Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Apr 1.
Published in final edited form as: Aging Ment Health. 2016 Dec 22;22(4):550–557. doi: 10.1080/13607863.2016.1268095

Social networks and alcohol use among older adults: a comparison with middle-aged adults

Seungyoun Kim a, Samantha L Spilman b, Diana H Liao b, Paul Sacco c, Alison A Moore d
PMCID: PMC5523450  NIHMSID: NIHMS857084  PMID: 28006983

Abstract

Objectives

This study compared the association between social networks and alcohol consumption among middle-aged (MA) and older adults (OA) to better understand the nature of the relationship between those two factors among OA and MA.

Method

We examined Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions. Current drinkers aged over 50 were subdivided into two age groups: MA (50–64, n = 5214) and OA (65 and older, n = 3070). Each age group was stratified into drinking levels (low-risk vs. at-risk) based on alcohol consumption. The size and diversity of social networks were measured. Logistic regression models were used to examine age differences in the association between the social networks (size and diversity) and the probability of at-risk drinking among two age groups.

Results

A significant association between the social networks diversity and lower odds of at-risk drinking was found among MA and OA. However, the relationship between the diversity of social networks and the likelihood of at-risk drinking was weaker for OA than for MA. The association between social networks size and at-risk drinking was not significant among MA and OA.

Conclusion

The current study suggests that the association between social networks diversity and alcohol use among OA differs from the association among MA, and few social networks were associated with alcohol use among OA. In the future, research should consider an in-depth exploration of the nature of social networks and alcohol consumption by using longitudinal designs and advanced methods of exploring drinking networks.

Keywords: Social ties, alcohol consumption, age differences, life-span perspective

Introduction

Although alcohol use generally declines with age (Moore et al., 2005), the health risks of drinking are higher among older adults than younger adults because of the increased vulnerability to the effects of alcohol among older adults (Blow & Barry, 2002). Normal age-related physiological changes (Novier, Diaz-Granados, & Matthews, 2015), medical comorbidity (Wu & Blazer, 2014), and increased use of medication that may interact with alcohol (Cousins et al., 2014) exacerbate the negative health effects of alcohol among older adults (Sklar, Gilbertson, Boissoneault, Prather, & Nixon, 2012).

Health risks related to alcohol use among older adults are expected to increase in the future due to an aging population and higher rates of alcohol use among baby boomers (Satre, 2015). Given increasing numbers of older alcohol users and the consequences associated with unhealthy drinking, it is critical to identify and understand factors that influence alcohol consumption among the older population.

One of the key factors associated with drinking behaviors may be social networks. Social networks, defined here as ‘the size and diversity of social ties̕ (Cohen, Doyle, Skoner, Rabin, & Gwaltney, 1997), are the structural aspects of social relationships (Berkman, Glass, Brissette, & Seeman, 2000). Recent advances in research have suggested mutual influences between social networks and alcohol use (Bullers, Cooper, & Russell, 2001; Moos, Brennan, Schutte, & Moos, 2010). Specifically, social networks may reinforce or constrain drinking behaviors, and the structure of one’s social network may be a product of patterns of alcohol use.

Significant relationships between social networks and alcohol use have been demonstrated among younger adults (e.g. Seaman & Ikegwuone, 2010). Consequently, intervention programs incorporating social network factors have been developed and have been shown to be effective in preventing risky alcohol use among younger adults (Valente, 2012; Valente et al., 2007). Compared to younger adults, our knowledge regarding the association between social networks and alcohol use among older adults remains limited (Ahlström, 2008). If we deepen our knowledge and understanding of the relationship, adequate interventions can be planned and developed to prevent unhealthy drinking among older adults.

Social networks and alcohol consumption among older adults can be understood within a life-span developmental perspective (Uchino, 2009). The life-span approach emphasizes continuity and change in the nature of social networks over the life span (Committee report, 2006; Erikson, 1964). Patterns of drinking behavior are also dynamic constructs that change across the life span (Umberson, Crosnoe, & Reczek, 2010). From a long-term perspective of the life span, it is expected that social networks may be differentially related to alcohol use in different age groups as individuals experience change and variability over the life span (Nesselroade, 2001). It is, therefore, important to examine the association between social networks and drinking behaviors at different points in the life span (Watt et al., 2014).

Although most developed countries have accepted the chronological age of 65 years as a definition of an ʻolder adultʼ (World Health Organization, 2003), much of the alcohol use-related research on older adults has included participants aged 50 and older as a single age group (e.g. Lang, Wallace, Huppert, & Melzer, 2007). These studies might not identify possible differences in the association between social networks and alcohol consumption between middle-aged (MA, age 50–64) and older adults (OA, age 65 and older). Therefore, to identify the association between social networks and alcohol consumption in OA (age 65 and older) and identify differences in the association compared to the earlier life stage (age 50–64), it is necessary to subdivide individuals aged 50 and older into subgroups by age (MA group: 50 64 years; OA group: 65 years and older) (Litwin & Stoeckel, 2013). Differentiating age groups also has implications in developing intervention programs. Differences in the relationship between social networks and alcohol use among MA and OA may shed light on age-specific approaches to intervention. Given health risks related to alcohol use, targeting social networks that might be associated with unhealthy drinking would be beneficial in promoting health among MA and OA. In the present study, guided by the life-span perspective, we explored and compared the association between social networks and alcohol consumption by two age groups: MA and OA.

Age-related changes in social networks

The life-span perspective suggests that changes in social networks occur throughout all stages of the life span (Wrzus, Hanel, Wagner, & Neyer, 2013). Therefore, social network characteristics may differ according to stages of life (Ajrouch, Blandon, & Antonucci, 2005). Socioemotional selectivity theory has provided a compelling explanation of how social networks change with advancing age (Carstensen, 1992). This theory states that having a limited future-time perspective motivates individuals to maintain the most intimate and more meaningful ties and to disband non-intimate and peripheral relationships in order to maximize positive emotional experiences. Due to the limited future-time perspective, OA are more likely to have smaller and more intimate social networks compared to younger adults. There is growing evidence supporting reductions in social networks among OA, particularly among less close social partners (Ajrouch, Antonucci, & Janevic, 2001; Wrzus et al., 2013). Age-related changes in social networks may lead to changes in the association between social networks and alcohol use in later life.

