ARCHIVED - Chronic Diseases in Canada


Volume 30, no. 2, March 2010

Gender, social relationships and depressive disorders in adults aged 65 and over in Quebec

S.-D. Mechakra-Tahiri, PhD (1,2); M. V. Zunzunegui, PhD (1,2); M. Préville, PhD (3); M. Dubé, PhD (4)

Author References

  1. 1 Department of Social and Preventive Medicine, Université de Montréal, Montréal, QC
  2. Institut de recherche en santé publique de l'Université de Montréal, Montréal, QC
  3. Department of Community Health, Université de Sherbrooke, Sherbrooke, QC
  4. Laboratory of Gerontology, Department of Psychology, Université du Québec à Trois-Rivières, Trois-Rivières, QC

Correspondence: Samia-Djemaa Mechakra-Tahiri, PhD, 911 Avenue Antonine Maillet, Outremont, QC H2V 2Y8, Tel.: 514-277-4688, E-mail:


The objective of this study was to examine if social relationships have a differential association with the presence of depression in men and women aged 65 and over. Data came from a survey of a representative sample of 2670 community-dwelling older adults in Quebec. Depressive disorders were measured using DSM-IV criteria. The prevalence of depression was 17.8% for women and 7.6% for men. Men reported a greater diversity of ties but less support than women. Having a confidant and/or being engaged in a good marital relationship was negatively associated with depression in both men and women. Compared with married people in general, widowhood was associated with a considerably higher risk of depression in men than in women. Compared with non-volunteers in general, men who volunteer were at considerably lower risk of depression than women who volunteer. This exploratory study could serve as a basis for future longitudinal studies on the impact of community activities and volunteering on the incidence and remission of depression in older men and women in Canada.

Keywords: gender, depression, social networks, volunteerism, widowhood, elderly, Quebec, older adults, conflict


Epidemiological studies agree that women present more depressive disorders than men, with a gender prevalence ratio between 1.5 and 2.1-7 In Canada, the prevalence of depression varies depending on the measuring instrument and diagnostic criteria used.8-10 According to Ostbye et al.9 in the population aged 65 and over, the prevalence of depression (diagnosed using DSM-III-R* criteria) was 9.3% in women and 2.9% in men; of major depression was 3.4% for women and 1.5% for men, and of minor depression was 6.0% for women and 1.4% for men. Risk of depression has been associated with lack of social networks and support,6,11a,12a and differences in the social networks of men and women have been documented.13a Thus, gender differences in depression could be at least partly explained by differences in the social relationships of men and women. Few studies have looked at the differential association between social relationships and depression in older populations, and their results are limited to some specific aspects of relationships, such as marital status or social support. According to some researchers, widowhood is associated with increased depressive symptomatology for both men and women13b and marriage appears to be more beneficial for men's mental health.3a,5,14 However, according to another study,15 the effect of marital status on mental health varies depending on the region or society in which the study was done. With respect to the influence of others, studies have looked mainly at the role played by offspring or friends in depression, and again, the results vary with the context of the study.13c,16a Some European studies reported that the contacts of offspring with elderly parents had a protective function,13d,17a,18a while in a study conducted in the United States, the support provided by offspring was found to be important mainly if the older adults were in a situation where they needed help (i.e. poverty, poor health or widowhood).19a

The association between social integration and depressive symptoms was reported by cross-sectional and longitudinal studies conducted in the elderly.11b,20,23a However, in these studies the differential effect of friends' support or social integration on mental health were not examined separately by gender.

Studies show that social support provided by the members of social networks is inversely associated with depression in older men,3b and that the presence of conflictual relationships is related to depression for both women and men.13e In other research, however, the association of conflictual relationships is stronger in women.24a In Canada, one study examined the association between social relationships and the presence of psychological distress and depression in older adults25 and showed that even though women receive more support than men, there are no gender differences in the impact of support on these.

In this study, we examined the association between depression in adults aged 65 and over and several structural aspects of social relationships such as marital status, relationships with family and friends and memberships in clubs, other groups and leisure organizations. We further contribute to the literature by examining the associations between depression and functional aspects of social relationships, such as support and conflict. Lastly, we examine if these associations differ between older men and older women.


Study population and data collection

Data used in this study came from the “Enquête sur la Santé des Aînés” (ESA), a cross-sectional survey conducted in 2005-2006 using a representative sample of French-speaking, community-dwelling adults over the age of 65 (94% of the Quebec population speaks French). A proportional sample of households was constituted according to the 16 administrative regions of Quebec and a random sampling method was used to select no more than one adult aged 65 years or over per household.26,27a Subjects who obtained a score of less than 22 on the Mini-mental State Examination (MMSE) were excluded because they could have presented a cognitive deficit28 that compromised the validity of their responses to the ESA questionnaire. The response rate was 66.5%. Data were collected through at-home, face-to-face interviews. Details on data collection procedures are given in a previous publication.27b The research procedure was reviewed and approved by the Ethics Committee of the Sherbrooke Geriatric University Institute.


