Background Previous studies have shown social activity is associated with reduced risk of health outcomes. However, among older people (≥65 years) who were socially inactive at baseline, limited study explored whether increased participation in social activity in later life was associated with reduced risk of health outcomes; therefore, using the data from the Chinese Longitudinal Healthy Longevity Survey, the study was performed. Methods The study outcomes were 10-year all-cause mortality (sample number = 9,984) and 10-year heart diseases (sample number = 7,496). The exposure was the change of social activity frequency. Cox regression analysis was used for data analysis. Results During the follow-up, there were 6,407 all-cause mortalities and 1,035 heart diseases, respectively. Kaplan–Meier analysis demonstrated that cumulative incidences of all-cause mortality were significantly lower in participants with changes into more frequent social activity (log-rank p < 0.001), while no significant difference was observed for heart diseases (log-rank p = 0.330). Compared with the subgroup who never participated in social activity at baseline, adjusted HRs of all-cause mortality were 0.79 (95% CI: 0.70–0.90, p < 0.001), 0.78 (95% CI: 0.63–0.96, p = 0.019), 0.74 (0.59–0.92, p = 0.006), and 0.70 (95% CI: 0.56–0.88, p = 0.002) for the subgroup of switching to sometimes, the subgroup of switching to once a month, the subgroup of switching to once a week, and the subgroup of switching to everyday, respectively. The corresponding HRs of heart diseases were 0.83 (95% CI: 0.65–1.08, p = 0.170), 0.82 (95% CI: 0.51–1.31, p = 0.412), 0.91 (0.58–1.42, p = 0.675) and 0.75 (95% CI: 0.47–1.20, p = 0.227), respectively. Stratified and sensitivity analyses revealed similar results. Conclusion Among older people who never participated in social activity, increased participation in social activity in later life was associated with reduced risk of all-cause mortality, but was not associated with reduced risk of heart diseases.