Abstract Body

Background: Among people living with HIV (PLHIV), depression has been associated with poorer HIV-related outcomes, including increased transmission risk, poorer ART adherence, and more rapid CD4 decline or progression to AIDS and death. Little is known about trends in depression at ART initiation among PLHIV in low- and middle-income countries (LMICs) during the early years of expansions in access to ART. We used data from the Uganda AIDS Rural Treatment Outcomes (UARTO) study, an ongoing cohort of treatment-naive PLHIV begun in 2005, to estimate secular trends in depression among PLHIV at ART initiation and to understand the factors explaining these trends. Methodology: A linear regression model was fitted to the baseline data with depression symptom severity, measured using a locally adapted version of the Hopkins Symptom Checklist (HSCL-D), as the outcome variable and year of cohort entry (2005-2012) as the explanatory variable. We adjusted our estimate using baseline socio-demographic characteristics (age, gender, educational attainment, marital status, household asset wealth, and employment status). Next, we assessed physical health score and CD4 count as potential mediators of the secular trend in depression symptom severity by adding the variables of interest to the model and re-assessing the statistical significance of the regression coefficient for year of entry. Results: 498 persons with complete data, including 355 women (71.3%), were included in the models. 140 (28.1%) respondents had HSCL-D scores consistent with probable depression (women, 34.7% vs. men, 11.9%; p<.001). A linear regression model revealed a statistically significant negative association between year of entry into cohort and HSCL-D (b = -.04; 95% CI, -.06 to -.02) at entry, suggesting a 2.5% relative decline in the mean depression symptom severity score at ART initiation in each year of study recruitment after the first year. After inclusion of baseline socio-demographic characteristics, the adjusted regression coefficient for year of entry remained statistically significant. The addition to the model of physical health score, but not CD4 count, caused a reduction in magnitude and loss of statistical significance for the regression coefficient for year of entry (b=-.014; 95% CI, -.034 to .007). Conclusions: Mean depression severity scores at ART initiation declined over time among PLHIV in a rural Ugandan cohort. This secular trend appeared to be explained by better physical health at ART initiation. Secular improvements in depression at ART initiation may be related to ART being initiated among patients with better physical health, which warrants further study and may provide an important further rationale for earlier ART initiation in LMICs.