Dependent variables: depressive disorders

The respondents' depression status (including major and minor depression) was measured using the computer-based ESA-Q developed by members of the research team27c and based on criteria in the DSM-IV.†The ESA-Q is similar to the Diagnostic Interview Schedule (DIS)29a and the Composite International Diagnostic Interview (CIDI),30 which demonstrated satisfactory reliability and validity.29b,31

A 12-month recall period was used. Subjects were classified as major depression cases if they displayed the essential features of depression (i.e. either depressed mood or the loss of interest or pleasure in usual daily activities, and reporting at least five of the nine symptoms associated with depression) nearly every day and most of the day for at least two consecutive weeks. In addition, subjects who displayed the essential features of depression and reported between two to four of the nine associated symptoms of depression within the same time period were classified as minor depression cases. For the purpose of this study, the definition of depression did not include impairment of respondents' usual social functioning. Participants whose symptoms were related to a physical illness or treatment were not classified as cases of major or minor depression. For the purpose of this study, and in agreement with our previous research,27d a dichotomous outcome was defined as 1 if the person fulfilled the criteria for major or minor depression and 0 if otherwise.

Independent variables: social relationships

The structural and functional characteristics of the respondents' social relationships were measured using four variables: social network, social integration, social support and conflictual relationships.

The social network was measured using two indicators: diversity of social ties and marital status.32,33a Diversity of social ties was measured by the number of different types of relationships that participants had, including those with a partner, adult children, siblings, friends and members of a community group.33b This variable was rated on a scale of 0 to 5 and recoded into three categories: low (0 to 2 relationships), medium (3 to 4 relationships) and high (5 relationships). Marital status was divided into four categories: with partner, separated or divorced, widowed, and never married. Social integration was measured using three questions related to participation in the community: 1) Do you regularly go to leisure, cultural or social centres? 2) Do you regularly do volunteer work? 3) Do you regularly attend religious services (at a church, synagogue, mosque or other centre of worship)?

Social support34 was measured using three questions about the availability of a confidant to talk to about various problems, the presence of someone who could provide instrumental help and the presence of someone who could provide emotional support.

Conflictual relationships were measured using five questions, three of which concerned feeling criticized or disapproved of by adult children, siblings and/or friends.13f,24b Each was evaluated on a scale of 0 to 4. Those who answered never (0) or rarely (1) were considered not to have any conflictual relationships, while those who answered sometimes (2), often (3) or always (4) were considered to have conflictual relationships with family or friends. For the analysis, those with conflict in their relationships with adult children (2 to 4) and those without conflict in these relationships (0 to 1) were compared with those without children (the reference category).

The remaining two questions concerned conflicts in the relationship with one's partner and evaluated the frequency with which the person argued with him or her as follows: 1) How often do you argue with your partner? and 2) How often do you and your partner get on each other's nerves? Each variable was measured on a scale from 0 (never) to 5 (always) to classify frequency of quarrelling. A single variable summing the responses to these two questions was created and divided into three categories: no partner, no conflict (rarely or never) and conflicts (sometimes or more often). Again those without a spouse were taken as the reference category to examine if the presence of conflict changes the association between depression and the fact of being married.

The following risk factors for depression, as identified in the literature, were considered in the analysis due to their possible confounding role: age, gender, region of residence, family annual income and health status.

Region of residence was divided into urban, metropolitan and rural, as defined by the Institut de la statistique du Québec.35 According to the Institute, the only metropolitan region in Quebec is the city of Montréal; an urban area is one with at least 1000 inhabitants and a population density of at least 400 inhabitants per square kilometre; and all other regions are classified as rural.

Family annual income was divided into four categories: less than $15,000, between $15,000 and $25,000, between $25,001 and $35,000, and over $35,000.

Health status was measured using two indicators: self-perceived physical health and morbidity. Perceived physical health was evaluated with the question: In general, compared to other people your age, would you say your physical health is excellent, very good, good, fair or poor? Morbidity was measured by the number of the participant's chronic health problems (defined in the ICD-10‡), which was divided into four categories: 0, 1 to 2, 3 to 4, and 5 or more chronic health problems. Chronic conditions were evaluated with the following question: To your knowledge and under the advice of a doctor, do you currently have one of the following diseases: heart disease, disorders of the digestive and endocrine systems, and/or 15 other conditions.§

Statistical analyses

A descriptive analysis of the socio- demographic characteristics and social relationships was done separately for the male and female participants, and the differences in the distributions were examined using the chi-square test. Confounding variables associated with depression in the bivariate analyses with (p < .25) were considered in the subsequent multivariate analyses. 36

Simple binary logistic regression models, done separately for men and women, were used to examine the associations between social relationships and the presence of a depressive disorder and to estimate the odds ratios (OR) with 95% confidence intervals (CI). The confounding variables were included in two blocks. The first included socio-demographic factors and the second, physical health. Staggered entry was used to assess the changes in the coefficients of the associations between depressive disorders and social relationships after adjusting first for socio-demographic variables and second for physical health.

