Abstract

Purpose: to investigate if the increased risk of disability onset among older people who live alone could possibly be moderated by either high social participation or by being satisfied with the social relations.

Design and methods: logistic regression models were tested using two waves in a study population of 2,697 non-disabled older men and women from The Danish Longitudinal Study on Preventive Home Visits.

Results: living alone and low social participation were significant risk factors for later male disability onset. Not being satisfied with the social relations was significantly associated with onset of disability for both genders. Among men who lived alone low social participation was a significant predictor of disability onset [odds ratio, OR = 2.30 (1.00–5.29)]; for cohabiting men social participation was not associated with disability onset, [adjusted OR = 0.91 (0.49–1.71)]. Similar results were present concerning satisfaction with the social relations among men. There was no significant interaction for women.

Conclusions: the study suggests that men who live alone can possibly alleviate their risk of disability onset by being socially active and by having access to satisfactory social relations. Women do not seem to benefit as much from cohabitation as men, although women who live alone and who are not satisfied with their social relations also constitute a significant risk category.

Introduction

During the last decades, a growing focus in gerontological research has been on factors that influence the pathway from disease to disability in older adults. Disability is defined as ‘limitations in performance of socially defined roles and tasks within a sociocultural and physical environment’ and results from the interaction of disease, impairment and functional limitations with the environmental challenges experienced by the individual [1]. It is most often measured as the ability to perform mobility and activities of daily living. Social relations may be regarded as an important environmental challenge as they play an important role in maintaining independence and the participation in family and community activities [1]. Being married/cohabiting has been shown to be protective against disability onset among both genders [2, 3], among men only [4, 5] and women only [6]. Other aspects of social relations have been associated with health outcomes among old people. For example, an association between low social participation and greater disability, and disability onset has been suggested [7, 8]. Satisfaction with the social relations has been suggested to influence mental well-being and depressive symptoms among older adults [9]; however, its association with disability onset is less well described.

It is unknown whether the deleterious effects of living alone on disability onset can be alleviated by other positive aspects of social relations. The purpose of the present study is to investigate if the increased risk of disability onset among older people who are living alone can be moderated by either high social participation or by satisfaction with the social relations. As earlier studies have identified possible gender differences in the effect of social relations on health outcomes [10], we are furthermore interested in studying the possible gender differences in the interaction between cohabitation status and each of the two other measures of social relations.

Methods

Study population

This study is based on secondary analyses on data from a randomised intervention study on preventive home visits, which is described in detail elsewhere [11]. Baseline and follow-up data were collected in 1998/1999, 2000, 2001–02 and 2003, all by mailed questionnaires. The study population included all non-institutionalised citizens in 34 Danish municipalities born in 1918 (80 years old) or 1923/1924 (75 years old), N = 4,060 (participation rate 70%) [11, 12].

Social relations of older individuals are influenced by their functional ability, e.g. impaired persons will be less socially active but will potentially receive more instrumental support. In order to keep the causal pathway clear, we restricted our study population to those with good functional ability at baseline [2, 8]. At baseline, 3,144 participants had good functional ability. Of these, 2,697 survived and answered the questions included in our analyses at 3-year follow-up.

Variables

Disability was estimated by the Mob-Help Scale, a validated scale [13] that measures the number of mobility activities performed without need of help (0–6); transfer, walking indoors, going outdoors, walking outdoors in nice/poor weather and climbing stairs [13, 14]. Mobility as measured by the Mob-Help Scale is strongly associated with isometric muscle strength, simple function tests [15] and postural balance [16]. An earlier study on the stability of the measure using the Rasch analysis showed that all participants at a given index score on the Mob-Help Scale had the same probability of scoring on an item regardless of age, gender, household composition and level of self-rated health [13]. Furthermore, we have found the same patterns of associations between the relevant social relations variables and onset of disability when we used the Mob-Help Scale as a continuous and as a dichotomous scale [17]. In the present study, the Mob-Help Scale was included as a dichotomised variable; manage all activities without help vs need of help in one or more activities (0, 1+). This cut point was used because of the clinical relevance for the old individuals of being able to manage all the mobility activities without help. Absence of disability at baseline was the inclusion criterion and absence/presence of disability at 3-year follow-up in 2001–02 was the outcome.

