Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Sep 1.
Published in final edited form as: Soc Sci Med. 2012 May 24;75(5):901–904. doi: 10.1016/j.socscimed.2012.04.031

Social network type and health-related behaviors: Evidence from an American national survey

Sharon Shiovitz-Ezra 1, Howard Litwin 2
PMCID: PMC3552155  NIHMSID: NIHMS380464  PMID: 22682660

Abstract

This study examined the association between social network type and engagement in physical activity, alcohol abuse and use of complementary and alternative medicine by older Americans. Data from the National Social Life, Health & Aging Project were employed. Multivariate logistic regressions conducted separately for each health behavior showed that older people embedded in less resourceful network types were at greater risk for alcohol abuse, physical inactivity and less use of complementary and alternative medicine, net of the effects of sociodemographic characteristics, health, and the quality of the social relationships. The study underscores the importance of the construct of social network type for understanding healthy lifestyle in late life.

Keywords: Network type, loneliness, alcohol abuse, physical activity, alternative medicine, complementary and alternative medicine, U.S.A

INTRODUCTION

A popular conceptual model put forth by Berkman and colleagues (2000) emphasizes the influence of social networks with respect to health outcomes. Networks are important because they provide social resources that enhance or restrain access to opportunities, which in turn, determine behaviors and attitudes. The model posits that social networks operating at the mezzo level impact health through psychosocial mechanisms operating at the micro level.

Health-related behavior, defined as "a range of personal actions that influence health, disability, and mortality,” (Umberson et al., 2010, p. 140), is one of the key mechanisms in the three downstream pathways to health in Berkman and colleagues’ model. According to Umberson (1987), health behavior is the mechanism that underlies the social relationships/ health and mortality association. Certain health behaviors, particularly exercise, eating well and adherence to medical regimens promote health and prevent illness. In contrast, smoking, excessive weight gain and substance abuse can compromise health (Umberson et al., 2010).

Empirical evidence supports the linkage between social ties and the adoption of health related behaviors. For example, people who reported having many social connections were more likely to have had colorectal cancer screening within the previous five years (Kinney et al., 2005). Being a member of a religious or other group was also predictive of taking the cancer screening. In a study of Mexican-American adults, aged 20–75, social relations were associated with motivations to engage in screening for blood cholesterol and blood pressure (Ashida et al., 2010). Correspondingly, heavy drinking was found to be associated with less self-rated social support (Kirchner et al., 2007). But, belonging to community groups and attending religious services was associated with less likelihood of engaging in physical activities among Latino women (Voorhees & Young, 2003).

Complementary and alternative medicine (CAM) is "a group of diverse medical and health care systems, therapies, and products that are not presently considered to be part of conventional medicine" that are used to treat or prevent illness and promote health (Barnes et al., 2004, p.54). Support and strain from different network members have been found to correlate with different types of CAM use. For example, perceived friend support was positively associated with the use of mind –body therapies (such as biofeedback), whereas perceived partner strain was associated with increasing odds of using biologically based therapy (such as herbal medicine) (Honda & Jacobson, 2005).

Studies of the linkages between social relationships and health behavior tend to look at specific social ties (e.g. Jeffery & Rick, 2002), at simple indices of social network indicators (e.g. Yun et al., 2010) or at isolated indicators of the social network (e.g. Voorhees & Young, 2003). The absence of a more complex picture of social ties in relation to health-related behavior reflects a gap in the research literature. This is particularly relevant for testing the social network/ health-behavior relationship in adulthood, a period in which one’s social world may undergo changes (Umberson et al., 2010).

The construct of social network type provides a way to take into account the complexity of the interpersonal environment in late life by incorporating a composite collection of network characteristics (Wenger, 1991). Fiori and colleagues (2006) maintain that composite measures that reflect several aspects of one’s social network provide added value for understanding the social milieu in which older individuals are embedded. Recent analyses have sought to identify social network typologies in different societies. Four core typologies found most often have been termed: "diverse," "family-focused," "friend-focused" and "restricted," with some cross-cultural variations also evident (Cheng et al., 2009; Cheon, 2010; Doubova et al., 2010; Fiori et al., 2008; Fiori et al., 2007; Litwin & Shiovitz-Ezra, 2006, 2011b).