Interaction terms of each variable of social relationships with gender were tested one-by-one in a series of logistic regression models fitted to the whole sample (men and women) and controlling for all confounders. Lastly, a final model including all significant interactions and confounders was fitted. Odds ratios were estimated with their 95% confidence intervals. All the analyses were done with SPSS version 15 for Windows.**


For this study we used the 2670 individuals with complete data on all the variables in the analyses. The 132 individuals for whom at least one item of information was missing differed from those with complete data on marital status (more were married: 70.8%, p < .001) and gender (more men [58.3%] than women [41.7%]; p < .001). No significant difference was found for the remaining socio-demographic and health characteristics.

Table 1 shows that the female participants in our study had higher exposure to those factors commonly associated with depression:2 they were older, had less schooling, were poorer, had more chronic health problems and were in poorer health than the male participants. They also had significantly higher prevalence of depression than men. The prevalence of depressive disorders (including major and minor depression) was 17.8% for women and 7.6% for men. For major depression, the prevalence was 5.3% for women and 2.6% for men; for minor depression, it was 12.5% for women and 5.0% for men.

As shown in Table 2, a higher proportion of men than women lived with a partner (57.2% vs. 36.6%, p < .001) and twice as many of the women were widowed (42.7% vs. 20.0%, p < .001). A high proportion of both men (87.5%) and women (89.5%) reported having friends; 86% reported having children and 90%, siblings. No significant difference with regards to the distributions of having friends, children or siblings was found between men and women. In addition, the results (Table 2) show that men reported a greater diversity of ties than did women (p < .001), while a greater proportion of women went to social centres (p = .008) and attended religious services (p < .001). One-third of both men and women reported being involved in volunteer work. A greater proportion of women than men reported having a confidant (88.7% vs. 85.4%, p < .012).

Of the married respondents, men reported fewer conflictual relationships with their spouse (70.7 vs. 79.2%, p < .001).

Table 1
Distribution of social relationships in women and men aged 65 and over in Quebec (N = 2670)
Variables Men (n = 1073) Women (n = 1597) p(χ2)
  % %  
Age (years)   < .001
65-69 33.7 28.8  
70-74 29.2 24.9  
75-79 21.5 21.5  
80-84 12.6 18.0  
85+ 3.0 6.7  
Schooling   < .001
Elementary or less 65.6 68.6  
Secondary 16.1 20.0  
Post-secondary and college 18.3 11.3  
Income   < .001
< $15,000 9.1 22.6  
$15,000-$25,000 18.8 28.3  
$25,001-$35,000 21.5 25.1  
> $35,000 50.6 24.1  
Area   0.11
Urban 15.5 17.3  
Metropolitan 43.6 45.6  
Rural 40.9 37.0  
Chronic health problems   < .001
0 10.2 6.1  
1-2 35.3 31.0  
3-4 33.2 34.4  
5 or more 21.2 28.6  
Perceived physical health   < .001
Excellent 22.4 17.1  
Very good 33.8 34.6  
Good 30.3 31.1  
Fair 11.9 15.0  
Poor  1.6  2.2  
Depression   < .001
Major 2.6 5.3  
Minor 5.0 12.5  
Combined 7.6 17.8  
Abbreviations: χ2, chi-square test; <, less then; n, sample size; N, overall sample size; %, percent; p, p-value.

Table 3 shows the associations between depression and social relationships, first adjusting for socio-economic factors and second adjusting for physical health. Being widowed, going less to social centres, and not volunteering were significantly associated with the probability of presenting a depressive disorder in older men but not in women. Lack of instrumental support lost significance in the fully adjusted model (OR=2.33; 95%CI 0.99, 5.46). While our results show that men involved in volunteer work had lower frequency of depressive disorders, this association was not significant in women. No statistically significant associations were found between depressive disorders and diversity of ties or attending religious services in either men or women.

With regards to social support, lack of instrumental support†† in men and lack of emotional support in women were associated with increased likelihood of presenting a depressive disorder. Unavailability of a confidant was associated with depression in both men and women.

Finally, the presence of conflictual relationships with friends, offspring and siblings was not associated with depression in men or women while nonconflictual relationships with one's partner was associated with decreased probability of reporting a depressive disorder in both men and women (Table 3). Nevertheless, among men marriage was associated with less depression, even in the presence of marital conflict. Among women marriage was associated with low likelihood of depression only when there was no conflict in the marital relationship.

