Browsing by Author "Gray, Christine"

Researchers have been searching for ways to improve outcomes for orphaned and separated children (OSC) worldwide. OSC have a particularly high rate of mental health disorders and lower emotional wellbeing. Stigma has been shown to be a predictor of mental health disorders and emotional wellbeing for HIV and children in poverty. However, no research has been conducted with OSC examining the relationship between self-stigma and emotional wellbeing. Using Round 10 of the Positive Outcomes for Orphans (POFO) study with 2013 orphans from Kenya, Ethiopia, Tanzania, India, and Cambodia, a linear model was implemented to examine the association between self-stigma and emotional wellbeing. Through the building of a linear regression model, self-stigma was shown to be a strong predictor of emotional wellbeing as measured by the Strengths and Difficulties Questionnaire (SDQ). This indicates that self-stigma may be a significant factor to address when looking at ways to improve emotional wellbeing among orphans.

Background: While nationwide health surveys commonly assess the prevalence of mental health conditions and disability status at the population level, they often fail to elucidate the relationship between mental health and disability. The aim of this study is to estimate the association between past-month poor mental health days and two indicators of disability (difficulty doing errands alone and difficulty making decisions) among sexual and gender minority (SGM) respondents to the 2021 BRFSS survey. Methods: A secondary analysis was conducted on the publicly accessible 2021 BRFSS data to estimate the association between past-month poor mental health days and indicators of disability among SGM. Logistic regression models were used to report odds ratios and 95% confidence intervals. We then examined potential effect modification by gender, sexual orientation, race, and SES factors, and present stratified estimates as indicated. Results: We observed increasing difficulty of completing errands alone with increasing past-month poor mental health days (OR 2.64, 95% CI 2.194, 3.178 at moderately poor mental health; OR 5.025, 95% CI 4.289, 5.889 at severely poor mental health). This association is modified by gender and SES. We also observed increasing difficulty of making decisions alone with increasing past-month poor mental health days (OR 3.298, 95% CI 2.871, 3.787 at moderately poor mental health; OR 6.792, 95% CI 5.979, 7.716 at severely poor mental health). This association is modified by sexual orientation, gender and race. Conclusions: There are clear dose response relationships between mental health and the two disability outcomes that are modified by socioeconomic status, gender and sexual identity.

Background: Alcohol use accounts for over 3 million deaths each year and causes negative mental and physical health outcomes. Both adolescent use and early use initiation are associated with future dependence and alcohol-related harm. Orphaned and separated children (OSC) in low- and middle-income countries (LMICs) are particularly vulnerable to alcohol use. Methods: A cross-sectional analysis was used to explore predictors of adolescent alcohol use and early alcohol initiation among a sample of OSC in 5 LMICs. Bivariable logistic regression models were used to evaluate associations between age, gender, education, care setting, OSC status (single orphan, double orphan, neither), behavior and emotional health, the experience of potentially traumatic events (PTEs), social support, and participation in chores or work and the outcomes of early initiation and recent alcohol use. Variables with p-value <0.15 were included in a multivariable model for that respective outcome and evaluated through multivariable logistic regressions. Results: In multivariable regressions, age [OR 1.28 (1.11-1.48)], community-based setting [OR 2.74 (1.57-5.02)], non-abuse trauma [OR 5.23 (2.51-12.10)], and physical or sexual abuse [OR 1.27 (0.75-2.11)] were associated with recent use. Non-abuse trauma [OR 12.86 (3.83-80.12)] and work [OR 1.72 (0.91-3.12)] were associated with early use. Conclusions: Interventions targeting alcohol use among OSC may be more effective if they target predictors identified in this study, such as a community care setting.

Background: Greenspaces have been shown to promote psychological health; they are understudied in regions facing epidemiologic and demographic shifts towards increased mental illness burden and a more urban population, including Ethiopia. This exploratory cross-sectional analysis characterizes greenspace distribution in Addis Ababa and estimates the overall and stratified (income and sex) associations between greenspaces and depression symptomology. Methods: Data are from young adults living in Addis Ababa in the Positive Outcomes for Orphans study. The normalized difference vegetation index (NDVI) measured greenspace exposure. The Patient Health Questionnaire 8-item scale measured depressive symptoms. Tests of spatial clustering characterized greenspace distribution. Adjusted linear regression models estimated associations between greenspace and depressive symptoms. Results: Among our analytic population (n=206; 51% male; 49% female) the median age was 22(IQR: 20-23), mean NDVI was 0.14(SD:0.44), and mean PHQ-8 was 5.33(SD:4.60). Tests of clustering revealed green-rich areas in the northeast region of the city and green-deprived areas in the center of the city. Adjusted models suggest a slight association between greenspace and depressive symptoms (β = -1.10, 95% CI: -2.76,0.57). The association was strengthened among those with income (β = -1.52, 95% CI: -4.09,1.05) and among males (β = -2.58, 95%CI: -5.38,0.21), but null among those with no income (β = -0.01, 95%CI: -1.89,1.87), and females (β = 0.26, 95%CI: -1.08, 1.58). Conclusion: Though the sample is small and estimates imprecise, this analysis suggests there may be an association between greenspace and depressive symptoms that should be explored further in pursuit of population mental health in Ethiopia.

Adolescent girls' sexual and reproductive health (SRH) is a significant public health issue, but little is known about the SRH outcomes of orphaned and separated adolescent girls, who face a higher risk of abuse and neglect in the absence of parental guidance, underscoring the critical need for research. This study examines the SRH outcomes (early pregnancy and condom use) and their predictors among adolescent girls aged 15 to 21 in five low- and middle-income countries, including orphaned and separated children. We conducted a cross-sectional analysis of 933 participants during the latest round of data collection from a longitudinal cohort study to assess predictors of early pregnancy and condom use. We first used bivariable logistic regression to identify potential predictors of each outcome. Potential predictors included education level, abuse, working status, relationship status, and age. Predictors that were significant (p<=0.10) in crude analyses were added to a multivariable model to assess associations in the presence of other predictors. In bivariable analysis, we found that education, marital status, history of abuse, work status, and age are associated with early pregnancy. However, we did not find a significant association for condom use with any predictors. It should be noted that the findings may be imprecise due to the small sample sizes and exclusion of children under 16 years old. Further research is needed to explore targeted interventions aimed at reducing early pregnancy and promoting condom use. Addressing education, work, and relationship status may enhance the effectiveness of such interventions.