Social networks and alcohol use among older adults

Current conceptual models posit a reciprocal relationship between social networks and alcohol use (Bullers et al., 2001; Moos et al., 2010). While social networks may function both as an inhibitory and facilitating influence on drinking behaviors (Perreira & Sloan, 2001), alcohol consumption may alter individuals’ social networks. The former, a social causation perspective, emphasizes that social network norms determine individual drinking behaviors. On the other hand, a social selection perspective posits that individuals’ alcohol consumption can have an impact on social networks (Moos et al., 2010).

Empirical evidence supports the existence of the reciprocal association between social networks and alcohol consumption among OA (Moos et al., 2010). To date, however, the findings of previous studies have been largely inconclusive on whether the association is positive or negative. Some studies have found that small social networks were related to less alcohol use (e.g. Bacharach, Bamberger, Cohen, & Doveh, 2007), others found that small social networks were associated with higher levels of alcohol consumption (e.g. Shiovitz-Ezra & Litwin, 2012). A few cross-sectional studies have examined the relationship between social networks and alcohol consumption among OA. Older adults (age 65 and older) with limited social networks have been shown to exhibit a greater risk of alcohol abuse (Shiovitz-Ezra & Litwin, 2012). Among individuals aged 60 and older, Watt et al. (2014) found that older male heavy drinkers are more likely to be divorced/separated (an absence of spousal relationship), and older women heavy drinkers are more likely to be single. Longitudinal studies support the long-term mutual relationships between social networks and alcohol consumption among older adults. In support of a social causation perspective, Perreira and Sloan (2001) found that at six-year follow-up, retirement and widowhood were associated with increased drinking, respectively, and marriage and divorce were associated with both increases and decreases in drinking among individuals between the ages of 51 and 61 at the baseline. Among men whose age ranged from 43 to 70, Bacharach et al. (2007)) found that retirement was associated with a decrease in the severity of drinking problems among those men with a history of problem drinking at two-year follow-up.

One explanation for these inconclusive findings from cross-sectional and longitudinal studies is that the findings may depend on the composition of the sample. With a few exceptions (e.g. Shiovitz-Ezra & Litwin, 2012), these studies consisted of individuals with a wide age range from middle age to old age (age 40 and older), who are in different life stages. As noted, the social network structure changes over the life span (Wrzus et al., 2013). These age-related changes in social networks may lead to heterogeneity in the association between social networks and alcohol use among MA and OA. Because previous studies were less likely to differentiate OA to MA, little is known about how the association between social networks and alcohol use among OA differs from that of adults in middle age.

Based on the life-span perspective, the first aim of the present study is to explore the association between social networks and alcohol consumption among two age groups: MA and OA. In this study, each age group was divided by two drinking groups (low-risk drinkers and at-risk drinkers) based on alcohol consumption. Low-risk drinkers were defined as individuals who drink within both the single-day and weekly limits, and high-risk drinkers were defined as individuals who drink more than these single-day or weekly limits (NIAAA, 2010). We categorized alcohol consumption into drinking levels (low-risk vs. at-risk) rather than examining alcohol consumption as a continuous variable because we were more interested in whether social networks are related to lower or greater probability of at-risk drinking, which will provide important clinical implications. Two different measures of social networks (size and diversity) were used in this study. Social network size is defined as the number of persons for which individuals report social interactions (Mowbray, Quinn, & Cranford, 2014). Social network diversity measures the number of types of social relationships (e.g. spouses, children). Social network researchers suggest that the size and diversity of social networks need to be differentiated because they provide different perspectives on the role of social networks. Social network size indicates social resources that can be beneficial for individualsʼ health and well-being (Wrzus et al., 2013; Erosheva, Kim, Emlet, & Fedriksen-Goldsen, 2015). Social network diversity can be considered an indicator of social network bridging, which provides connections to individuals who are otherwise not connected. Cornwell (2009) argued that social bridging is beneficial because it allows the individual to have access to a variety of resources from different social domains and to be more independent from the control of others.

To examine the association between social networks and alcohol consumption among MA and OA (aim 1), we explored the relationship between social networks (size and diversity) between low-risk drinkers and at-risk drinkers among two age groups. The second aim of this study is to examine differences in the association between social networks and alcohol consumption between two age groups. Our study tested the following hypotheses:

Hypothesis 1

The size and diversity of social networks will be significantly associated with the odds of at-risk drinking among MA and OA. The direction of the association (higher odds of at-risk drinking or lower odds of at-risk drinking) is not stated because of inconsistency in empirical work regarding the association.

Hypothesis 2

Drawing on the socioemotional selectivity theory in the sense of social networks reductions among OA, the association between social networks and the odds of at-risk drinking among OA will be weaker than that found among MA.

Methods

Data source and participants

The National Epidemiologic Survey on Alcohol and Related Conditions (NESARC: Waves 1 [2001–2002] and 2 [2004–2005]) was conducted by the National Institute of Alcohol Abuse and Alcoholism (NIAAA, Chen, Yi, Dawson, Stinson, & Grant, 2010). NESARC is among the largest nationally representative comorbidity surveys ever conducted and includes extensive questions on alcohol and drug use and related disorders. The data were collected through in-person interviews in household settings with 34,653 civilians, non-institutionalized adults (aged 18 and older) residing in the USA. From each household, one adult was selected to be interviewed. The overall response rate for NESARC was 81%. The methods, quality control procedures, and test–retest reliability have been previously published (Grant et al., 2004). The cumulative response rate over the two surveys was 70.2% (Dawson, Goldstein, & Grant, 2007). We used the second wave of the study (2004–2005), as data were collected on social networks in Wave 2 only. The data were weighted to account for oversampling, survey non-response, and to be representative of the US civilian population using the 2000 decennial census (U.S. Census Bureau, 2004). Participants of this study were restricted to current drinkers aged 50 and older (n = 8194). Non-drinkers were excluded in this study because older non-drinkers are more likely to show poorer mental and physical health outcomes (e.g. depression, more falls, increased cardiovascular mortality, poorer general health status) compared to light and moderate drinkers (Anderson, Scafato, & Galluzzo, 2012; Blow & Barry, 2002; Kirchner et al., 2007; Moriconi & Nadeau, 2015; Naimi et al., 2005). As non-drinking individuals are qualitatively different than current alcohol users due to health, they may represent a different population.

