Citation

Peng S, Yang XY, Rockett IRH. Soc. Sci. Med. (1982) 2019; 243: e112640.

Affiliation

School of Public Health, West Virginia University, Morgantown WV, USA; Department of Psychiatry, University of Rochester Medical Center, Rochester NY, USA.

Copyright

(Copyright © 2019, Elsevier Publishing)

DOI

10.1016/j.socscimed.2019.112640

PMID

31665656

Abstract

RATIONALE: Social capital is an important theoretical construct often used to explain positive behavioral outcomes. However, suicide research is often hampered by a limited definition of social capital, at the expense of its multidimensional nature, since suicide may occur when individuals are beset by overwhelming and conflicting relationships. We hypothesize not all dimensions of social capital protect individuals from suicidal ideation, and the beneficial return from some types also gradually diminishes.

METHODS: We conducted a hierarchical ordinal regression analysis of combined samples of university students from 22 cities across all regions of China (n = 5025, response rate 97%).

RESULTS: 1) whereas social capital, measured as network connection, embedded-resource, and generalized trust, is negatively associated with self-reported suicidal ideation, controlling for mental stress and other covariates, when measured as social participation, shows a positive association; 2) the moderate curvilinear effects of embedded resource and network connection imply that the return from some dimensions of social capital gradually diminishes.

CONCLUSIONS: We leveraged the results to discuss the theoretical relevance of social capital in the suicide literature -particularly its dark side- and the need for more contextualized public health intervention programs for suicide prevention.

Copyright © 2019. Published by Elsevier Ltd.

Language: en

Keywords

Hierarchical ordinal regression; Social capital; Suicidal ideation; Youth