Network types have been shown to predict mental and physical health outcomes. Better health is observed among older people embedded in network types characterized by greater social capital, regardless of cultural setting (e.g. Doubova et al., 2010; Litwin, 2001; Litwin & Shiovitz-Ezra, 2011a). However, there is still only limited research exploring the associations between social network types and health-related behaviors. One study showed that respondents in the most socially endowed “diverse network” had the highest likelihood for engaging in physical activity while those in the less endowed exclusively “family” and/or “restricted” networks had the lowest (Litwin, 2003).

The present study aims to fill this gap by testing associations between social network type and three health-related behaviors: risky health behavior—abuse of alcohol, health-promoting behavior—engaging in physical activity, and health-related help-seeking. We hypothesize that respondents embedded in socially resourceful network types engage more frequently in health-enhancing behavior whereas respondents embedded in network types characterized by lesser social capital engage more frequently in risky behavior.

An additional aspect of the interpersonal environment that requires attention when exploring associations between social relationships and health behaviors is the quality of the social relationships. The literature shows that social relationships may be characterized as supportive and beneficial, but also as a source for stress, strain and conflict (Cohen, 2004; Due et al., 1999). Loneliness as a discrete subjective concept reflects the perceived quality of one's social relationships. It occurs when a desired degree of intimacy does not accompany one's social interactions (de Jong Gierveld, 1998). Embeddedness in a socially resourceful network, in terms of quantity and variety of ties, is not necessarily an antidote for loneliness.

There is some inconsistency in findings regarding the loneliness/ health behaviors relationships. Cacioppo and colleagues (2002) found no association between loneliness and obesity, smoking, alcohol use and physical activity among relatively small samples of undergraduate students and older adults. However, in a random Australian sample, lonely and non-lonely adults differed in BMI index and smoking, with lonely individuals more likely to engage in health-compromising behaviors (Lauder et al., 2006). Data on participants in the Chicago Health, Aging, and Social Relations Study revealed that loneliness was associated with transitioning from physically active to sedentary status (Hawkley et al., 2009). Thus, we further hypothesize that loneliness is positively associated with undermining health behaviors and negatively associated with health promoting behaviors.

METHODS

Data from the first wave of the National Social Life, Health and Aging Project (2005–6) (NSHAP) were used for the present secondary data analysis. NSHAP is a nationally representative multi-stage stratified area probability sample of community-dwelling older Americans aged 57 to 85, with oversampling of African-Americans, Latinos, men, and the older old (O’Muircheartaigh et al., 2009). In-person interviews were conducted in the respondents' homes using a computer-assisted personal interview and a self-administrated post-interview questionnaire. The survey spanned 3,005 respondents with a weighted sample response rate of 75.5% and 84% for the main questionnaire and the post-interview, respectively (Smith et al., 2009). Approval was granted for NSHAP from the Social and Behavioral Sciences Institutional Review Board at the University of Chicago (the Federal Wide Assurance [FWA] No. is FWA00005565) and the Institutional Review Board at the National Opinion Research Center (NORC) (FWA00000142). In the current inquiry, we address only respondents aged 65 and over.

Study Variables

Health-related behaviors

Three outcome variables representing three domains of health-related habits were employed:

  1. Negative (risky) health behavior—abuse of alcohol—was measured by CAGE, a clinical screening instrument used effectively with older adults (Beullens & Aertgeerts, 2004). The instrument has a yes/no response scale on four-questions, for example, whether respondents had ever felt that they should cut down on drinking. A dichotomous alcohol abuse variable was generated based upon the National Institute on Alcohol Abuse and Alcoholism's recommendation of a clinical screening cut-point of one positive response (Bradley et al., 2001; NIAAA, 1995). The CAGE measure was tapped in the post-interview questionnaire.

  2. Positive health-promoting engagement in physical activity was assessed with a single item, “How often do you participate in physical activity, such as walking, dancing, gardening, physical exercise, or sports?”, on a 5-point Likert scale ranging from “3 or more times per week” to “never.” The five response categories were collapsed here into two (0=never; 1=engagement to any degree).

  3. Health-related help-seeking regarding CAM was assessed by questions about utilization of up to seven alternative medicines during the past year, e.g. acupuncture and massage therapies. A dichotomized measure was generated in which a score of zero reflected "none" and a score of one represented use of one or more CAM. Engagement in physical activity and use of CAM were both queried in the in-person questionnaire.