Measures

Social networks

The Social Network Index utilized by the NESARC measured two different aspects of social networks: size and diversity. The Index consists of 12 questions assessing participation in 12 types of social relationships (Cohen et al., 1997). The questions ask about the presence (yes/no) of certain types of the relationships (spouse/partner, parents, and parents-in-law) or the number of people the participants interacts with for the various types of relationships (children, friends, relative, student/teachers, co-workers, neighbors, fellow volunteers, religious group, and other group). The size of social networks was calculated by adding the number of people within each type of social relationships across all 12 groups in which participants see or talk with someone at least once every two weeks (if respondents answered ‘yes’ on a ‘yes/no’ question, it was counted as one). The diversity of social networks was created by counting the number of types of social relationships participants reported that they see or speak to for each type of relationship at least once every two weeks (score range 0–12). Higher scores in the diversity of social networks reflect greater variety of social connections.

Alcohol consumption (drinking levels)

To identify current drinkers, we used the variable ‘drank alcohol in the last 12 months of wave 2′. Participants (current drinkers) were categorized as two drinking groups: low-risk drinkers vs. at-risk drinkers. We used a variable called ‘W2EXCEED’ coded as positive if the respondent was positive for either exceeding the daily or weekly drinking limits recommended in NIAAA’s Physician Guidelines in the past 12 months (NIAAA, 2008). These guidelines for general population are gender specific. Men who drink more than 14 standard drinks per week or more than 5 standard drinks on any day are defined as at-risk drinkers. Women who drink more than seven standard drinks per week or more than four drinks on any day are defined as at-risk drinkers (NIAAA, 2008). Among MA (age 50–64), if respondents drink exceeding either daily or weekly limits based on guidelines for general population, they were categorized as at-risk drinkers. For OA, we followed independent drinking guidelines for older adults suggested by NIAAA (2010). According to the guidelines, OA (aged 65 and older) who drink more than seven standard drinks per week or more than three standard drinks on any day were defined as at-risk drinkers in this study. Based on the risk factors associated with alcohol use by older adults, drinking guidelines for this population are lower than those set for other adults. In addition, the guidelines for older adults should be no more liberal to men because older men who exceed drinking guidelines are more likely to have drinking problems than are older women (Moos, Schutte, Brennan, & Moos, 2004).

Covariates

Socio-demographic correlates

Participants were assessed for several socio-demographic characteristics including age (in years), gender, race/ethnicity (White, Black, American Indian/Alaska Native, Asian/Pacific Islander, and Hispanic), household income (from $0–19,999 to $70,000 and over), education, marital status (married, widowed, divorced/separated, and never married), and employment status (retired/not retired).

Mental health disorders

Major depressive disorder and anxiety disorders were examined in this study. These covariates were selected because of their higher prevalence in the US population compared to other mental health disorders (Hasin, Stinson, Ogburn, & Grant, 2007). Moreover, literature suggests that the two mental disorders are highly related to social networks and alcohol use among OA (e.g. Ivan et al., 2014). We used a binary variable examining past year history of major depressive disorder and a binary variable examining past year history of an anxiety disorder, including either the presence of social phobia, panic disorder, and generalized anxiety disorder.

Self-report physical health

Self-report physical health was measured by one item with a response ranges from 1 = excellent to 5 = poor. Thus, lower scores reflect better perceived physical health status.

Data analyses

Statistical data analyses were performed using STATA (Version 14). All analyses incorporated survey weights, clustering on primary sampling units, and stratification (Grant, Kaplan, Shepard, & Moore, 2003). We used Taylor series linearization (STATA ʻ svy̕ command) to obtain unbiased estimates of standard errors. Before testing two hypotheses, we first examined significant differences in demographic variables, mental and physical health characteristics, and social networks between two drinking groups (low-risk vs. at-risk) stratified by age group. We used F-statistic approximations to the Rao–Scott statistic for categorical variables (gender, race/ethnicity, income, education, marital status, employment status, and mental health) and adjusted Wald F-tests for continuous variables (age, social networks, and physical health). To test our first hypothesis, a weighted logistic regression model examined the association between social networks (size and diversity) and the probability of at-risk drinking among MA and OA, while adjusting for potentially confounding socio-demographic and health-related variables. To determine whether there were age-group differences in the association between social networks (size and diversity) and the probability of at-risk drinking (hypothesis 2), we added the interactions terms in the next step. We entered interaction terms for age group and each of the social networks indicators (age group × social network size, age group social × network diversity). Odds ratios (OR) and 95% confidence intervals (CIs) were reported.

Results

Drinking status and levels among middle-aged and older adults

The number of current drinkers was 5214 in the MA group and 3070 in the OA group. Of the 5214 currently drinking MA included in the analysis, 63.5% (n = 3281) were low-risk drinkers, and 36.5% (n = 1933) were at-risk drinkers. Among 3070 currently drinking OA, 72.7% (n = 2231) were low-risk drinkers, and 27.2% (n = 839) were at-risk drinkers.

Socio-demographic, mental and physical health, and social network characteristics by drinking levels

Middle-aged adults (MA)

Table 1 shows differences in socio-demographic, mental and physical health, and social networks characteristics between two drinking groups (low-risk and at-risk) among MA. At-risk drinkers were more likely than low-risk drinkers to be younger, male, and White. In terms of income, at-risk drinkers were more likely to be included in the lowest ($0–19,999) or the highest ($70,000 and over) income group than low-risk drinkers. At-risk drinkers were less likely to be married, but more likely to be divorced/separated or never married compared to low-risk drinkers. Moreover, at-risk drinkers reported lower education levels and more past year anxiety disorder than low-risk drinkers. With regard to social networks, at-risk drinkers reported smaller and less diverse social networks compared to low-risk drinkers.

Table 1.

Socio-demographics, clinical characteristics, and social networks, stratified by drinking groups among middle-aged adults.