In order to obtain combined variables (with cohabitation status) with sufficient numbers in each category, the measures of satisfaction with the social relations and social participation were both dichotomised in our analyses. ‘Cohabitation status’ was measured at baseline by the question ‘Do you live alone?’ (yes/no). ‘Social participation’ was measured at baseline by two questions: ‘Have you had anyone visit you in your home within the past month?’ and ‘Have you yourself visited others within the past month?’ (no; yes, once or twice; yes, several times). These two variables were combined and categorised into ‘low social participation’: ‘no’ to both questions or ‘no’ to one and ‘yes, once or twice’ to the other question, otherwise the coding was ‘high’. ‘Satisfaction with the social relations’ at baseline was measured by one question: ‘How satisfied are you, all in all with the contacts you have with other people?’ (very satisfied; reasonably satisfied; neither satisfied nor dissatisfied; a little dissatisfied; very dissatisfied). The variable was dichotomised: very satisfied vs any other answer.

Two combined variables (cohabitation status/social participation, cohabitation status/satisfaction with the social relations) were constructed: (i) cohabiting with either high social participation or being ‘very satisfied’ with the social relations, (ii) living alone with either high social participation or being ‘very satisfied’ with the social relations, (iii) cohabiting with low social participation or being ‘not very satisfied’ with the social relations and (iv) living alone with either low social participation or being ‘not very satisfied’ with the social relations.

‘Age’ (75 or 80 years old at baseline), ‘intervention status’, ‘financial assets’, ‘mental well-being’ and ‘physical activity’ were included in the analyses as potential confounders. ‘Financial assets’ in 1999 were used as a measure of socioeconomic status. Financial assets have been suggested to be more important than other measures of socioeconomic status when regarding older people [18, 19]. Furthermore, social support has been shown to vary with socioeconomic position among older adults [20, 21]. Data were obtained by merging the participants’ civil registration number to Statistics Denmark. Financial assets were based on bonds and stocks deposited in financial institutions and mortgage debt, and were categorised into five groups. ‘Mental well-being’ has been suggested to predict subsequent functional ability [22] and was estimated at baseline by questions concerning feeling down, being aggressive and feeling tired without a specific reason (often/sometimes vs rarely/never) [23]. Answering rarely or never was assigned one point (0–3). In these analyses, a dichotomised variable was used: good mental well-being (2–3) vs poor mental well-being (0–1). ‘Physical activity’ was estimated at baseline by two questions regarding moderate and vigorous activity, respectively. (i) Do you get any moderate exercise (e.g. light gardening, short walks, short bike rides, housekeeping?); (ii) Do you get any vigorous exercise (e.g. sports, workout, dancing, long bike rides, long walks, heavy gardening). The response categories were: no, never; yes, seldom (not every month); yes, once or twice a month; yes, several times a month; and yes, several times a week (1–5).

Statistical analyses

Crude and multivariate logistic regression analysis was performed using SAS version 9 (PROC LOGISTIC). In the logistic regression analyses, model 1 describes the crude association between each of the two combined variables: cohabitation status/social participation, cohabitation status/satisfaction with social relations and incident disability at 3-year follow-up. The final multivariate model for each analysis was assessed by backwards elimination from a full model including: age, financial assets, intervention status, mental well-being, moderate and vigorous physical activity, and either social participation (in analysis of the effect of satisfaction with social relations) or satisfaction with the social relations (in the analyses of the effect of social participation) at baseline. Covariates with P > 0.15 were excluded from the final model in order to obtain the most statistically simple models. All analyses performed were gender specific. We tested the possible statistical interaction between cohabitation status and either social participation or satisfaction with the social relations by including an interaction term in the multivariate model and by stratification on cohabitation status in gender-specific analyses.

Results

Fourteen percent of men and 23% of women experienced onset of disability by the 3-year follow-up. A quarter of the men and well over half of the women lived alone at baseline. Significantly more men than women had low social participation and were not very satisfied with their social relations. Combining cohabitation status and social participation showed that the majority (69%) of the males were cohabiting and had a high social participation. Nineteen percent of the men and 56% of the women lived alone and had high social participation. Among both men and women, 3–4% lived alone and had a low social participation. The combination of cohabitation status and satisfaction with social relations showed that almost 60% of the men and >30% of the women lived with somebody and were very satisfied with their social relations. Sixteen percent of the men lived alone and were very satisfied with their social relations, whereas this was the case for almost half of the women. One-fifth of the men but less than a tenth of the women were cohabiting and not satisfied. Seven percent of the men and 14% of the women lived alone and were not very satisfied. The gender differences in the two combined measures of cohabitation status and social relations were highly significant. There were no gender differences in age and intervention status. Significantly more women experienced poor mental well-being, had low financial assets and never exercised vigorously (Table 1).