Social network types

As described in detail in other works (Litwin & Shiovitz-Ezra, 2011a; 2011b) the construct of “network type” was developed through the application of K-means cluster analysis to seven social capital indicators which are relevant to older adults (e.g., current marital status and number of friends). Five network types were derived: "Diverse", "Friends", "Congregant", "Family" and "Restricted." The first three types were more endowed in terms of social capital, and the last two were less so.

Loneliness was measured on the sixth item of the CES-D Depression scale. Participants were asked to indicate how often they felt lonely during the past week. The 4-point answer scale was collapsed into a dichotomous response: 0 (never or rarely felt lonely), and 1 (felt lonely sometimes or more often).

Covariates included age (65–74/75–85), gender (men/women), education (<high school/high school/some college/BA or more), race/ethnicity (whites/blacks/other); subjective household income—respondents were asked to compare their household income with those of American families (below average income/average income/above average income); functional impairment (no ADL difficulty/ one or more ADL difficulties) and subjective health status (fair or less/ good/very good or excellent.

Statistical Analysis

Separate analysis was carried out for each health behavior outcome. A recent study by Cohen-Mansfield & Kivity underscores that different health-related behaviors constitute separate and mostly unrelated factors among older adults (Cohen-Mansfield & Kivity, 2011). The three outcomes in the current inquiry were regressed separately on the two social predictors (network type and loneliness) and the possible confounders by means of adjusted logistic regression. An additional analysis explored the entry of interaction terms for social network types by loneliness. In all the analyses, estimates were weighted, using Stata 10.

RESULTS

The sample had slightly more women (53.4%), about 60 per cent young-old (ages 65–74) and more than 80 percent whites. The majority category reported average income (42.8%) and at least a high school education (77.7%). More than 70 percent reported no ADL difficulties and a similar percentage rated their health as good or very good. Almost one sixth of the sample engaged in alcohol abuse, close to two thirds had been physically active in some activity and almost four out of ten used CAM. The most prevalent network type was the friend network (28.9%); the least prevalent was the family network (14%). Most respondents reported no feelings of loneliness, but 30% indicated loneliness to some degree. The bivariate analysis (not shown) revealed that each of the main predictors and covariates were statistically associated with at least one of the health-related outcomes.

Multivariate analyses carried out separately for each of the outcomes (Table 1) were adjusted for age, gender, income, race/ethnicity, self-rated health and functional impairments. The "friends" network served as the reference category as it was the most prevalent network type (28.9%) and the only one with health promoting behaviors on all three outcome measures. Compared to the other network types (not shown), respondents in the friend network had less likelihood of alcohol abuse (OR=.64, p<.01); greater likelihood of having engaged in physical activity (OR=1.57, p<.01); and a greater likelihood of having used CAM (OR=1.27, p<.05).

Table 1.

Associations between social network typology and health related behaviors: Results from Multivariate logistic regressions

Alcohol Abuse1
OR (SE) [95% CI]		 Physical Activity1
OR (SE) [95% CI]		 Alternative Meds1
OR (SE) [95% CI]
Network types
Friends2
Diverse 1.40 (.38) [.82–2.39] .91 (.19) [.61–1.39] 1.43 (.28)+ [.98–2.10]
Congregant 1.06 (.29) [.61–1.85] .80 (.17) [.52–1.23] .70 (.15)+ [.46–1.06]
Family 1.71 (.44)* [1.02–2.86] .64 (.14)* [.42–.98] .55 (.12)** [.36–.85]
Restricted 1.76 (.44)* [1.07–2.87] .77 (.15) [.52–1.13] .78 (.15) [.54–1.13]
Lonely to some
degree
No2
Yes 1.27 (.24) [.88–1.83] .80 (.11) [.60–1.07] 1.21 (.18) [.91–1.61]
Open in a new tab
+

.05> p <.10,

*

p <.05,

**

p <.01,

***

p <.001

Note: Estimates are weighted to account for differential probabilities of selection, differential non-response and to account for survey sampling design through incorporation of sampling strata and clusters.

1

Regressions are adjusted for age, gender, education, income, ethnicity, SRH and ADL difficulties

2

Reference categories

Net of the effect of the socioeconomic and health characteristics, respondents embedded in the family and restricted network groupings were more likely to have abused alcohol compared to the reference category. Members of the family cluster were also less likely to have been physically active and to have used CAM. A borderline association also suggested that respondents in the diverse network were more likely to have used CAM when compared to their counterparts in the friends network, whereas those in the congregant network cluster were less likely.

