Loneliness linked to depression in Chinese men who acquired HIV sexually

Men living with HIV in China who acquired the virus through sexual transmission face a markedly elevated risk of depressive symptoms when they feel lonely, according to a cross-sectional study published in Scientific Reports. The research, led by Rubin Chen and Yixiong Zheng of Xiamen Xianyue Hospital, examined 436 men and found that those reporting high loneliness had nearly five times the odds of moderate-to-severe depressive symptoms compared with men reporting lower loneliness, even after accounting for a range of demographic and clinical factors. Social support, by contrast, did not independently predict the depression outcome in the fully adjusted model, although a small statistical pathway linking loneliness to depressive symptom scores through social support emerged in a more sophisticated mediation analysis.

The study addresses a gap that has persisted in the psychosocial literature on HIV. Loneliness and social support are widely recognized as correlates of depression among people living with HIV, but the precise interrelationships among the three constructs have remained insufficiently understood. Depression is itself a clinically consequential condition in this population: it is associated with poorer antiretroviral therapy adherence, faster disease progression, reduced quality of life, and elevated suicide risk. Understanding which psychosocial variables carry the strongest independent signal for depressive symptoms, and whether some of their influence operates indirectly through others, therefore has direct implications for screening and intervention design in HIV care settings.

Participants were recruited from the Department of Infectious Diseases and the Treatment and Follow-up Clinic of the First Affiliated Hospital of Xiamen University, with additional involvement of a hospital in the Pingtan Comprehensive Experimental Zone. All were men who had acquired HIV through sexual transmission, the dominant transmission route in China’s current epidemic. Each participant completed three standardized instruments: the Patient Health Questionnaire-9, a nine-item screening tool that yields a continuous depression score and supports a conventional cutoff for moderate-to-severe symptoms; the Social Support Rating Scale, a measure developed for Chinese populations that captures objective support, subjective support, and support utilization; and the UCLA Loneliness Scale, a widely used self-report index of perceived social isolation. Electronic informed consent was obtained from all participants, and the protocol was approved by the Ethics Committee of Xiamen Xianyue Hospital in accordance with the Declaration of Helsinki.

The analytical strategy combined three complementary techniques. Correlation analysis first established the bivariate associations among loneliness, social support, and depressive symptoms. Multivariable logistic regression then modeled the odds of moderate-to-severe depressive symptoms, dichotomized on the Patient Health Questionnaire-9, as a function of loneliness and social support while adjusting for covariates. Finally, latent-variable mediation analysis tested whether social support statistically transmitted part of the association between loneliness and depressive symptom severity. Latent-variable approaches treat each construct as a factor underlying its indicator items rather than as a simple summed score, which reduces measurement error and yields more precise estimates of the structural pathways among the variables.

The headline result concerns loneliness. In the fully adjusted logistic regression model, high loneliness was associated with an adjusted odds ratio of 4.790 for moderate-to-severe depressive symptoms, with a 95 percent confidence interval spanning 2.778 to 8.259. In other words, after controlling for the other variables in the model, men in the high-loneliness category had roughly 4.8 times the odds of crossing the moderate-to-severe depression threshold compared with their less lonely counterparts. The confidence interval lies entirely above unity, indicating a statistically robust association that is unlikely to reflect sampling fluctuation alone. An effect of this magnitude in psychosocial epidemiology is substantial, particularly for a modifiable perceived state rather than a fixed demographic characteristic.

Social support told a different story. Total social support was not independently associated with the dichotomized depressive symptom outcome in the fully adjusted model, with an adjusted odds ratio of 0.985 and a 95 percent confidence interval of 0.957 to 1.013. Because this interval includes 1.0, the data do not provide evidence that total social support, on its own, predicts whether a man crosses the moderate-to-severe depression cutoff once loneliness and the other covariates are held constant. The authors note that this null finding for the dichotomized outcome contrasts with the persistent popular assumption that bolstering social support will directly reduce clinically significant depression in this population; the picture that emerges instead is one in which loneliness carries the dominant independent signal.

The mediation analysis added nuance to that conclusion. In the adjusted latent-variable mediation model, loneliness showed a small indirect association with depressive symptom scores through social support, with an indirect effect of 0.011 and a 95 percent bias-corrected bootstrap confidence interval of 0.001 to 0.034. Because the bootstrap interval excludes zero, the indirect pathway is statistically distinguishable from no effect, but its magnitude is small. The interpretation is that some of the statistical association between loneliness and depression operates through reduced social support, yet the bulk of the loneliness-depression relationship is direct rather than mediated. Bootstrap resampling, the technique used to derive that confidence interval, repeatedly draws samples with replacement from the observed data to estimate the sampling distribution of the indirect effect without assuming normality, which is the preferred approach for mediation effects that are known to be non-normally distributed.

Several technical caveats frame these findings. The study is cross-sectional, meaning loneliness, social support, and depressive symptoms were measured at a single point in time. This design cannot establish temporal ordering: loneliness may precipitate depression, depression may amplify feelings of isolation, or third variables may drive both. Reverse causation is a live possibility, since depressed individuals often withdraw from social contact and subsequently perceive themselves as more isolated. The sample was also recruited from clinical settings in Fujian province, which may limit generalizability to men living with HIV elsewhere in China or to those not engaged in follow-up care. The dichotomization of depressive symptoms, while clinically interpretable, discards information contained in the continuous scores, and the authors accordingly modeled both the categorical outcome and the continuous symptom severity.

Within those constraints, the practical implications are concrete. The strong adjusted association between loneliness and moderate-to-severe depressive symptoms suggests that loneliness deserves routine assessment in HIV follow-up clinics, potentially alongside standard depression screening with the Patient Health Questionnaire-9. Interventions that directly target perceived isolation, such as peer support programs, structured group activities, and cognitive approaches to reframing social perception, may hold more promise for depression prevention in this group than generic support provision alone. The modest indirect pathway through social support implies that support networks matter, but that their influence on depression is largely channeled through the subjective experience of loneliness rather than through the objective availability of help. Screening tools such as the UCLA Loneliness Scale could identify men at elevated risk before depressive symptoms become severe.

The research team, which included investigators from Xiamen Xianyue Hospital, the First Affiliated Hospital of Xiamen University, Pingtan Comprehensive Experimental Zone Hospital, and Fujian Medical University, was funded by the Fujian Provincial Health Technology Innovation Project and the Xiamen Science and Technology Guidance Project. The authors declare no competing interests, and the article is open access under a Creative Commons license. The team emphasizes that longitudinal studies among men living with HIV are needed to clarify the temporal ordering and direction of the associations observed here. Until such data accumulate, the study stands as a technically rigorous signal that loneliness, more than social support per se, marks the men in China’s sexual transmission epidemic who are most likely to be struggling with depression, and that addressing perceived isolation directly may be the most efficient route to protecting their mental health.

Subject of Research: Loneliness, social support, and depressive symptoms among men who acquired HIV through sexual transmission in China

Article Title: Loneliness, social support, and depressive symptoms among men who acquired HIV through sexual transmission in China

Article References: Chen, R., Que, J., Zeng, J., Lin, Y., Wei, W., Huang, Y., Zheng, C., & Zheng, Y. (2026). Loneliness, social support, and depressive symptoms among men who acquired HIV through sexual transmission in China. Scientific Reports. https://doi.org/10.1038/s41598-026-76157-w

Image Credits: AI Generated

DOI: 10.1038/s41598-026-76157-w

Keywords: Loneliness, social, support, depressive, symptoms, acquired, sexual, transmission, China, scientific research