Table 1

Distribution of baseline covariates by gender (N = 2,796)

  Men (n = 1,222) Women (n = 1,475) 
Cohabitation statusa 
 Not living alone 941 (77%) 608 (41%) 
 Living alone 281 (23%) 867 (59%)*** 
Social participationa 
 High 1,075 (88%) 1,376 (93%) 
 Low 147 (12%) 99 (7%)*** 
Satisfaction with social relationsa 
 Very satisfied 897 (73%) 1,126 (76%) 
 Reasonably satisfied, neither satisfied nor dissatisfied, a little dissatisfied or very dissatisfied with social relations 325 (27%) 349 (24%)* 
Cohabitation status and social participationa 
 Cohabiting with high social participation 839 (69%) 556 (38%) 
 Living alone with high social participation 236 (19%) 820 (56%) 
 Cohabiting with low social participation 102 (8%) 52 (4%) 
 Living alone with low social participation 45 (4%) 47 (3%)*** 
Cohabitation status and satisfaction with social relationsa 
 Cohabiting and very satisfied 703 (58%) 466 (32%) 
 Living alone and very satisfied 194 (16%) 660 (45%) 
 Cohabiting not very satisfied 238 (20%) 142 (10%) 
 Living alone not very satisfied 87 (7%) 207 (14%)*** 
Agea 
 75 years old 923 (76%) 1,093 (74%) 
 80 years old 290 (24%) 382 (26%) 
Financial assets (1999)b 
 Less than $0 59 (5%) 85 (6%) 
 $0–6,000 152 (12%) 367 (25%) 
 $6,001–53,000 228 (19%) 455 (31%) 
 $53,001–131,000 370 (30%) 307 (21%) 
 More than $131,000 413 (34%) 261 (18%)*** 
Mental well-beinga 
 Good 1,098 (89%) 1,208 (82%) 
 Poor 134 (11%) 267 (18%)*** 
Physical activity (moderate)b 
 Never 19 (2%) 29 (2%) 
 Yes, rarely (not every month) 22 (2%) 24 (2%) 
 Yes, once or twice a month 39 (3%) 39 (3%) 
 Yes, several times a month 193 (16%) 167 (11%) 
 Yes, several times a week 949 (78%) 1,216 (82%)** 
Physical activity (vigorous)b 
 Never 250 (21%) 495 (34%) 
 Yes, rarely (not every month) 123 (10%) 100 (7%) 
 Yes, once or twice a month 122 (10%) 119 (8%) 
 Yes, several times a month 265 (22%) 276 (19%) 
 Yes, several times a week 462 (38%) 485 (33%)*** 
Municipalitya 
 Intervention 638 (52%) 770 (52%) 
 Control 584 (48%) 705 (48%) 
  Men (n = 1,222) Women (n = 1,475) 
Cohabitation statusa 
 Not living alone 941 (77%) 608 (41%) 
 Living alone 281 (23%) 867 (59%)*** 
Social participationa 
 High 1,075 (88%) 1,376 (93%) 
 Low 147 (12%) 99 (7%)*** 
Satisfaction with social relationsa 
 Very satisfied 897 (73%) 1,126 (76%) 
 Reasonably satisfied, neither satisfied nor dissatisfied, a little dissatisfied or very dissatisfied with social relations 325 (27%) 349 (24%)* 
Cohabitation status and social participationa 
 Cohabiting with high social participation 839 (69%) 556 (38%) 
 Living alone with high social participation 236 (19%) 820 (56%) 
 Cohabiting with low social participation 102 (8%) 52 (4%) 
 Living alone with low social participation 45 (4%) 47 (3%)*** 
Cohabitation status and satisfaction with social relationsa 
 Cohabiting and very satisfied 703 (58%) 466 (32%) 
 Living alone and very satisfied 194 (16%) 660 (45%) 
 Cohabiting not very satisfied 238 (20%) 142 (10%) 
 Living alone not very satisfied 87 (7%) 207 (14%)*** 
Agea 
 75 years old 923 (76%) 1,093 (74%) 
 80 years old 290 (24%) 382 (26%) 
Financial assets (1999)b 
 Less than $0 59 (5%) 85 (6%) 
 $0–6,000 152 (12%) 367 (25%) 
 $6,001–53,000 228 (19%) 455 (31%) 
 $53,001–131,000 370 (30%) 307 (21%) 
 More than $131,000 413 (34%) 261 (18%)*** 
Mental well-beinga 
 Good 1,098 (89%) 1,208 (82%) 
 Poor 134 (11%) 267 (18%)*** 
Physical activity (moderate)b 
 Never 19 (2%) 29 (2%) 
 Yes, rarely (not every month) 22 (2%) 24 (2%) 
 Yes, once or twice a month 39 (3%) 39 (3%) 
 Yes, several times a month 193 (16%) 167 (11%) 
 Yes, several times a week 949 (78%) 1,216 (82%)** 
Physical activity (vigorous)b 
 Never 250 (21%) 495 (34%) 
 Yes, rarely (not every month) 123 (10%) 100 (7%) 
 Yes, once or twice a month 122 (10%) 119 (8%) 
 Yes, several times a month 265 (22%) 276 (19%) 
 Yes, several times a week 462 (38%) 485 (33%)*** 
Municipalitya 
 Intervention 638 (52%) 770 (52%) 
 Control 584 (48%) 705 (48%) 