After controlling for background and health characteristics, and for the composite network type construct, the loneliness variable did not retain its inverse bivariate association with engagement in physical activity. The additional analysis exploring the entry of interaction terms for social network types by loneliness (not shown) revealed that none of the interaction terms reached significance levels in relation to the three health behavior outcomes.

DISCUSSION

Consistent with the first two research hypotheses, our findings revealed that older people embedded in less resourceful network types were at greater risk for alcohol abuse, physical inactivity and less use of CAM, after controlling for sociodemographic characteristics, health, and the quality of the social relationships. Umberson and colleagues (1987; 1992; 2010) explain the observed associations between social network type and health behaviors in terms of health-related social control. That is, people embedded in social resource-rich networks, as characterized by multiple social interactions, are exposed to the control of a variety of social agents (such as intimate partners, friends and family members). The latter might pose positive pressure for the adoption and maintenance of health-promoting behaviors or might impose informal sanctions that encourage abandonment or avoidance of health-damaging habits.

Another possible explanation for the results of the present inquiry is that socially connected people are exposed to a greater variety of sources of information relevant to their health (Cohen, 2002). For example, being socially connected has the potential for increasing the odds of being exposed to new alternative treatment types that have been found to be efficient in certain medical conditions. This is especially relevant to the current findings in regard to the utilization of CAM. Whereas the family network—the social network type with limited ties and greater dependence on children—had lower odds of using CAM treatments, the diverse network—a grouping characterized by its greater variety of social connections—showed greater odds.

Members of the congregant grouping were also less likely to use CAM. This may also be explained by the fact that their social world is less varied. Members of this particular social cluster showed the lowest rate of attendance at organized group meetings other than religious services. Consequently, the place of worship constituted a major source for social contacts for them, with their social exchange occurring primarily with other congregants (Litwin & Shiovitz-Ezra, 2011a).

The importance of the objective social constellation in which one is embedded for the adoption of healthy behaviors was strengthened in the current inquiry when the quality of social interactions was explored simultaneously. Contrary to our third hypothesis, the measure of loneliness was not associated with any of the health behavior outcomes in the adjusted models, whereas different social network types were associated with engagement in health/ risk behaviors. The absence of a loneliness/ health behaviors association was also found in the work by Cacioppo et al (2002) among young and old adults. A significant association between loneliness, on the one hand, and smoking and overweight, on the other, was indeed found in a study by Lauder and his colleagues (2006). However, those associations may be partially explained by the fact that objective social network measures were not included in the multivariate equations, not to mention the more complicated measures such as the composite construct of social network type.

A limitation of the present analysis should be pointed out. The current study was restricted to cross-sectional data. Therefore we were unable to establish causal relationships between the social network types and engagement in health related behaviors. Health related behaviors may also have the potential to affect the formation of network types and to act as a source of change in the composition of one’s interpersonal environment. Therefore, examination of a causal model will become a priority when the second wave of NSHAP data becomes available for analysis. Panel data will also allow the evaluation of transitions in one's social network in relation to health-related behaviors. For example, it was recently found that marital transitions and especially marital dissolution due to widowhood were more important than marital status in predicting weight changes over time (Umberson et al., 2009). The composition of the entire social network is presumably not static, especially in old age when losses in one's network become more prevalent as well as deterioration of health. Thus, future longitudinal inquiry should address the effect of these transitions on the formulation of different social networks which may, in turn, effect the adoption of health promoting/ compromising behaviors.

Research Highlights.

  • We employ the notion of social network type, which takes into account the complexity of the interpersonal environment.

  • People in less resourceful network types were found to be at greater risk for alcohol abuse and physical inactivity.

  • People in less resourceful network types found to make less use of complementary and alternative medicine.

  • Social tie quality— as measured by loneliness—was not associated with the three health behaviors in the adjusted models.

Acknowledgements

NSHAP is supported by the National Institute on Aging, Office of Women’s Health Research, Office of AIDS Research, and Office of Behavioral and Social Science Research (5R01AG021487).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Sharon Shiovitz-Ezra, Email: sharonshi@mscc.huji.ac.il.

Howard Litwin, Email: mshowie@mscc.huji.ac.il.

REFERENCES

  1. Ashida S, Wilkinson AV, Koehly LM. Motivation for health screening: Evaluation of social influence among Mexican-American adults. American Journal of Preventive Medicine. 2010;38(4):396–402. doi: 10.1016/j.amepre.2009.12.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barnes PM, Powell-Griner E, McFann K, Nahin RL. Complementary and alternative medicine use among adults: United States, 2002. Seminars in Integrative Medicine. 2004;2(2):54–71. [PubMed] [Google Scholar]
  3. Berkman LF, Glass T, Brissette I, Seeman TE. From social integration to health: Durkheim in the new millennium. Social Science & Medicine. 2000;51:843–857. doi: 10.1016/s0277-9536(00)00065-4. [DOI] [PubMed] [Google Scholar]
  4. Beullens J, Aertgeerts B. Screening for alcohol abuse and dependence in older people using DSM criteria: A review. Aging & Mental Health. 2004;28:76–82. doi: 10.1080/13607860310001613365. [DOI] [PubMed] [Google Scholar]
  5. Bradley KA, Kivlahan DR, Bush KR, McDonell MB, Fihn SD. Variations on the CAGE alcohol screening questionnaire: Strengths and limitations in VA general medical patients. Alcoholism: Clinical and Experimental Research. 2001;25:1472–1478. doi: 10.1097/00000374-200110000-00010. [DOI] [PubMed] [Google Scholar]
  6. Cacioppo JT, Hawkley LC, Crawford LE, Ernst JM, Burleson MH, Kowalewski RB, et al. Loneliness and health: Potential mechanisms. Psychosomatic Medicine. 2002;64:407–417. doi: 10.1097/00006842-200205000-00005. [DOI] [PubMed] [Google Scholar]
  7. Cheng ST, Lee CKL, Chan ACM, Leung EMF, Lee JJ. Social network types and subjective well-being in Chinese older adults. Journals of Gerontology: Social Sciences. 2009;64:S713–S722. doi: 10.1093/geronb/gbp075. [DOI] [PubMed] [Google Scholar]
  8. Cheon EY. Correlation of social network types on health status of Korean elders. Journal of Korean Academy of Nursing. 2010;40(1):88–98. doi: 10.4040/jkan.2010.40.1.88. [DOI] [PubMed] [Google Scholar]
  9. 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: Oxford University Press; 2002. pp. 101–123. [Google Scholar]
  10. Cohen S. Social relationships and health. American Psychologist. 2004:676–684. doi: 10.1037/0003-066X.59.8.676. [DOI] [PubMed] [Google Scholar]
  11. Cohen-Mansfield J, Kivity Y. The relationships among health behaviors in older persons. Journal of Aging and Health. 2011;23(5):822–842. doi: 10.1177/0898264311398130. [DOI] [PubMed] [Google Scholar]
  12. de Jong Gierveld J. A review of loneliness: Concept and definitions, determinants and consequences. Review in Clinical Gerontology. 1998;8:73–80. [Google Scholar]
  13. Doubova SV, Perez-Cuevas R, Espinosa-Alarcon P, Flores-Hernandez S. Social network types and functional dependency in older adults in Mexico. BMC Public Health. 2010;10:104. doi: 10.1186/1471-2458-10-104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Due P, Holstein B, Lund R, Modvig J, Avlund K. Social relations: network, support and relational strain. Social Science & Medicine. 1999;48:661–673. doi: 10.1016/s0277-9536(98)00381-5. [DOI] [PubMed] [Google Scholar]
  15. Fiori KL, Antonucci TC, Akiyama H. Profiles of social relations among older adults: A cross-cultural approach. Ageing & Society. 2008;28:203–231. [Google Scholar]
  16. Fiori KL, Antonucci TC, Cortina KS. Social network typologies and mental health among older adults. Journals of Gerontology: Psychological Sciences. 2006;61(1):P25–P32. doi: 10.1093/geronb/61.1.p25. [DOI] [PubMed] [Google Scholar]
  17. Fiori KL, Smith J, Antonucci TC. Social network types among older adults: A multidimensional approach. Journals of Gerontology: Psychological Sciences. 2007;62(6):P322–P330. doi: 10.1093/geronb/62.6.p322. [DOI] [PubMed] [Google Scholar]
  18. Hawkley LC, Thisted RA, Cacioppo JT. Loneliness predicts reduced physical activity: Cross sectional & longitudinal analyses. Health Psychology. 2009;28(3):354–363. doi: 10.1037/a0014400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Honda K, Jacobson JS. Use of complementary and alternative medicine among United States adults: The influences of personality, coping strategies, and social support. Preventive Medicine. 2005;40:46–53. doi: 10.1016/j.ypmed.2004.05.001. [DOI] [PubMed] [Google Scholar]
  20. Jeffery RW, Rick AM. Cross-sectional and longitudinal associations between body mass index and marriage-related factors. Obesity Research. 2002;10(8):809–815. doi: 10.1038/oby.2002.109. [DOI] [PubMed] [Google Scholar]
  21. Kinney AY, Bloor LE, Martin C, Sandler RS. Social ties and colorectal cancer screening among blacks and whites in North Carolina. Cancer Epidemiology, Biomarkers & Prevention. 2005;14(1):192–189. [PubMed] [Google Scholar]
  22. Kirchner JE, et al. Alcohol consumption among older adults in primary care. Society of General Internal Medicine. 2007;22:92–97. doi: 10.1007/s11606-006-0017-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lauder W, Mummery K, Jones M, Caperchione C. A comparison of health behaviors in lonely and non-lonely populations. Psychology, Health & Medicine. 2006;11(2):233–245. doi: 10.1080/13548500500266607. [DOI] [PubMed] [Google Scholar]
  24. Litwin H. Social predictors of physical activity in later life: The contribution of social-network type. Journal of Aging and Physical Activity. 2003;11(3):389–406. [Google Scholar]
  25. Litwin H. Social network type and morale in old age. Gerontologist. 2001;41(4):516–524. doi: 10.1093/geront/41.4.516. [DOI] [PubMed] [Google Scholar]
  26. Litwin H, Shiovitz-Ezra S. Network type and mortality risk in later life. Gerontologist. 2006;46(6):735–743. doi: 10.1093/geront/46.6.735. [DOI] [PubMed] [Google Scholar]
  27. Litwin H, Shiovitz-Ezra S. Social network type and subjective well-being in a national sample of older Americans. The Gerontologist. 2011a;51(3):379–388. doi: 10.1093/geront/gnq094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Litwin H, Shiovitz-Ezra S. The association of background and network type among older Americans: Is "who you are" related to "who you are with"? Research on Aging. 2011b;33(6):735–759. [Google Scholar]
  29. National Institute on Alcohol Abuse and Alcoholism (NIAAA) The physician's guide to helping patients with alcohol problems (NIH Publication 95-3769) Bethesda, MD: Author U.S. Department of Health; 1995. [Google Scholar]
  30. O’Muircheartaigh C, Eckman S, Smith S. Statistical design and estimation for the National Social Life, Health and Aging Project. Journal of Gerontology: Social Sciences. 2009;64(S1):12–19. doi: 10.1093/geronb/gbp045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Smith S, Jaszczak A, Graber J, Lundeen K, Leitsch S, Wargo E, et al. Instrument development, study design implementation, and survey conduct for the National Social Life, Health and Aging Project. Journal of Gerontology: Social Sciences. 2009;64(S1):20–29. doi: 10.1093/geronb/gbn013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Umberson D. Family status and health behaviors: Social Control as a dimension of social integration. Journal of Health and Social Behavior. 1987;28:306–319. [PubMed] [Google Scholar]
  33. Umberson D. Gender, marital status, and the social control of health behavior. Social Science and Medicine. 1992;34:907–917. doi: 10.1016/0277-9536(92)90259-s. [DOI] [PubMed] [Google Scholar]
  34. Umberson D, Crosnoe R, Reczek C. Social relationships and health behavior across the life course. The Annual Review of Sociology. 2010;36:139–157. doi: 10.1146/annurev-soc-070308-120011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Umberson D, Liu H, Powers D. Marital status, marital transitions, and body weight. Journal of Health and Social Behavior. 2009;50:327–343. doi: 10.1177/002214650905000306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Voorhees C, Young DR. Personal, social, and physical environmental correlates of physical activity levels in urban Latinas. American Journal of Preventive Medicine. 2003;25:61–68. doi: 10.1016/s0749-3797(03)00166-1. [DOI] [PubMed] [Google Scholar]
  37. Wenger GC. A network typology: From theory to practice. Journal of Aging Studies. 1991;5(2):147–162. [Google Scholar]
  38. Yun EH, Kang YH, Lim MK, Oh J-K, Son JM. The role of social support and social networks in smoking behavior among middle and older aged people in rural areas of South Korea: A cross- sectional study. BMC Public Health. 2010;10:78. doi: 10.1186/1471-2458-10-78. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES