The relationship between depressive symptomatology and risky sexual behaviors has been controversial in literature.

The current study aims to reexamine the relationship between depression and sexual behaviors among men who have sex with men (MSM) using different sets of analytical assumption.

Six hundred twenty MSM were recruited in a lesbian, gay, bisexual, and transgender community center in Taiwan to participate in a cross‐sectional survey. An additional variable of squared depressive symptomatology was used to detect nonlinearity between depressive symptomatology and logit‐transformed unprotected anal intercourse (UAI) and unprotected oral sex (UOS). Multivariable logistic regression was applied to further estimate the relationship among the three variables.

Depressive symptomatology was measured by the Beck Depression Inventory (BDI), UAI, and UOS.

Depressive symptomatology had a nonlinear relationship with unprotected anal and oral sex. While linear BDI scores of MSM were not associated with unprotected sexual behaviors in the logistic model, their scores were significantly associated with unprotected sexual behaviors in the model that included both the linear (UAI, odds ratio [OR] = 1.087; P < 0.01; UOS, OR = 1.099, P < 0.01) and quadratic BDI scores (UAI, OR = 0.998, P < 0.01; UOS, OR = 0.997, P < 0.01). The relationship between BDI scores and the probability of unprotected sexual behaviors corresponded to an inverted U‐shaped curve, as opposed to a straight line.

Our findings suggest that depressive symptomatology has a significant curvilinear relationship with unprotected sexual behaviors. MSM with moderate levels of depression may be at elevated risk of engaging in unprotected sexual behaviors in comparison to their peers who exhibit either significantly higher or lower depression scores.