Fox S, Bell C, Galyean P, Babbel D, Zickmund S (2025) “They don’t just need a handshake or a handoff, they need a hug”: a qualitative assessment of the care transition experience of patients with substance use disorders after hospital discharge. J Gen Intern Med. https://doi.org/10.1007/s11606-025-09614-3CrossRefPubMed
Zurück zum Zitat Khan A, Leventhal RM, Khan S, Brown WA (2002) Suicide risk in patients with anxiety disorders: a meta-analysis of the FDA database. J Affect Disord 68(2–3):183–190CrossRefPubMed
Zurück zum Zitat Kim S, Lee H-K, Lee K (2021) Which PHQ-9 items can effectively screen for suicide? Machine learning approaches. Int J Environ Res Public Health 18(7):3339CrossRefPubMedPubMedCentral
Zurück zum Zitat Kino S, Stickley A, Nishioka D, Ueno K, Saito M, Ojima T, Kondo N (2022) Suicidal ideation and suicide attempts among older recipients of public welfare assistance in Japan. J Epidemiol Community Health 76(10):873–879. https://doi.org/10.1136/jech-2022-218893CrossRefPubMed
Zurück zum Zitat Kroenke K, Spitzer RL, Williams JB (2001) The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 16(9):606–613CrossRefPubMedPubMedCentral
Zurück zum Zitat Kroenke K, Strine TW, Spitzer RL, Williams JBW, Berry JT, Mokdad AH (2009) The PHQ-8 as a measure of current depression in the general population. J Affect Disord 114(1):163–173. https://doi.org/10.1016/j.jad.2008.06.026CrossRefPubMed
Zurück zum Zitat Louzon SA, Bossarte R, McCarthy JF, Katz IR (2016) Does suicidal ideation as measured by the PHQ-9 predict suicide among VA patients? Psychiatr Serv 67(5):517–522CrossRefPubMed
Zurück zum Zitat Murphy ML, Janicki-Deverts D, Cohen S (2018) Receiving a hug is associated with the attenuation of negative mood that occurs on days with interpersonal conflict. PLoS ONE 13(10):e0203522CrossRefPubMedPubMedCentral
Zurück zum Zitat Na PJ, Yaramala SR, Kim JA, Kim H, Goes FS, Zandi PP, Vande Voort JL, Sutor B, Croarkin P, Bobo WV (2018) The PHQ-9 item 9 based screening for suicide risk: a validation study of the Patient Health Questionnaire (PHQ)−9 item 9 with the Columbia Suicide Severity Rating Scale (C-SSRS). J Affect Disord 232:34–40. https://doi.org/10.1016/j.jad.2018.02.045CrossRefPubMed
Zurück zum Zitat Östberg D, Nordin S (2022) Three-year prediction of depression and anxiety with a single self-rated health item. J Ment Health 31(3):402–409CrossRefPubMed
Zurück zum Zitat Pan A, Sun Q, Okereke OI, Rexrode KM, Hu FB (2011) Depression and risk of stroke morbidity and mortality: a meta-analysis and systematic review. JAMA 306(11):1241–1249CrossRefPubMedPubMedCentral
Zurück zum Zitat Pelton M, Ciarletta M, Wisnousky H, Lazzara N, Manglani M, Ba DM, Chinchillli VM, Du P, Ssentongo AE, Ssentongo P (2021) Rates and risk factors for suicidal ideation, suicide attempts and suicide deaths in persons with HIV: a systematic review and meta-analysis. Gen Psychiatry 34(2):e100247CrossRef
Zurück zum Zitat Plummer F, Manea L, Trepel D, McMillan D (2016) Screening for anxiety disorders with the GAD-7 and GAD-2: a systematic review and diagnostic metaanalysis. Gen Hosp Psychiatry 39:24–31. https://doi.org/10.1016/j.genhosppsych.2015.11.005CrossRefPubMed
Zurück zum Zitat Radfar M, Ahmadi F, Fallahi Khoshknab M (2014) Turbulent life: the experiences of the family members of patients suffering from depression. J Psychiatr Ment Health Nurs 21(3):249–256CrossRefPubMed
Zurück zum Zitat Rosenberg M, Luetke M, Hensel D, Kianersi S, Fu T-c, Herbenick D (2021) Depression and loneliness during April 2020 COVID-19 restrictions in the United States, and their associations with frequency of social and sexual connections. Soc Psychiatry Psychiatr Epidemiol 56(7):1221–1232. https://doi.org/10.1007/s00127-020-02002-8CrossRefPubMedPubMedCentral