Controlling for health indicators did not substantially change the odds ratios except for the attenuation of the association between lack of instrumental support and depression in men and lack of emotional support and depression in women (Table 3).

Introduction of multiplicative terms to test for gender interactions revealed two differential and significant associations: between marital status and depression and between volunteering and depression (Table 4). Widowhood is more strongly associated with depression in men than in women: a widowed man is 2.85 times more likely to be depressed than a married man, while a widowed woman is only 1.35 times (3.34/2.47) more likely to be depressed than a married woman. Differential effects of separation/divorce and of never-married status are much smaller.

Volunteering may be more beneficial to men (odds ratio for depression 0.43 compared to non-volunteers) than to women (odds ratio = 0.89).

Table 2
Distribution of social relationships in women and men aged 65 and over in Quebec (N = 2670)
Variables Men (n = 1073) Women (n = 1597) p(χ2)
  % %  
Social Networks
Marital status   < .001
With partner 57.2 36.6  
Separated or divorced 15.5 12.0  
Widowed 20.0 42.7  
Never married 7.4 8.8  
Children   .89
Yes 86.4 86.6  
No 13.6 13.4  
Siblings   .49
Yes 91.0 90.2  
No 9.0 9.8  
Siblings   .10
Yes 87.5 89.5  
No 12.5 10.5  
Diversity of ties   < .001
Low (0-2 relationships) 13.5 18.0  
Medium (3-4 relationships) 64.1 72.6  
High (5 relationships) 22.3 9.4  
Visit social centres
Diversity of ties   < 0.001
Yes 39.5 44.6  
No 60.5 55.4  
Volunteering   .27
Yes 34.5 32.4  
No 65.5 67.6  
Attending religious services   < 0.001
Yes 44.4 55.5  
No 55.6 44.5  
Social support
Presence of a confidant   0.12
Yes 85.4 88.7  
No 14.6 11.3  
Instrumental support   .97
Yes 95.9 95.9  
No 4.1 4.1  
Emotional support   0.35
Yes 95.2 96.0  
No 4.8 4.0  
Conflictual relationships
Children   .08
No conflicts 66.1 62.6  
Conflicts 20.3 24.0  
No children 13.6 13.4  
Siblings   .44
No conflicts 73.6 71.3  
Conflicts 17.1 18.7  
No siblings 9.4 10.0  
Friends   .20
No conflicts 71.6 74.2  
Conflicts 15.7 15.2  
No friends 12.7 10.6  
Partner   < .001
No conflicts 16.8  7.6  
Conflicts 40.4 29.0  
No partner 42.8 63.4  

Abbreviations: χ2, chi-square test; <, less then; n, sample size; N, overall sample size; %, percent; p, p-value.