*0.05 < P < 0.1, **P = 0.07, ***P < 0.0001.

a

Chi-square test.

b

Chi-square test for trend.

In unadjusted analyses, men who lived alone, had low social participation or who were not very satisfied with their social relations had a significantly increased odds ratio of disability onset compared with the unexposed men. For women, not being satisfied with the social relations was significantly associated with disability onset (Table 2).

Table 2

Crude odds ratios (95% CI) for onset of disability during the 3-year follow-up by cohabitation status, socialparticipation and satisfaction with social relations

Men N = 1,222 
Cohabitation status 
 Not living alone 1.00 
 Living alone 1.75 (1.23–2.49) 
Social participation 
 High 1.00 
 Low 1.80 (1.17–2.78) 
Satisfaction with social relations 
 Very satisfied 1.00 
 Reasonably satisfied, neither satisfied nor dissatisfied, a little dissatisfied or very dissatisfied with social relations 1.57 (1.11–2.22) 
    
Women N = 1,475 
Cohabitation status 
 Not living alone 1.00 
 Living alone 1.23 (0.96–1.58) 
Social participation 
 High 1.00 
 Low 1.18 (0.74–1.88) 
Satisfaction with social relations 
 Very satisfied 1.00 
 Reasonably satisfied, neither satisfied nor dissatisfied, a little dissatisfied or very dissatisfied with social relations 1.70 (1.30–2.24) 
Men N = 1,222 
Cohabitation status 
 Not living alone 1.00 
 Living alone 1.75 (1.23–2.49) 
Social participation 
 High 1.00 
 Low 1.80 (1.17–2.78) 
Satisfaction with social relations 
 Very satisfied 1.00 
 Reasonably satisfied, neither satisfied nor dissatisfied, a little dissatisfied or very dissatisfied with social relations 1.57 (1.11–2.22) 
    
Women N = 1,475 
Cohabitation status 
 Not living alone 1.00 
 Living alone 1.23 (0.96–1.58) 
Social participation 
 High 1.00 
 Low 1.18 (0.74–1.88) 
Satisfaction with social relations 
 Very satisfied 1.00 
 Reasonably satisfied, neither satisfied nor dissatisfied, a little dissatisfied or very dissatisfied with social relations 1.70 (1.30–2.24) 

In the multivariate logistic regression analyses, men who lived alone and either had a low social participation or were not very satisfied with their social relations had a significantly three times higher odds ratio for disability onset compared with the cohabiting men with high social participation or cohabiting men who were satisfied with their social relations. Inclusion of covariates attenuated these associations; however, they remained statistically significant. Among men, a significantly increased risk of disability onset was also seen among those living alone who had a high social participation. This association remained close to significance after adjustment for covariates [odds ratio, OR = 1.50 (0.98–2.29)]. A significantly increased risk of disability onset was present only among women who lived alone and who were not very satisfied with their social relations in the fully adjusted models (Table 3).

