The purpose of this research was to investigate the effects of social support on the health and safety of blue-collar employees (N = 220) in a unionized appliance manufacturer. The research design encompassed two time periods: four months preceding and four months following knowledge of anticipated plant closure. The key variables were injury occurrences, safety, social support, and job strain, in addition to gender, work type, age, education, and employment tenure. The first two hypotheses were based upon a comparison between Primary (n = 90), Assembly (n = 101), and Maintenance (n = 29) groups of employees. They predicted that the Assembly group would have fewer injury occurrences during both time periods, due to the buffering effects of social support received within their work teams. The second two hypotheses compared men and women. They predicted that women would have fewer injury occurrences than men during both time periods, due to the moderating effects of social support. The theoretical demand-control model of job strain (Karasek, 1979) and the expanded demand-control-support model (Johnson, 1986) were used as the basis for the investigation.The results did not support the first two hypotheses. The Assembly group did not have fewer injury occurrences than the other two groups during Time 1, and they sustained a significantly higher mean number of injury occurrences than the other two groups during Time 2 (p $$.01). However, the results supported the basic predictions of the theoretical model that low support, low control, and high job demands, contributed to adverse health outcomes. The results did not support the third hypothesis; women and men had relatively similar injury occurrences during Time 1. The fourth hypothesis was supported; men had significantly higher mean number of injury occurrences as compared to women during Time 2 (p $$.01). The results from the latter two hypotheses were difficult to explain due to limited research on gender responses to work-related stress. The association between work, status, role identity, and potential job loss may have contributed to the extent of job strain observed in men.