Introduction: To explore gender differences in the relationship between loneliness and health-related behavioral risk factors (BRFs) among the Hakka elderly.

Methods: Loneliness was measured by the UCLA Loneliness Scale Short-form (ULS-8). Seven BRFs were examined. Mann-Whitney U, Kruskal-Wallis, and post hoc tests were conducted to compare the differences in ULS-8 scores among the Hakka elderly with different BRFs. Generalized linear regression models were employed to examine the associations of specific BRF and its number with the ULS-8 scores among the Hakka elderly in male, female, and total samples.

Results: Physical inactivity (B = 1.96, p < 0.001), insufficient leisure activities participation (B = 1.44, p < 0.001), unhealthy dietary behavior (B = 1.02, p < 0.001), and irregular sleep (B = 2.45, p < 0.001) were positively correlated with the ULS-8 scores, whereas drinking (B = -0.71, p < 0.01) was negatively associated with the ULS-8 scores in the total sample. In males, insufficient leisure activities participation (B = 2.35, p < 0.001), unhealthy dietary behavior (B = 1.39, p < 0.001), and irregular sleep (B = 2.07, p < 0.001) were positively associated with the ULS-8 scores. In females, physical inactivity (B = 2.69, p < 0.001) and irregular sleep (B = 2.91, p < 0.001) was positively correlated with the scores of ULS-8, while drinking (B = -0.98, p < 0.05) was negatively associated with the ULS-8 scores. More BRFs were significantly related to greater loneliness (p < 0.001).

Conclusion: There are gender differences in the relationship between loneliness and BRFs among the Hakka elderly, and individuals with more BRFs were more likely to feel loneliness. Therefore, the co-occurrence of multiple BRFs requires more attention, and integrated behavioral intervention strategies should be adopted to reduce the loneliness of the elderly.