Middle-aged adults (50–64 years, n = 5214)
Low-risk drinkers (63.52%)
At-risk drinkers (36.48%)
Weighted Weighted
n % n % F
Age in years (M, SD) 56.24   4.13 55.61   4.16 18.97**
Gender 60.61**
 Male 1416 47.96 1125 62.01
 Female 1865 52.04 808 37.99
Race/ethnicity 3.52*
 White 2203 79.37 1343 81.34
 Black 544   8.47 296   8.39
 American Indian/Alaska Native 58   2.33 33   1.94
 Asian/Pacific Islander 76   3.49 20   1.54
 Hispanic 400   6.31 241   6.78
Household income 3.69*
 $0–19,999 434   9.79 297 11.69
 $20,000–39,999 685 17.46 385 15.85
 $40,000–69,999 953 29.44 489 25.88
 $70,000 and over 1209 43.31 762 46.58
Education 3.58**
 <Some high school 289   7.51 214   8.95
 Completed high school 693 20.88 436 22.38
 <Completed college 1576 48.37 919 49.62
 >Graduate studies 723 23.24 364 19.04
Marital status 8.75**
 Married 2067 76.20 1128 70.36
 Widowed 192   3.64 77   3.17
 Divorced/Separated 718 14.27 540 20.03
 Never married 304   5.88 188   6.43
Employment status 0.57
 Retired 533 16.41 301 15.48
 Employed 2748 83.59 1632 84.52
Past year major depressive disorder 267   7.67 169   7.67 0.00
Past year anxiety disorder 199   5.45 155   7.15 5.32*
Self-rated health (M, SD) 2.37   1.07 2.35   1.11 0.23
Social networks size (M, SD) 30.23 27.08 28.30 25.54 4.31*
Social networks diversity (M, SD) 6.32   1.83 6.08   1.82 15.90**
Open in a new tab

Notes: M = mean; SD = standard deviation.

*

p < .05;

**

p < .001.

Older adults (OA)

Among OA, at-risk drinkers were more likely to be younger, male, and married and divorced/separated than were low-risk drinkers (see Table 2). At-risk drinkers tended to report higher income and education and better physical health status compared to low-risk drinkers. No differences were found in the size and diversity of social networks between two drinking groups.

Table 2.

Socio-demographics, clinical characteristics, and social networks, stratified by drinking groups among older adults.

Older adult (65 years and older, n = 3070)
Low-risk drinkers (72.7%)
At-risk drinkers (27.2%)
Weighted Weighted
n % n % F
Age in years (M, SD) 74.49   6.65 72.73   6.03 38.45**
Gender 92.81**
 Male 934 45.16 549 66.88
 Female 1297 54.58 290 33.12
Race/ethnicity 1.03
 White 1748 88.46 657 89.06
 Black 218   4.70 82   4.33
 American Indian/Alaska Native 23   1.26 12   1.80 23
 Asian/Pacific Islander 18   1.16 4   0.47
 Hispanic 224   4.40 84   4.31
Household income 9.91**
 $0–19,999 681 24.03 194 17.01
 $20,000–39,999 760 35.24 258 29.42
 $40,000–69,999 476 23.95 221 30.20
 $70,000 and over 314 16.78 166 23.38
Education 9.24**
 <Some high school 387 15.24 134 10.81
 Completed high school 664 30.95 193 24.60
 <Completed college 892 40.05 358 44.60
 >Graduate studies 288 13.75 154 20.00
Marital status 5.24*
 Married 1132 62.52 451 66.74
 Widowed 735 25.15 200 19.33
 Divorced/separated 268   8.35 150 10.89
 Never married 96   2.97 38   3.03
Employment status 0.19
 Retired 1837 81.83 665 81.02
 Employed 394 18.17 174 18.98
Past year major depressive disorder 92   4.01 30   3.11 1.24
Past year anxiety disorder 54   2.36 31   3.19 1.60
Self-rated health (M, SD) 2.59   1.04 2.47   1.05 5.95*
Social networks size (M, SD) 33.60 30.57 33.38 28.75 0.03
Social networks diversity (M, SD) 5.36   1.59 5.38   1.62 0.06
Open in a new tab

Note: M = mean; SD = standard deviation.

*

p < .05;

**

p < .001.

The association between social networks and at-risk drinking among MA and OA

Social network size

As depicted in Table 3, the model showed that social network size was not associated with at-risk drinking among MA and OA. We also found no significant age-group differences in the relationship between the size of social networks and the odds of at-risk drinking. Regression coefficients associated with the relationship between the size of social networks and the probability of at-risk drinkers for MA and OA are presented in Figure 1.

Table 3.

Logistic regressions of social network size on at-risk drinking.

Participants (n = 8194)
Model 1
Model 2
OR 95% CI OR 95% CI
Age group 0.74 0.64–0.86 0.47** 0.30–0.72
Social network size 0.99 0.99–1.00 0.99 0.99–1.00
Social network diversity 0.96 0.92–0.99 0.94** 0.89–0.98
Network size × age group – – 0.99 0.99–1.00
Network diversity ×age group – – 1.08** 1.00–1.18
Gender
 Male
 Female 0.49 0.43–0.55 0.49** 0.43–0.55
Race/ethnicity
 White
 Black 0.93 0.78–1.11 0.93 0.78–1.11
 American Indian/Alaska Native 0.89 0.56–1.42 0.89 0.56–1.41
 Asian/Pacific Islander 0.38 0.24–0.63 0.38** 0.24–0.64
 Hispanic 1.01 0.82–1.24 1.01 0.83–1.24
Household income
 $0–19,999
 $20,000–39,999 0.98 0.81–1.16 0.97 0.81–1.15
 $40,000–69,999 1.13 0.93–1.37 1.12 0.92–1.36
 $70,000 and over 1.44 1.17–1.76 1.44 1.17–1.77
Education
 <Some high school
 Completed high school 1.10 0.90–1.33 1.08 0.89–1.31
 <Completed college 1.10 0.91–1.33 1.09 0.90–1.32
 >Graduate studies 0.95 0.77–1.18 0.94 0.75–1.17
Marital status
 Married
 Widowed 1.04 0.86–1.27 1.08 0.88–1.32
 Divorced/separated 1.64** 1.40–1.91 1.64** 1.46–1.91
 Never married 1.20 0.96–1.51 1.19 1.03–1.50
Employment status
 Retired
 Employed 1.06 0.91–1.24 1.05 0.90–1.23
Past year major depressive disorder 0.93 0.72–1.20 0.93 0.72–1.19
Past year anxiety disorder 1.51* 1.18–1.92 1.50* 1.17–1.91
Self-report health (M, SE) 0.94* 0.85–0.99 0.93* 0.87–0.98
Open in a new tab

Note: OR = odds ratio; CI = confidence interval; M = mean; SD = standard deviation.