Table 3

Odds ratios (95% CI) for disability onset (3-year follow-up) by cohabitation status/social participation and cohabitation status/satisfaction with social relations at baseline, stratified by gender

  Model 1 Model 2a 
Men N = 1,222 
Cohabiting + high social participation 1.00 1.00 
Living alone + high social participation 1.53 (1.02–2.28) 1.50 (0.98–2.29) 
Cohabiting + low social participation 1.39 (0.78–2.47) 0.91 (0.49–1.70) 
Living alone + low social participation 3.74 (1.94–7.19) 3.20 (1.57–6.51) 
      
  Model 1 Model 2a 
Cohabiting + satisfied with soc. rel. 1.00 1.00 
Living alone + satisfied with soc. rel. 1.33 (0.84–2.10) 1.30 (0.80–2.11) 
Cohabiting + not satisfied with soc. rel. 1.22 (0.79–1.88) 0.85 (0.52–1.38) 
Living alone + not satisfied with soc. rel. 3.23 (1.93–5.40) 2.60 (1.48–4.56) 
      
  Model 1 Model 2b 
Women N = 1,475 
Cohabiting + high social participation 1.00 1.00 
Living alone + high social participation 1.22 (0.94–1.58) 1.13 (0.86–1.50) 
Cohabiting + low social participation 1.05 (0.52–2.11) 0.67 (0.32–1.40) 
Living alone + low social participation 1.66 (0.86–3.21) 1.02 (0.49–2.10) 
  Model 1 Model 2c 
Cohabiting + satisfied with soc. rel. 1.00 1.00 
Living alone + satisfied with soc. rel. 1.15 (0.85–1.54) 1.11 (0.81–1.52) 
Cohabiting + not satisfied with soc. rel. 1.46 (0.94–2.27) 1.30 (0.82–2.08) 
Living alone + not satisfied with soc. rel. 2.15 (1.48–3.11) 1.74 (1.17–2.58) 
  Model 1 Model 2a 
Men N = 1,222 
Cohabiting + high social participation 1.00 1.00 
Living alone + high social participation 1.53 (1.02–2.28) 1.50 (0.98–2.29) 
Cohabiting + low social participation 1.39 (0.78–2.47) 0.91 (0.49–1.70) 
Living alone + low social participation 3.74 (1.94–7.19) 3.20 (1.57–6.51) 
      
  Model 1 Model 2a 
Cohabiting + satisfied with soc. rel. 1.00 1.00 
Living alone + satisfied with soc. rel. 1.33 (0.84–2.10) 1.30 (0.80–2.11) 
Cohabiting + not satisfied with soc. rel. 1.22 (0.79–1.88) 0.85 (0.52–1.38) 
Living alone + not satisfied with soc. rel. 3.23 (1.93–5.40) 2.60 (1.48–4.56) 
      
  Model 1 Model 2b 
Women N = 1,475 
Cohabiting + high social participation 1.00 1.00 
Living alone + high social participation 1.22 (0.94–1.58) 1.13 (0.86–1.50) 
Cohabiting + low social participation 1.05 (0.52–2.11) 0.67 (0.32–1.40) 
Living alone + low social participation 1.66 (0.86–3.21) 1.02 (0.49–2.10) 
  Model 1 Model 2c 
Cohabiting + satisfied with soc. rel. 1.00 1.00 
Living alone + satisfied with soc. rel. 1.15 (0.85–1.54) 1.11 (0.81–1.52) 
Cohabiting + not satisfied with soc. rel. 1.46 (0.94–2.27) 1.30 (0.82–2.08) 
Living alone + not satisfied with soc. rel. 2.15 (1.48–3.11) 1.74 (1.17–2.58) 
a

Adjusted for baseline: age, mental well-being, financial assets, moderate and vigorous physical activity, intervention/control municipality.

b

Adjusted for baseline: age, moderate and vigorous physical activity, intervention/control municipality, satisfaction with social relations.

c

Adjusted for baseline: age, moderate and vigorous physical activity, intervention/control municipality.

The inclusion of interaction terms between cohabitation status and either social participation or satisfaction with the social relations revealed significant interactions for men but not for women.

Stratified analyses showed that, among cohabiting men, social participation did not influence later onset of disability [adjusted OR = 0.91 (0.49–1.71)], whereas men who lived alone had a significantly increased risk of disability onset if they also had low social participation [adjusted OR = 2.30 (1.00–5.29)]. Similar results were present concerning satisfaction with the social relations for men: ORcohabiting men = 0.82 (0.50–1.35) and ORnot cohabiting men = 2.00 (0.99–4.06). Among women, there was no effect of social participation on onset of disability in either strata of cohabitation status. Not being very satisfied with the social relations was associated with higher odds of disability onset among both the cohabiting [OR = 1.35 (0.83–2.21)] and for those who lived alone [OR = 1.47 (1.00–2.16)].

From baseline to follow-up, 176 respondents died. One may argue that dying is the ultimate loss of functional ability, and it is likely that the excluded individuals have experienced a period of disability onset before dying; they just had this experience before follow-up. By excluding individuals who have deteriorated faster, important information will be lost. In order to deal with the risk of attrition due to this exclusion procedure, we performed all analyses including individuals who died during follow-up (N = 2,950). Overall, the results were in the same direction (data not shown).

Discussion

In summary, this study showed gender differences in the association between cohabitation status, social participation and satisfaction with social relations, respectively, and disability onset. The significant interactions between cohabitation status and each of the other variables for social relations suggest that men who live alone can possibly alleviate their risk of disability onset by having higher social participation or by having access to satisfactory social relations.

In contrast, women do not seem to be as vulnerable to the effects of living alone as men. However, being both not very satisfied with the social relations and living alone also constitutes a significant risk category for disability onset for women.

The findings suggest that social participation and access to satisfactory social relations could be intervention candidates to alleviate the hazardous effects of living in a single-person household for men.

Generally, men seem to benefit more from spousal support and women more from support from others [24, 25]. This is in line with our main findings of cohabitation status as the most important predictor of disability onset among men, which also support the findings of earlier studies where marriage/not living alone was found to protect against disability onset among men [4, 5, 26]. We only found a protective effect of social participation for men. Some studies indicate that the social relations tend to change towards the core of the network in old age [27], i.e. older persons tend to increase their social interaction with children and grandchildren and limit connections with friends and acquaintances. Interaction with children includes both positive and negative social exchange, and women seem to react more to negative social exchanges than men [28]. With this background, the present results may also indicate that women who live alone do not benefit from social participation to the same degree as their male counterparts because they experience more strain from meeting with the close network.

Measuring overall satisfaction with the social relations with just one item may be problematic. Earlier studies have included measures of satisfaction with the social relations employing several items, e.g. satisfaction specifically related to the perceived tangible and emotional and informational support [9, 29]. In the present study, the intention was to describe the overall function of social relations, and hence to describe if the support the respondent perceived from the social relations was experienced as sufficient. The categorisation of responses (very satisfied vs other responses) was based on the general reluctance of older people to express dissatisfaction [30]. Consequently, the ‘reasonably satisfied’ were classified as ‘not satisfied’. Very few participants (5–6%) were less than ‘very satisfied with social relations’. In addition, we performed analyses dichotomising between ‘very satisfied’/‘reasonably satisfied’ vs other responses and found stronger estimates for disability onset among men and weaker estimates for women. The final conclusions for men did, however, not alter substantially. For women, the significant association between cohabitation status/satisfaction with the social relations and disability onset disappeared after adjusting for covariates.

The present study reached a high response rate (70%). Furthermore, data are longitudinal in nature, which make it possible to better evaluate the causality of the studied associations. Poor health influences the social relations [31]. Consequently, to examine the impact of social relations on disability onset, it was important to only include participants without disability at baseline. Other strengths include the well-validated measure of functional ability [13] and the opportunity to obtain data on socioeconomic status for all participants from registers at Statistics Denmark.

We included physical activity as a possible confounder in our models since low physical activity may be predictive of social participation and also satisfaction with social relations as well as of disability onset. We are aware that low physical activity may also be a result of e.g. low social participation and thereby a possible mediating factor, which should not be adjusted for. We chose to keep physical activity in our models and treated the variable as a potential confounder. Physical health at baseline is a candidate for confounding since it may influence the social relations and increase risk of disability onset. Unfortunately, we did not have access to a measure of physical health at baseline (or even better before baseline) and consequently our results may have been biased by residual confounding from this factor. However, physical health measured at baseline may also be a mediating factor between social relations and disability onset, which consequently should not be adjusted for. Furthermore, it is well known that the measure of disability summarises most of the consequences of poor physical health [32]. Thus, by excluding those with disability at baseline, we have strongly reduced the risk of confounding from physical health.

Individuals excluded due to missing values on the selected variables were more often males, belonged to the older age group, more often lived alone, had more often poor mental well-being and had lower financial assets. We believe that the exclusion of these individuals with a high risk profile for disability onset may have underestimated the effects in our analyses.

Conclusion

Men seem to benefit more from a cohabiting partner than women when it comes to risk of disability onset. Significant interaction between cohabitation status and each of the other measures of social relations was suggested for men but not for women. Consequently, men who live alone can possibly alleviate their risk of disability onset by being socially active and by having access to satisfactory social relations. Women do not seem to benefit as much from cohabitation as men, although women who live alone and who are not satisfied with their social relations also constitute a significant risk category.