Table 3
Odds ratios for depressive disorders by social relationships and gender in women and
men aged 65 and over in Quebec (N = 2670)
  Men (n = 1073) Women (n = 1597)
  ORa (95% CI) ORb (95% CI) ORa (95% CI) ORb (95% CI)
Marital status
With partner 1 1 1 1
Separated or divorced 1.28 (0.62, 2.65) 1.26 (0.61, 2.60) 1.43 (0.90, 2.25) 1.52 (0.95, 2.43)
Widowed 3.00 (1.62, 5.53) 2.97 (1.60, 5.51) 1.28 (0.90, 1.82) 1.35 (0.95, 1.93)
Never married 1.38 (0.54, 3.51) 1.39 (0.55, 3.53) 1.11 (0.65, 1.87) 1.22 (0.72, 2.09)
Yes 1 1 1 1
No 0.80 (0.39, 1.62) 0.77 (0.38, 1.57) 1.24 (0.85, 1.81) 1.24 (0.84, 1.81)
Yes 1 1 1 1
No 0.84 (0.35, 1.99) 0.78 (0.33, 1.86) 1.31 (0.85, 2.02) 1.26 (0.81, 1.95)
Yes 1 1 1 1
No 0.75 (0.36, 1.58) 0.75 (0.36, 1.58) 0.96 (0.62, 1.48) 0.85 (0.55, 1.33)
Diversity of ties
Low 1.27 (0.58, 2.78) 1.18 (0.54, 2.59) 1.78 (1.00, 3.14) 1.60 (0.90, 2.85)
Medium 0.72 (0.41, 1.26) 0.68 (0.39, 1.20) 1.39 (0.85, 2.25) 1.37 (0.84, 2.23)
High 1 1 1 1
Social integration
Social centres
Yes 1 1 1 1
No 1.74 (1.05, 2.87) 1.69 (1.02, 2.80) 1.04 (0.80, 1.36) 0.92 (0.70, 1.21)
Yes 1 1 1 1
No 2.43 (1.35, 4.36) 2.33 (1.29, 4.20) 1.24 (0.93, 1.65 ) 1.09 (0.81, 1.46)
Attending religious services
Yes 1 1 1 1
No 0.91 (0.56, 1.47) 0.92 (0.60, 1.48) 1.18 (0.91, 1.55) 1.12 (0.85, 1.47)
Social support
Presence of a confidant
Yes 1 1 1 1
No 1.87 (1.10, 3.17) 1.88 (1.11, 3.21) 1.73 (1.20, 2.49) 1.56 (1.08, 2.26)
Instrusmental upport
Yes 1 1 1 1
No 2.42 (1.04, 5.63) 2.33 (0.99, 5.46) 1.55 (0.89, 2.70) 1.32 (0.74, 2.32)
Emotional support
Yes 1 1 1 1
No 1.28 (0.53, 3.10) 1.23 (0.51, 2.98) 1.80 (1.05, 3.10) 1.55 (0.89, 2.69)
Conflictual relationships
No children 1 1 1 1
No conflicts 1.11 (0.53, 2.29) 1.13 (0.55, 2.34) 0.68 (0.46, 1.00) 0.70 (0.47, 1.04)
Conflicts 1.83 (0.83, 4.06) 1.81 (0.82, 4.01) 1.17 (0.77, 1.79) 1.11 (0.72, 1.72)
No siblings 1 1 1 1
No conflicts 1.08 (0.44, 2.63) 1.15 (0.50, 2.84) 0.67 (0.43, 1.04) 0.70 (0.45, 1.10)
Conflicts 1.36 (0.50, 3.70) 1.41 (0.52, 3.85) 1.07 (0.65, 1.76) 1.10 (0.66, 1.83)
No friends 1 1 1 1
No conflicts 1.00 (0.47, 2.15) 1.00 (0.47, 2.14) 1.04 (0.66, 1.62) 1.18 (0.75, 1.86)
Conflicts 1.58 (0.65, 3.81) 1.61 (0.66, 3.89) 1.04 (0.60, 1.79) 1.16 (0.67, 2.01)
No partner 1 1 1 1
No conflicts 0.35 (0.15, 0.81) 0.36 (0.15, 0.83) 0.45 (0.24, 0.85) 0.44 (0.23, 0.83)
Conflicts 0.58 (0.33, 1.01) 0.59 (0.34, 1.02) 0.88 (0.63, 1.23) 0.82 (0.58, 1.15)

Abbreviations: CI, confidence interval; n, sample size; N, overall sample size; OR, odds ratio.

aadjusted for age, income and type of region

Table 4
Odds ratio for depression in women and men aged 65 and over in Quebec: interactions between marital status and gender, and volunteering and gender
    OR (95% CI)a
Marital status
Married Men 1
Women 2.47 (1.60, 3.83)
Separated/ divorced Men 1.15 (0.59, 2.24)
Women 3.68 (1.44, 9.40)
Widowed Men 2.85 (1.64, 4.96)
Women 3.34 (1.48, 7.57)
Never married Men 1.66 (0.68, 4.02)
Women 2.94 (0.98, 8.80)
No Men 1
Women 2.47 (1.60, 3.83)
Yes Men 0.43 (0.24, 0.77)
Women 2.20 (1.01, 4.77)

Abbreviations: CI, confidence interval; OR, odds ratio; p, p-value.

a adjusted for age, income, type of region, number of chronic health problems and perceived physical health Lemeshow-Hosmer goodness of fit statistic p = .24


The objective of this study was to examine if social relationships were associated with the probability of presenting a depressive disorder in women and men aged 65 and over. Our results showed that marital status, being socially active and doing volunteer work were strongly inversely associated with the presence of depression only in older men. Women were more depressed than men generally, except widowers.

Our study did not find any association between relationships with adult children, siblings and friends and depression in older men and women in Quebec. This result is different from those reported in some studies that showed that offspring were salient for older adults' mental health,17b,18b,37 or others that reported the beneficial effect of friends on older adults' well-being.12b,13g,7b,16b In the United States, Silverstein et al. showed that support from adult children was important for older adults only if they were in a situation where they needed help (in poor health or widowed). However, we could not compare our results with those reported in these studies because they did not examine gender differences.

To reiterate, according to our results offspring do not seem to play an important role in depression among adults over 65 years old in Quebec. Maybe the social life of elderly Canadians does not revolve around their children or other family members, contrary to what has been reported in other countries where similar studies were done.17c,19b It would be interesting to study ethnic differences in Quebec and to compare these results with the results of our study.

Our results suggest that some social integration activities, such as going to community social centres and, to an even greater extent, being involved in volunteer work, are associated with less depression in men but not in women. However, no association was found between attending religious services and the presence of a depressive disorder in either men or women. This suggests that attendance at a religious centre, in spite of increasing the opportunities for social interaction, does not play a key role in depression in older adults in Quebec, unlike other types of group activities.