*

p < .05;

**

p < .001.

Figure 1.

Figure 1

Open in a new tab

Probability of at-risk drinking according to age group and the size of social networks.

Social networks diversity

In comparison with the social network size, social network diversity was significantly associated with lower risk of at-risk drinking among MA and OA. In addition, we found age group difference in the association between the diversity of social networks and the probability of at-risk drinking (OR = 1.08, p < 0.05). Regression coefficients associated with the relationship between the diversity of social networks and the probability of at-risk drinkers for MA and OA are presented in Figure 2. As seen in the figure, among MA, the association between diversity of social networks and the probability of at-risk drinking declines with each additional social network type with 8% lower odds of being an at-risk drinker for additional network type. Among OA, the probability of at-risk drinking is stable regardless of social network diversity.

Figure 2.

Figure 2

Open in a new tab

Probability of at-risk drinking according to age group and the diversity of social networks.

Discussion

The current study investigated the association between social networks and alcohol consumption (low-risk drinking and at-risk drinking) among MA and OA. Our findings revealed that there was no significant relationship between the size of social networks and at-risk drinking among MA and OA. But our findings did suggest that the diversity of social networks was associated with the probability of at-risk drinking among MA and OA and that the association was weaker for OA than for MA.

The association between social networks and at-risk drinking

We found mixed support for our first hypothesis. We only found that more diverse social networks were associated with a lower risk of at-risk drinking among MA and OA. Our study supports prior work that found a negative association between social networks diversity and alcohol consumption (Mowbray et al., 2014). One possible explanation for the negative association between more diverse social networks and at-risk alcohol use among MA and OA is that social networks serve a regulatory function, in that individuals with diverse social interactions are less likely to engage in risky drinking behaviors such as excessive drinking (Anson, 1989; Ewart, 1991). Social control has been considered to be central to the association between social networks and health behaviors in adulthood (Umberson et al., 2010). Control from social network members can be related to alcohol consumption by helping (e.g. reminding, urging, or threatening) drinkers to decrease alcohol use (e.g. Tucker, Klein, & Elliott, 2004). On the other hand, a greater variety of resources and information from the link from different networks together may be related to health risk behaviors such as drinking (Cohen, 2002; Eriksson, 2011; Putnam, 2000). For example, being socially connected to diverse population has the potential for increasing the odds of being exposed to new alternative behaviors to drinking.

Although we found significant relationship between diverse social networks and a lower risk of at-risk drinking among MA and OA, the association among OA was weaker than that among MA. One possible reason for this finding is a decrease in diversity of social networks among OA. According to the socioemotional selectivity theory (Carstensen, 1992), we assumed that OA would report less diverse social networks than MA and that social networks among OA would be more likely to consist of close and intimate relationships. In consistent with our assumption, we found that OA had significantly less diverse social networks than MA (p < .001). It is reasonable to think that the relationship between social networks and the probability of at-risk drinking among OA might be weaker than that of MA, as less diverse social networks are more normative in older adulthood. Reduction in social network might be related to a decrease in the role of social networks as a regulatory function on alcohol consumption (Anson, 1989; Ewart, 1991). Alternatively, overall low alcohol consumption among OA (e.g. Moore et al., 2005) may be associated with the weaker relationship between the diversity of social networks and at-risk drinking among OA than MA.

It is interesting to note that the size of social networks was not significantly associated with alcohol use among MA and OA. The results support findings from a previous study that the diversity of social networks was significant predictor of Alcohol Use Disorder (AUD), while social network size was not associated with AUD (Mowbray et al., 2014). Given the non-significant relationship between the size of social networks and alcohol use among MA and OA in this study, the diversity of social networks might be more critical to alcohol use than network size among this population. This finding suggests that low-risk drinkers among MA and OA are more likely to have diverse social networks, rather than just have a large number of social connections.

Socio-demographic correlates of alcohol consumption

We found some demographic factors related to alcohol use among MA and OA. First, age and female gender were associated with decreased odds of being an at-risk drinker. This is consistent with the literature suggesting that a younger age and male gender are risk factors for excessive alcohol use (Hasin et al., 2007; Moos et al., 2004). Regarding education level, a higher education level was related to an increased OR of at-risk drinking among OA. Our findings align with prior research suggesting OA with higher socioeconomic statuses, including education levels were more likely to be at-risk drinkers (Iparraguirre, 2015). Finally, in both age groups, we found that the presence of a past year anxiety disorder was significantly related to higher risk of at-risk drinking. Previous studies also indicate high comorbidity between anxiety disorders and alcohol abuse (Ivan et al., 2014; Wolitzky-Taylor, Castriotta, Lenze, Stanley, & Craske, 2010).

Implications

Understanding the relationship between social networks and alcohol use is important for developing interventions for unhealthy alcohol use. Literature shows the effectiveness of social network-based interventions in decreasing alcohol use among mixed-age populations (Copello et al., 2002; Litt, Kadden, Kabela-Cormier, & Petry, 2007). Our study suggests that targeting the diversity of social networks may be promising approach to reduce alcohol use among MA. For example, a recent study has found that a pilot test of a computer-assisted motivational social network intervention targeting the diversity of social contacts enhanced motivation to reduce future high-risk alcohol use (Kennedy et al., 2016). Few structural aspects of social relationships were associated with at-risk drinking among OA, suggesting a need to examine other aspects of social relationships that may influence alcohol consumption among this population. Examination of social integration (e.g. frequency of contact with network members) or the quality of relationships (e.g. conflict, support) may provide a better understanding of the association between social networks and alcohol use among OA. Moreover, alcohol consumption among OA might be explained significantly better by other factors controlled for in the analysis. Therefore, exploring the association between social networks and alcohol use with these other factors would afford a more nuanced representation of the association between social networks and alcohol use among OA. Finally, the current study suggests that the size and diversity of social networks need to be differentiated in social networks research (Erosheva et al., 2015).