A future challenge for our society will be the special care needs of the older people who live alone. Future interventions to increase social participation and access to satisfactory social relations especially among men who live alone may delay their risk of disability onset.

Key points

  • This study showed gender differences in the association between cohabitation status, social participation and satisfaction with social relations, respectively, and disability onset.

  • The significant interactions between cohabitation status and each of the other variables for social relations suggest that men who live alone can possibly alleviate their risk of disability onset by having a higher social participation or by having access to satisfactory social relations.

  • Women do not seem to be as vulnerable to the effects of living alone as men. However, being both not very satisfied with the social relations and living alone also constitutes a significant risk category for disability onset for women.

We thank MD Mikkel Vass who has been responsible for the study concept, design and acquisition of data in the Danish Intervention Study of Preventive Home Visits as principal investigator together with K.A. We thank Dr Natalie Chan for language editing.

Funding

This study is supported by grants from The Danish National Council for Health Research, The Research Foundation for General Practice and Primary Care, East Danish Research Forum, The County Value-Added Tax Foundation and The Danish Ministry of Social Affair.

References

1
Guralnik
JM
Ferrucci
L
Assessing the building blocks of function: utilizing measures of functional limitation
Am J Prev Med
 , 
2003
, vol. 
25
 (pg. 
112
-
21
)
2
Mor
V
Murphy
J
Masterson-Allen
S
, et al.  . 
Risk of functional decline among well elders
J Clin Epidemiol
 , 
1989
, vol. 
42
 (pg. 
895
-
904
)
3
Waite
LJ
Hughes
ME
At risk on the cusp of old age: living arrangements and functional status among black, white and Hispanic adults
J Gerontol B Psychol Sci Soc Sci
 , 
1999
, vol. 
54
 (pg. 
S136
-
44
)
4
Strawbridge
WJ
Camacho
TC
Cohen
RD
Kaplan
GA
Gender differences in factors associated with change in physical functioning in old age: a 6-year longitudinal study
Gerontologist
 , 
1993
, vol. 
33
 (pg. 
603
-
9
)
5
Nilsson
CJ
Lund
R
Avlund
K
Cohabitation status and onset of disability among older Danes: is social participation a possible mediator?
J Aging Health
 , 
2008
, vol. 
20
 (pg. 
235
-
53
)
6
Avlund
K
Due
P
Holstein
BE
Sonn
U
Laukkanen
P
Changes in household composition as determinant of changes in functional ability among old men and women
Aging Clin Exp Res
 , 
2002
, vol. 
14
 (pg. 
65
-
74
)
7
Mendes de Leon
CF
Glass
TA
Berkman
LF
Social engagement and disability in a community population of older adults: the New Haven EPESE
Am J Epidemiol
 , 
2003
, vol. 
157
 (pg. 
633
-
42
)
8
Avlund
K
Lund
R
Holstein
BE
Due
P
Social relations as determinant of onset of disability in aging
Arch Gerontol Geriatr
 , 
2004
, vol. 
38
 (pg. 
85
-
99
)
9
Clay
OJ
Roth
DL
Wadley
VG
Haley
WE
Changes in social support and their impact on psychosocial outcome over a 5-year period for African American and White dementia caregivers
Int J Geriatr Psychiatr
 , 
2008
, vol. 
23
 (pg. 
857
-
62
)
10
Shumaker
SA
Hill
DR
Gender differences in social support and physical health. [Review] [84 refs]
Health Psychol
 , 
1991
, vol. 
10
 (pg. 
102
-
11
)
11
Vass
M
Avlund
K
Hendriksen
C
Andersen
CK
Keiding
N
Preventive home visits to older people in Denmark
Aging Clin Exp Res
 , 
2002
, vol. 
14
 
12
Vass
M
Avlund
K
Lauridsen
J
Hendriksen
C
Feasible model for prevention of functional decline in older people: municipality-randomized, controlled trial
J Am Geriatr Soc
 , 
2005
, vol. 
53
 (pg. 
563
-
8
)
13
Avlund
K
Kreiner
S
Schultz-Larsen
K
Functional ability scales for the elderly. A validation study
Eur J Publ Health
 , 
1996
, vol. 
6
 (pg. 
35
-
42
)
14
Avlund
K
Holstein
BE
Functional ability among elderly people in three service settings: the discriminatory power of a new functional ability scale
Eur J Epidemiol
 , 
1998
, vol. 
14
 (pg. 
783
-
90
)
15
Avlund
K
Schroll
M
Davidsen
M
Løvborg
B
Rantanen
T
Maximal isometric muscle strength and functional ability in daily activities among 75-year-old men and women
Scand J Med Sci Sports
 , 
1994
, vol. 
4
 (pg. 
32
-
40
)
16
Era
P
Avlund
K
Jokela
J
, et al.  . 
Postural balance and self-reported functional ability in 75-year-old men and women: a cross-national comparative study
J Am Geriatr Soc
 , 
1997
, vol. 
45
 (pg. 
21
-
9
)
17
Avlund
K
Lund
R
Holstein
BE
Due
P
Social relations as determinant of onset of disability in aging
Arch Gerontol Geriatr
 , 
2004
, vol. 
38
 (pg. 
85
-
99
)
18
Mackenbach
JP
Borsboom
GJ
Nusselder
WJ
Looman
CW
Schrijvers
CT
Determinants of levels and changes of physical functioning in chronically ill persons: results from the GLOBE Study
J Epidemiol Community Health
 , 
2001
, vol. 
55
 (pg. 
631
-
8
)
19
Robert
S
House
JS
SES differentials in health by age and alternative indicators of SES
J Aging Health
 , 
1996
, vol. 
8
 (pg. 
359
-
88
)
20
Ajrouch
KJ
Blandon
AY
Antonucci
TC
Social networks among men and women: the effects of age and socioeconomic status
J Gerontol B Psychol Sci Soc Sci
 , 
2005
, vol. 
60
 (pg. 
S311
-
7
)
21
Krause
N
Borawski-Clark
E
Social class differences in social support among older adults
Gerontologist
 , 
1995
, vol. 
35
 (pg. 
498
-
508
)
22
Raji
MA
Ostir
GV
Markides
KS
Goodwin
JS
The interaction of cognitive and emotional status on subsequent physical functioning in older Mexican Americans: findings from the Hispanic established population for the epidemiologic study of the elderly
J Gerontol A Biol Sci Med Sci
 , 
2002
, vol. 
57
 (pg. 
M678
-
82
)
23
Avlund
K
Vass
M
Hendriksen
C
Onset of mobility disability among community-dwelling old men and women. The role of tiredness in daily activities
Age Ageing
 , 
2003
, vol. 
32
 (pg. 
579
-
84
)
24
Fuhrer
R
Stansfeld
SA
How gender affects patterns of social relations and their impact on health: a comparison of one or multiple sources of support from “close persons”
Soc Sci Med
 , 
2002
, vol. 
54
 (pg. 
811
-
25
)
25
Antonucci
TC
Akiyama
H
An examination of sex differences in social support among older men and women
Sex Roles
 , 
1987
, vol. 
17
 (pg. 
737
-
49
)
26
Goldman
N
Korenman
S
Weinstein
R
Marital status and health among the elderly
Soc Sci Med
 , 
1995
, vol. 
40
 (pg. 
1717
-
30
)
27
van Tilburg
T
Losing and gaining in old age: changes in personal network size and social support in a four-year longitudinal study
J Gerontol B Psychol Sci Soc Sci
 , 
1998
, vol. 
53B
 (pg. 
S313
-
23
)
28
Okun
MA
Keith
VM
Effects of positive and negative social exchanges with various sources on depressive symptoms in younger and older adults
J Gerontol B Psychol Sci Soc Sci
 , 
1998
, vol. 
53
 (pg. 
4
-
20
)
29
Krause
N
Negative interaction and satisfaction with social support among older adults
J Gerontol B Psychol Sci Soc Sci
 , 
1995
, vol. 
50
 (pg. 
59
-
73
)
30
Due
P
“When you are alone you can do as you want” Course of life and health among elderly, poor Danish women living alone. Life situation among elderly women on lowest superannuation in Denmark, Norway and Sweden. [Danish]
Norwegian Institute of Gerontology
 , 
1993
Oslo
Norwegian Institute of Gerontology
(pg. 
21
-
120
)
31
Lund
R
Avlund
K
Modvig
J
Due
P
Holstein
BE
Development in self-rated health among older people as determinant of social relations
Scand J Publ Health
 , 
2004
, vol. 
32
 (pg. 
419
-
25
)
32
Avlund
K
Rantanen
T
Schroll
M
Factors underlying tiredness in older adults
Aging Clin Exp Res
 , 
2007
, vol. 
19
 (pg. 
16
-
25
)

Comments

0 Comments