According to Canadian data published by the 2007 Canada Survey of Giving, Volunteering and Participating,38a 36% of older adults are involved in volunteerism, defined as unpaid, freely chosen activities within an organization.

The benefits of volunteerism for depression are supported by the results of various studies.21,22,39 According to other research,23b,40 volunteerism could be conducive to mental health through social integration, use of one's skills and reinforcement of the feeling of being useful. However, our study is cross-sectional and it is also possible that depressed people are unable to volunteer.

In addition, our results showed that gender differences in depression were very large among those engaged in volunteering. In fact, volunteerism was not related to depression in women but was strongly associated with reduced depression in men. We could suggest two not exclusive explanations. First, in this cohort of older adults, regardless of whether or not they had worked outside the home, women maintained their role as homemakers as they aged whereas for men, who had lost their role as active workers upon retirement, volunteerism gave them the opportunity to continue being productive. Second, the number and nature of volunteer activities seem to differ for men and women. The 2007 Canada Survey of Giving, Volunteering and Participating38b showed that women devoted more time to volunteer work than men, but that men devoted more time to volunteering in physical activities.

Also, since volunteering usually stems from a sense of community, it could be important to determine if the benefits of volunteering for mental health depend on the duration and history of volunteering. Further studies could document other aspects of volunteer activities (the nature and intensity, etc.) to identify if specific aspects could play a protective role in depression in older adults.

The lack of a confidant was positively related to depression in both men and women. Since associations with depression and lack of instrumental support in men and lack of emotional support in women were attenuated after adjusting for health status, we tested for the presence of interactions between both types of social support and physical health, to verify if social support was only effective among those with poor health as it has been previously reported for mild depression (analysis and results of these tests are not shown).41 However, these interactions were not significant in men or women.

Finally, our results show that conflict in the relationships with offspring and siblings do not seem to have any effect on depression. Non-conflictual relationships with a partner seem to be associated with a low probability of depression in both men and women. This fits with the findings of Antonucci et al. that negative social relations are associated with depression for both older women and men.13h


The limitations of this study include the fact that it is cross-sectional, which limits our ability to establish a chronological link between social relationships and depression. Some associations may be inverse, e.g. the presence of depression may lead to poor perceived health, less volunteering, less social integration and to more conflicts with the partner.

The ESA survey response rate of 66.5% is similar to that of other Canadian health surveys of older adults.42 This could result in a selection bias and under-evaluation of depression since those who did not participate in the study could be the oldest or more cognitively impaired elderly people, as has been shown in other studies.43 As no information was available on non- respondents to the ESA study, assessment of this selection bias was not possible. In addition, all potential participants with a score of less than 22 on the MMSE were excluded from the study. Therefore, the results can be generalized only to the older adult population with good cognitive functions and not living in the remote Northern areas of Quebec (Côte Nord, Gaspé Peninsula, Magdalen Islands, Saguenay-Lac-Saint-Jean and Abitibi-Témiscaminque).

Despite these limitations, the study used a large representative sample of community- dwelling older adults in Quebec. In addition, depression was measured using recognized diagnostic clinical criteria, unlike some other studies that used psychological distress measures. The study contributes to the scarce literature on the effects of social relationships, social support, and conflict on depression in older men and women.


Our results suggest that some specific aspects of social relationships could play a role in depression in older men and women in Quebec. The availability of a confidant and the absence of conflict with one's partner appear to be the specific aspects of social relationships most strongly associated with lack of depression. In addition, gender differential associations between marital status and depression and between social activities and depression were observed. Other longitudinal studies on the effect of volunteerism are needed to learn more about its effect on the incidence and remission of depressive disorders in older men and women.

Furthermore, the absence in this cross-sectional study of associations between networks of family and friends and the presence of depression in older adults does not exclude the possibility that these relationships could be important for some subgroups of older adults with specific needs. These subgroups could include individuals with functional disabilities or some ethnic groups. Therefore, it would be interesting to conduct research on the role of offspring and other family and friends in depression among the elderly population according to ethnicity.


This study was supported by the Canadian Institutes of Health Research (200403MOP). The authors thank the members of the Scientific Committee of the ESA Study who provided significant scientific advice.


*Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised

†Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.

‡International Statistical Classification of Diseases and Related Health Problems, 10th Revision.

§high blood pressure; hypercholesterolemia; diabetes mellitus; anaemia; B12 deficiency; asthma; chronic bronchitis; liver disease; kidney disease;skin disease; eye disease; arthritis; chronic back pain; headache; schizophrenia

**Statistical Package for the Social Sciences:

††Instrumental support: presence of someone who could provide practical help.