Limitations

When interpreting these findings, there are some limitations to consider. First, this study is limited by its cross-sectional design, which does not allow one to address causal relations. For instance, the current study was unable to assess whether social networks alter alcohol use or vice versa. Longitudinal research may help to shed light on relationships between social networks and alcohol consumption among OA. In addition, the measurement of alcohol use did not assess other alcohol use-related information such as duration of drinking or change in drinking behavior that may be related to social networks and age. The study could be strengthened if more information related to alcohol use were included. Finally, this study did not reflect the heterogeneity of the elderly population. Given continuous changes in social networks and alcohol use in later life (Wiscott, Kopera-Frye, & Begovic, 2002), future studies need to distinguish cohorts of older adults, such as young–old and old–old groups.

Conclusions

This study revealed that the association between the diversity of social networks and alcohol consumption (at-risk drinking) varies according to age, and the structural aspect of social networks (size and diversity) plays a somewhat modest role in at-risk drinking among older adults once socio-demographic, mental, and physical health factors are considered. In the future, research should consider a more in-depth exploration of the nature of social networks and alcohol consumption over time. Sophisticated methods of exploring drinking networks, such as social network analysis may present a more nuanced picture than survey methods. Similarly, longitudinal studies should consider both alcohol use and social network characteristics as time varying and reciprocal. The current study provides a foundation for further work on social networks and alcohol consumption, an understudied area of research in older adults compared to younger age groups.

Acknowledgments

Funding

This work was supported by the National Institutes of Drug Abuse [grant number T32 DA07272-23]; National Institute of Alcohol Abuse and Alcoholism [grant number K24 AA15957]; National Institute of Aging [grant number P30 AG028748]; National Institute of Health/National Institute of Aging [grant number P30-AG021684].