Zurück zum Zitat Rossom RC, Coleman KJ, Ahmedani BK, Beck A, Johnson E, Oliver M, Simon GE (2017) Suicidal ideation reported on the PHQ9 and risk of suicidal behavior across age groups. J Affect Disord 215:77–84. https://doi.org/10.1016/j.jad.2017.03.037CrossRefPubMedPubMedCentral
Zurück zum Zitat Shin C, Lee SH, Han KM, Yoon HK, Han C (2019) Comparison of the usefulness of the PHQ-8 and PHQ-9 for screening for major depressive disorder: analysis of psychiatric outpatient data. Psychiatry Investig 16(4):300–305. https://doi.org/10.30773/pi.2019.02.01CrossRefPubMedPubMedCentral
Zurück zum Zitat Simon GE, Rutter CM, Peterson D, Oliver M, Whiteside U, Operskalski B, Ludman EJ (2013) Does response on the PHQ-9 depression questionnaire predict subsequent suicide attempt or suicide death? Psychiatr Serv 64(12):1195–1202. https://doi.org/10.1176/appi.ps.201200587CrossRefPubMedPubMedCentral
Zurück zum Zitat Skundberg-Kletthagen H, Wangensteen S, Hall-Lord ML, Hedelin B (2014) Relatives of patients with depression: experiences of everyday life. Scand J Caring Sci 28(3):564–571CrossRefPubMed
Zurück zum Zitat Spitzer RL, Kroenke K, Williams JB, Löwe B (2006) A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 166(10):1092–1097CrossRefPubMed
Zurück zum Zitat Statistisches Bundesamt (2025) Gut 4 % der Bevölkerung im Alter von 16 bis 74 Jahren in Deutschland sind offline https://www.destatis.de/DE/Presse/Pressemitteilungen/2025/04/PD25_N017_63.html
Zurück zum Zitat Stickley A, Shirama A, Sumiyoshi T (2023) Financial debt, worry about debt and mental health in Japan. BMC Psychiatry 23(1):761CrossRefPubMedPubMedCentral
Zurück zum Zitat Tassone VK, Duffy SF, Dunnett S, Boparai JK, Zuluaga Cuartas V, Jung H, Wu M, Goel N, Lou W, Bhat V (2024) Decreased odds of depressive symptoms and suicidal ideation with higher education, depending on sex and employment status. PLoS ONE 19(4):e0299817CrossRefPubMedPubMedCentral
Zurück zum Zitat Tunçgenç B, van Mulukom V, Newson M (2023) Social bonds are related to health behaviors and positive well-being globally. Sci Adv 9(2):eadd3715CrossRefPubMedPubMedCentral
Zurück zum Zitat UNESCO (2006) International Standard Classification of Education. ISCED 1997 (Re-edition ed.). UNESCO
Zurück zum Zitat Walker ER, McGee RE, Druss BG (2015) Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatr 72(4):334–341CrossRef
Zurück zum Zitat Wang X, Liu Q, Li X, Zheng X, Hommel B, Ma K (2025) Sadness is reduced by virtual hugging. Appl Psychol Health Well-Being 17(1):e70007. https://doi.org/10.1111/aphw.70007CrossRefPubMed
Zurück zum Zitat Wanjau MN, Möller H, Haigh F, Milat A, Hayek R, Lucas P, Veerman JL (2023) Physical activity and depression and anxiety disorders: a systematic review of reviews and assessment of causality. AJPM Focus 2(2):100074CrossRefPubMedPubMedCentral
Zurück zum Zitat Wu Y, Levis B, Riehm KE, Saadat N, Levis AW, Azar M, Rice DB, Boruff J, Cuijpers P, Gilbody S (2020) Equivalency of the diagnostic accuracy of the PHQ-8 and PHQ-9: a systematic review and individual participant data meta-analysis. Psychol Med 50(8):1368–1380CrossRefPubMed
Zurück zum Zitat Yuan K, Zheng Y-B, Wang Y-J, Sun Y-K, Gong Y-M, Huang Y-T, Chen X, Liu X-X, Zhong Y, Su S-Z (2022) A systematic review and meta-analysis on prevalence of and risk factors associated with depression, anxiety and insomnia in infectious diseases, including COVID-19: a call to action. Mol Psychiatry 27(8):3214–3222CrossRefPubMedPubMedCentral