  1. ^ Alvarado BE, Zunzunegui MV, Beland F, Sicotte M, Tellechea L. Social and gender inequalities in depressive symptoms among urban older adults of Latin America and the Caribbean. J Gerontol B Psychol Sci Soc Sci. 2007 Jul;62(4):S226-36.
  2. ^ Bebbington PE. Sex and depression. Psychol Med. 1998 Jan;28(1):1-8.
  3. a,b Sonnenberg CM, Beekman AT, Deeg DJ, van Tilburg W. Sex differences in late-life depression. Acta Psychiatr Scand. 2000 Apr; 101(4):286-92.
  4. ^ Chou KL, Chi I. Prevalence and correlates of depression in Chinese oldest-old. Int J Geriatr Psychiatry. 2005 Jan;20(1):41-50.
  5. ^ Djernes JK. Prevalence and predictors of depression in populations of elderly: a review. Acta Psychiatr Scand. 2006 May; 113(5):372-87.
  6. ^ Jeon GS, Jang SN, Rhee SJ, Kawachi I, Cho SI. Gender differences in correlates of mental health among elderly Koreans. J Gerontol B Psychol Sci Soc Sci. 2007 Sep; 62(5):S323-9.
  7. ^ Piccinelli M, Wilkinson G. Gender differences in depression. Critical review. Br J Psychiatry. 2000 Dec;177:486-92.
  8. ^ Newman SC, Bland RC, Orn HT. The prevalence of mental disorders in the elderly in Edmonton: a community survey using GMS-AGECAT. Geriatric Mental State-Automated Geriatric Examination for Computer Assisted Taxonomy. Can J Psychiatry. 1998 Nov; 43(9):910-4.
  9. ^ Ostbye T, Kristjansson B, Hill G, Newman SC, Brouwer RN, McDowell I. Prevalence and predictors of depression in elderly Canadians: The Canadian Study of Health and Aging. Chronic Dis Can. 2005 Fall;26(4):93-9.
  10. ^ Streiner DL, Cairney J, Veldhuizen S. The epidemiology of psychological problems in the elderly. Can J Psychiatry. 2006 Mar;51(3):185-91.
  11. a,b Glass TA, De Leon CF, Bassuk SS, Berkman LF. Social engagement and depressive symptoms in late life: longitudinal findings. J Aging Health. 2006 Aug;18(4):604-28.
  12. a,b Dean A, Kolody B, Wood P. Effects of social support from various sources on depression in elderly persons. J Health Soc Behav. 1990 Jun;31(2):148-61.
  13. a,b,c,d,e,f,g,h Antonucci TC, Lansford JE, Akiyama H, Smith J, Baltes MM, Takahahashi K, Fuhrer R, Dartigues JF. Differences between men and women in social relations, resource deficits, and depressive symptomatology during later life in four nations. J Soc Issues. 2002;58(4):767-83.
  14. ^ Prince MJ, Harwood RH, Blizard RA, Thomas A, Mann AH. Social support deficits, loneliness and life events as risk factors for depression in old age. The Gospel Oak Project VI. Psychol Med. 1997 Mar;27(2):323-32.
  15. ^ Bebbington P, Dunn G, Jenkins R, Lewis G, Brugha T, Farrell M, Meltzer H. The influence of age and sex on the prevalence of depressive conditions: report from the National Survey of Psychiatric Morbidity. Int Rev Psychiatry. 2003 Feb-May;15 (1-2):74-83.
  16. a,b Crohan S, Antonucci T. Friends as a source of social support in old age. In: Adams, R, Blieszner, R, editors. Older adult friendship: structure and process. Newbury Park, London (CA): Sage; 1989. p. 129-46.
  17. a,b,c Buber I, Engelhardt H. Children's impact on the mental health of their older mothers and fathers: findings from the Survey of Health, Ageing and Retirement in Europe. Eur J of Ageing. 2008;5(1):31-45.
  18. a,b Zunzunegui MV, Beland F, Otero A. Support from children, living arrangements, self-rated health and depressive symptoms of older people in Spain. Int J Epidemiol. 2001 Oct;30(5):1090-9.
  19. a,b Silverstein M, Bengtson VL. Does intergenerational social support influence the psychological well-being of older parents? The contingencies of declining health and widowhood. Soc Sci Med. 1994 Apr;38(7):943-57.
  20. ^ Kawachi I, Berkman LF. Social ties and mental health. J Urban Health. 2001 Sep;78(3):458-67.
  21. ^ Greenfield EA, Marks NF. Formal volunteering as a protective factor for older adults' psychological well-being. J Gerontol B Psychol Sci Soc Sci. 2004 Sep;59(5):S258-64.
  22. ^ Morrow-Howell N, Hinterlong J, Rozario PA, Tang F. Effects of volunteering on the well-being of older adults. J Gerontol B Psychol Sci Soc Sci. 2003 May;58(3):S137-45.