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

  1. Ahlström S. Alcohol use and problems among older women and men: A review. Nordic Studies on Alcohol and Drugs. 2008;25(2):154–161. [Google Scholar]
  2. Ajrouch KJ, Antonucci TC, Janevic MR. Social networks among blacks and whites: The interaction between race and age. Journal of Gerontology. 2001;56(2):S112–118. doi: 10.1093/geronb/56.2.S112. [DOI] [PubMed] [Google Scholar]
  3. Ajrouch KJ, Blandon AY, Antonucci TC. Social networks among men and women: The effects of age and socioeconomic status. Journal of Gerontology. 2005;60(6):S311–S317. doi: 10.1093/geronb/60.6.S311. [DOI] [PubMed] [Google Scholar]
  4. Anderson P, Scafato E, Galluzzo L. Alcohol and older people from a public health perspective. Annali dell’Istituto Superiore di Sanità. 2012;48(3):232. doi: 10.4415/ANN_12_03_04. http://dx.doi.org/10.4415/ANN_12_03_04. [DOI] [PubMed] [Google Scholar]
  5. Anson O. Marital status and women’s health revisited: The importance of a proximate adult. Journal of Marriage & the Family. 1989;51:185–194. doi: 10.2307/352379. [DOI] [Google Scholar]
  6. Bacharach SB, Bamberger PA, Cohen A, Doveh E. Retirement, social support, and drinking behavior: A cohort analysis of males with a baseline history of problem drinking. Journal of Drug Issues. 2007;37(3):525–548. doi: 10.1177/002204260703700303. [DOI] [Google Scholar]
  7. Berkman LF, Glass T, Brissette I, Seeman TI. From social integration to health: Durkheim in the new millennium. Social Science & Medicine. 2000;51(6):843–857. doi: 10.1016/S0277-9536(00)00065-4. [DOI] [PubMed] [Google Scholar]
  8. Blow FC, Barry KL. Use and misuse of alcohol among older women. Alcohol Research & Health. 2002;26:308–315. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12875042. [PMC free article] [PubMed] [Google Scholar]
  9. Bullers S, Cooper ML, Russell M. Social network drinking and adult alcohol involvement: A longitudinal exploration of the direction of influence. Addictive Behaviors. 2001;26(2):181–199. doi: 10.1016/S0306-4603(00)00099-X. [DOI] [PubMed] [Google Scholar]
  10. Carstensen LL. Social and emotional patterns in adulthood: Support for socioemotional selectivity theory. Psychology and Aging. 1992;7:331–338. doi: 10.1037/0882-7974.7.3.331. [DOI] [PubMed] [Google Scholar]
  11. Chen C, Yi H, Dawson D, Stinson F, Grant B. Alcohol use and alcohol use disorders in the United States, A 3-year follow-up: Main findings from the 2004–2005 Wave 2 NESARC. Bethesda, MD: NIH; 2010. 10-7677. [Google Scholar]
  12. Cohen S. Psychosocial stress, social networks and susceptibility to infection. In: Koenong HG, Cohen HJ, editors. The link between religion and health: Psychoneuroimmunology and the faith factor. New York, NY: Oxford University Press; 2002. pp. 101–123. [Google Scholar]
  13. Cohen S, Doyle WJ, Skoner DP, Rabin BS, Gwaltney JM., Jr Social ties and susceptibility to the common cold. JAMA. 1997;277(24):1940–1944. doi: 10.1016/s0020-7292(98)90474-x. [DOI] [PubMed] [Google Scholar]
  14. Committee report. The social side of human aging. In: Carstensen LL, Hartel CR, editors. When I’m 64. Washington, DC: The National Academies Press; 2006. pp. 19–33. (Committee on Aging Frontiers in Social Psychology, Personality, and Adult Developmental Psychology). [Google Scholar]
  15. Copello A, Orford J, Hodgson R, Tober G, Barrett C, UKATT Research Team Social behaviour and network therapy: Basic principles and early experiences. Addictive Behaviors. 2002;27(3):345–366. doi: 10.1016/s0306-4603(01)00176-9. http://dx.doi.org/10.1016/S0306-4603(01)00176-9. [DOI] [PubMed] [Google Scholar]
  16. Cornwell B. Network bridging potential in later life life-course experiences and social network position. Journal of Aging and Health. 2009;21(1):129–154. doi: 10.1177/0898264308328649. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Cousins G, Galvin R, Flood M, Kennedy MC, Motterlini N, Henman M. Fahey, T. Potential for alcohol and drug interactions in older adults: Evidence from the Irish longitudinal study on ageing. BMC Geriatrics. 2014;14(1):57. doi: 10.1186/1471-2318-14-57. doi:org/10.1186/1471-2318-14-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Dawson DA, Goldstein RB, Grant BF. Rates and correlates of relapse among individuals in remission from DSM—IV alcohol dependence: A 3—year follow up. Alcoholism: Clinical and Experimental Research. 2007;31(12):2036–2045. doi: 10.1111/j.1530-0277.2007.00536.x. [DOI] [PubMed] [Google Scholar]
  19. Erikson EH. Insight and responsibility. New York, NY: Norton; 1964. [Google Scholar]
  20. Eriksson M. Social capital and health-implications for health promotion. Global Health Action. 2011;4:1–11. doi: 10.3402/gha.v4i0.5611. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Erosheva EA, Kim HJ, Emlet C, Fredriksen-Goldsen KI. Social networks of lesbian, gay, bisexual, and transgender older adults. Research on Aging. 2015;38(1):98–123. doi: 10.1177/0164027515581859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Ewart CK. Social action theory for a public health psychology. American Psychologist. 1991;46(9):931–946. doi: 10.1037//0003-066x.46.9.931. doi:org/10.1037/0003-066X.46.9.931. [DOI] [PubMed] [Google Scholar]
  23. Grant BF, Dawson DA, Stinson FS, Chou SP, Dufour MC, Pickering RP. The 12-month prevalence and trends in DSM-IV alcohol abuse and dependence: United States, 1991–1992 and 2001–2002. Drug and Alcohol Dependence. 2004;74(3):223–234. doi: 10.1016/j.drugalcdep.2004.02.004. [DOI] [PubMed] [Google Scholar]
  24. Grant BF, Moore TC, Shepard J, Kaplan K. Source and accuracy statement: Wave 1 national epidemiologic survey on alcohol and related conditions (NESARC) Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism; 2003. [Google Scholar]
  25. Hasin DS, Stinson FS, Ogburn E, Grant BF. Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Arch Gen Psychiatry. 2007;64(7):830–842. doi: 10.1001/archpsyc.64.7.830. [DOI] [PubMed] [Google Scholar]
  26. Iparraguirre J. Socioeconomic determinants of risk of harmful alcohol drinking among people aged 50 or over in England. BMJ Open. 2015;5(7):1–14. doi: 10.1136/bmjopen-2015-007684. doi:org/10.1136/bmjopen-2015-007684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ivan MC, Amspoker AB, Nadorff MR, Kunik ME, Cully JA, Wilson N, Stanley MA. Alcohol use, anxiety, and insomnia in older adults with generalized anxiety disorder. The American Journal of Geriatric Psychiatry. 2014;22(9):875–883. doi: 10.1016/j.jagp.2013.04.001. doi:org/10.1016/j.jagp.2013.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kennedy DP, Hunter SB, Osilla KC, Maksabedian E, Golinelli D, Tucker JS. A computer-assisted motivational social network intervention to reduce alcohol, drug and HIV risk behaviors among Housing First residents. Addiction Science & Clinical Practice. 2016;11(1):1. doi: 10.1186/s13722-016-0052-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kirchner JE, Zubritsky C, Cody M, Coakley E, Chen H, Ware JH, Levkoff S. Alcohol consumption among older adults in primary care. Journal of General Internal Medicine. 2007;22(1):92–97. doi: 10.1007/s11606-006-0017-z. http://doi.org/10.1007/s11606-006-0017-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lang I, Wallace RB, Huppert FA, Melzer D. Moderate alcohol consumption in older adults is associated with better cognition and well-being than abstinence. Age and Ageing. 2007;36(3):256–261. doi: 10.1093/ageing/afm001. [DOI] [PubMed] [Google Scholar]
  31. Litt MD, Kadden RM, Kabela-Cormier E, Petry N. Changing network support for drinking: Initial findings from the Network Support Project. Journal of Consulting and Clinical Psychology. 2007;75(4):542. doi: 10.1037/0022-006X.75.4.542. http://dx.doi.org/10.1037/0022-006X.75.4.542. [DOI] [PubMed] [Google Scholar]
  32. Litwin H, Stoeckel K. Social networks and subjective wellbeing among older Europeans: Does age make a difference? Ageing and Society. 2013;33:1263–1281. doi: 10.1017/S0144686&#x000d7;12000645. [DOI] [Google Scholar]
  33. Moore AA, Gould R, Reuben DB, Greendale GA, Carter MK, Zhou K, Karlamangla A. Longitudinal patterns and predictors of alcohol consumption in the United States. American Journal of Public Health. 2005;95(3):458–464. doi: 10.2105/AJPH.2003.019471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Moos RH, Brennan PL, Schutte KK, Moos BS. Social and financial resources and high-risk alcohol consumption among older adults. Alcoholism: Clinical and Experimental Research. 2010;34:646–654. doi: 10.1111/j.1530-0277.200901133.x. [DOI] [PubMed] [Google Scholar]
  35. Moos RH, Schutte KK, Brennan P, Moos BS. Ten-year patterns of alcohol consumption and drinking problems among older women and men. Addiction. 2004;99(7):829–838. doi: 10.1111/j.1360-0443.2004.00760x. [DOI] [PubMed] [Google Scholar]
  36. Moriconi PA, Nadeau L. A cross-sectional study of self-rated health among older adults: Association with drinking profiles and other determinants of health. Current Gerontology and Geriatrics Research. 2015;3(5):29–47. doi: 10.1155/2015/352947. http://doi.org/10.1155/2015/352947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mowbray O, Quinn A, Cranford JA. Social networks and alcohol use disorders: Findings from a nationally representative sample. The American Journal of Drug and Alcohol Abuse. 2014;40(3):181–186. doi: 10.3109/00952990.2013.860984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Naimi TS, Brown DW, Brewer RD, Giles WH, Mensah G, Serdula MK, Stroup DF. Cardiovascular risk factors and confounders among nondrinking and moderate-drinking US adults. American Journal of Preventive Medicine. 2005;28(4):369–373. doi: 10.1016/j.amepre.2005.01.011. http://dx.doi.org/10.1016/j.amepre.2005.01.011. [DOI] [PubMed] [Google Scholar]
  39. National Institute of Alcohol Abuse and Alcoholism. Wave 2 NESARC data notes. Bethesda, MD: National Institutes of Health; 2008. [Google Scholar]
  40. National Institute of Alcohol Abuse and Alcoholism. Rethinking drinking: Alcohol and your health. Bethesda, MD: U.S. Department of Health and Human Services; 2010. [Google Scholar]
  41. Nesselroade JR. Intraindividual variability in development within and between individuals. European Psychologist. 2001;6(3):187–193. doi: org/10.1027//1016-9040.6.3.187. [Google Scholar]
  42. Novier A, Diaz-Granados JL, Matthews DB. Alcohol use across the lifespan: An analysis of adolescent and aged rodents and humans. Pharmacology Biochemistry and Behavior. 2015;133:65–82. doi: 10.1016/j.pbb.2015.03.015. doi:org/10.1016/j.pbb.2015.03.015. [DOI] [PubMed] [Google Scholar]
  43. Perreira KM, Sloan FA. Life events and alcohol consumption among mature adults: A longitudinal analysis. Journal of Studies on Alcohol. 2001;62:501–508. doi: 10.15288/jsa.2001.62.501. doi:org/10.15288/jsa.2001.62.501. [DOI] [PubMed] [Google Scholar]
  44. Putnam RD. Bowling alone: America’s declining social capital. In: Crothers L, Lockhart C, editors. Culture and politics. New York, NY: St. Martin’s Press; 2000. pp. 223–234. [Google Scholar]
  45. Satre DD. Alcohol and drug use problems among older adults. Clinical psychology: Science and practice. 2015;22(3):238–254. doi: 10.1111/cpsp.12104. [DOI] [Google Scholar]
  46. Seaman P, Ikegwuonu T. Drinking to belong. Understanding young adults alcohol use within social networks. York: Joseph Rowntree Foundation; 2010. [Google Scholar]
  47. Shiovitz-Ezra S, Litwin H. Social network type and health-related behaviors: Evidence from an American national survey. Social Science & Medicine. 2012;75(5):901–904. doi: 10.1016/j.socscimed.2012.04.031. doi:org/10.1016/j.socscimed.2012.04.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sklar AL, Gilbertson R, Boissoneault J, Prather R, Nixon SJ. Differential effects of moderate alcohol consumption on performance among older and younger adults. Alcoholism: Clinical and Experimental Research. 2012;36(12):2150–2156. doi: 10.1111/j.1530-0277.2012.01833.x. doi:org/10.1111/j.1530-0277.2012.01833x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tucker JS, Klein DD, Elliott MN. Social control of health behaviors: A comparison of young, middle-aged, and older adults. Journal of Gerontology. 2004;59(4):147–150. doi: 10.1093/geronb/59.4. [DOI] [PubMed] [Google Scholar]
  50. Uchino BN. Understanding the links between social support and physical health: A life-span perspective with emphasis on the separability of perceived and received support. Perspectives on Psychological Science. 2009;4(3):236–255. doi: 10.1111/j.1745-6924.2009.01122.x. doi:org/10.1111/j.1745-6924.2009.01122.x. [DOI] [PubMed] [Google Scholar]
  51. Umberson D, Crosnoe R, Reczek C. Social relationships and health behavior across life course. Annual Review of Sociology. 2010;36:139–157. doi: 10.1146/annurev-soc-070308-120011. http://doi.org/10.1146/annurev-soc-070308-120011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Census Bureau US. 2000 census of population and housing, population and housing unit counts PHC-3-1 United States summary. Washington, DC: Author; 2004. [Google Scholar]
  53. Valente TW. Network interventions. Science. 2012;337(6090):49–53. doi: 10.1126/science.1217330. [DOI] [PubMed] [Google Scholar]
  54. Valente TW, Olson Ritt A, Stacy A, Unger BJ, Okamoto J, Sussman S. Peer acceleration: Effects of a social network tailored substance abuse prevention program among high risk adolescents. Addiction. 2007;102(11):1804–1815. doi: 10.1111/j.1360-0443.2007.01992x. [DOI] [PubMed] [Google Scholar]
  55. Watt RG, Heilmann A, Sabbah W, Newton T, Chandola T, Aida J, Tsakos G. Social relationships and health related behaviors among older US adults. BMC Public Health. 2014;14:533. doi: 10.1186/1471-2458-14-533. http://doi.org/10.1186/1471-2458-14-533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wiscott R, Kopera-Frye K, Begovic A. Binge drinking in later life: Comparing young-old and old-old social drinkers. Psychology of Addictive Behaviors. 2002;16(3):252–255. doi:org/10.1037/0893-164X.16.3.252. [PubMed] [Google Scholar]
  57. Wolitzky-Taylor KB, Castriotta N, Lenze EJ, Stanley MA, Craske MG. Anxiety disorders in older adults: A comprehensive review. Depress and Anxiety. 2010;27:190–211. doi: 10.1002/da.20653. [DOI] [PubMed] [Google Scholar]
  58. World Health Organization. Definition of an older or elderly person. Geneva: Author; 2003. (WHO Technical Report series 916). [Google Scholar]
  59. Wrzus C, Hanel M, Wagner J, Neyer FJ. Social network changes and life events across the life span: A meta-analysis. Psychological Bulletin. 2013;139:53–80. doi: 10.1037/a0028601. doi:org/10.1037/a0028601. [DOI] [PubMed] [Google Scholar]
  60. Wu LT, Blazer DG. Substance use disorders and psychiatric comorbidity in mid and later life: A review. International Journal of Epidemiology. 2014;43(2):304–317. doi: 10.1093/ije/dyt17. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES