Social Networks and Disability Transitions Across Eight Intervals of Yearly Data in the New Haven EPESE

• Published in: The journals of gerontology. Series B, Psychological sciences and social sciences Psychological sciences and social sciences, 1999-05, Vol.54B (3), p.S162-S172
• Main Authors: de Leon, Carlos F. Mendes, Glass, Thomas A., Beckett, Laurel A., Seeman, Teresa E., Evans, Denis A., Berkman, Lisa F.
• Format: Article
• Language: English
• Subjects: Activities of Daily Living; Adult. Elderly; Aged; Aging - psychology; Biological and medical sciences; Cross-Sectional Studies; Developmental psychology; Disability; Disabled Persons - psychology; Female; Fundamental and applied biological sciences. Psychology; Gerontology; Health; Humans; Life Style; Male; Older people; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Risk Assessment; Social research; Social Support
• ISSN: 1079-5014; 1758-5368
• DOI: 10.1093/geronb/54B.3.S162

Objectives. There is considerable evidence that social networks are strongly related to survival and other health outcomes. However, findings regarding the effect of social networks on disability outcomes have been inconsistent. This study examines this relationship with respect to the risk of developing disability and recovering from disability. Methods. Data come from a community-based sample of the New Haven population aged 65 years and older, with nine annual interviews conducted between 1982 and 1991. Disability was measured by a 6-item index of activities of daily living (ADL), and a 3-item Rosow-Breslau index, with disability defined as impairment in one or more tasks on each measure. Social network variables were constructed for each of four domains of ties: children, relatives, friends, and a confidant, and a summary measure of total social networks. A Markov model was used to estimate one-year disability transitions averaged across all 8 intervals, after controlling for sociodemographic and health-related variables. Results. Total social networks was associated with a significantly reduced risk of developing ADL disability (β = -0.009, p < .01), and a significantly increased likelihood of ADL recovery (β = 0.017, p < .01). Emotional and instrumental support did not affect the protective effect of social networks against disability, but partially accounted for their effect on enhanced recovery. Network variables related to relatives and friends were significantly associated with disability and recovery risks, but those related to children or a confidant were not. The associations with disability transitions as measured by the Rosow-Breslau index were generally smaller and nonsignificant. Discussion. The findings lend further support for the role of social relationships in important health outcomes in old age. They suggest that being “embedded” in a social network of relatives and friends reduces risk for ADL disability, and enhances recovery from ADL disability.