  23. a,b Musick MA, Wilson J. Volunteering and depression: the role of psychological and social resources in different age groups. Soc Sci Med. 2003 Jan;56(2):259-69.
  24. a,b Seeman TE. Health promoting effects of friends and family on health outcomes in older adults. Am J Health Promot. 2000 Jul-Aug;14(6):362-70.
  25. ^ Cairney J, Krause N. The social distribution of psychological distress and depression in older adults. J Aging Health. 2005 Dec;17(6):807-35.
  26. ^ Préville M, Boyer R, Grenier S, Dubé M, Voyer P, Punti R, Baril MC, Streiner DL, Cairney J, Brassard J; Scientific Committee of the ESA Study. The epidemiology of psychiatric disorders in Quebec's older adult population. Can J Psychiatry. 2008;(12):822-32. Available from:
  27. a,b,c,d Mechakra-Tahiri S, Zunzunegui MV, Préville M, Dubé M. Social relations and depression among people 65 years and over living in rural and urban areas of Quebec. Int J Geriatr Psychiatry. 2009 Nov; 24:1226-36
  28. ^ Folstein MF, Folstein SE, McHugh PR. « Mini-mental states ». A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975 Nov;12(3):189-98.
  29. a,b Robins LN, Regier DA, editors. Psychiatric disorders in America: the epidemiologic catchment area study. New York: The Free Press; 1991.
  30. ^ Andrews G, Peters L. The psychometric properties of the Composite International Diagnostic Interview. Soc Psychiatry Psychiatr Epidemiol. 1998 Feb;33(2):80-8.
  31. ^ Murphy JM, Monson RR, Laird NM, Sobol AM, Leighton AH. A comparison of diagnostic interviews for depression in the Stirling County study: challenges for psychiatric epidemiology. Arch Gen Psychiatry. 2000 Mar;57(3):230-6.
  32. ^ Berkman LF, Glass T, Brissette I, Seeman TE. From social integration to health: Durkheim in the new millennium. Soc Sci Med. 2000 Sep;51(6):843-57.
  33. a,b Cohen S, Doyle WJ, Skoner DP, Rabin BS, Gwaltney JM Jr. Social ties and susceptibility to the common cold. JAMA. 1997 Jun 25; 277(24):1940-4.
  34. ^ Institut de la statistique du Québec. Enquête sociale et de santé 1998 [Internet]. Quebec (QC). Institut de la statistique du Quebec. Chapter 25. Le milieu social. 2001 Jan [cited on 2009 Nov 12]. Available from: publications/sante/e_soc-sante98_pdf.htm
  35. ^ Institut de la statistique du Québec. Lexique du système du code géographique du Québec. Institut de la statistique du Québec [Internet]. 2007 Sep [cited 2009 Nov 12]. Available from:
  36. ^ Hosmer DW, Lemeshow S. Applied logistic regression. Toronto: Wiley; 1989.
  37. ^ Okabayashi H, Liang J, Krause N, Akiyama H, Sugisawa H. Mental health among older adults in Japan: do sources of social support and negative interaction make a difference? Soc Sci Med. 2004 Dec;59(11):2259-70.
  38. a,b Statistics Canada. Caring Canadians, involved Canadians: highlights from the 2007 Canada survey of giving, volunteering and participating [Internet]. Ottawa (ON): Statistics Canada; 2009 Jun [cited on 2009 Nov 12]. Catalogue No.: 71-542-XPE. Available from:
  39. ^ Li Y, Ferraro KF. Volunteering and depression in later life: social benefit or selection processes? J Health Soc Behav. 2005 Mar;46(1):68-84.
  40. ^ Butrica BA, Johnson RW, Zedlewski SR. Volunteer transitions among older Americans. The Retirement Project [Internet]. Washington (DC): The Urban Institute; 2007 Oct [cited on 2009 Nov 12]. Available from: volunteer_transitions.pdf
  41. ^ Beekman AT, Penninx BW, Deeg DJ, Ormel J, Braam AW, Van Tilburg W. Depression and physical health in later life: results from the Longitudinal Aging Study Amsterdam (LASA). J Affect Disord. 1997 Dec;46(3):219-31.
  42. ^ Canadian Study of Health and Aging Working Group. Canadian study of health and aging: study methods and prevalence of dementia. CMAJ. 1994 Mar 15; 150(6):899-913.
  43. ^ Van Beijsterveldt CE, van Boxtel MP, Bosma H, Houx PJ, Buntinx F, Jolles J. Predictors of attrition in a longitudinal cognitive aging study: the Maastricht Aging Study (MAAS). J Clin Epidemiol. 2002 Mar;55(3):216-23.
Report a problem or mistake on this page
Please select all that apply:

Thank you for your help!

You will not receive a reply. For enquiries, contact us.

Date modified: