Analyzing Multilevel Factors As Predictors of Mental Health with Diverse Young Adults Seeking Care for HIV Services

Background and Purpose:

8.8 million young adults report having a mental illness (2019). Mental health does not exist in isolation, rather many multi-level factors shape mental health. Various factors, individual (gender, race/ethnicity, age), mezzo (access to care; discrimination), macro (policies; geographic location) shape one’s mental health. Little is known about how multilevel predictors in the US South impact mental health. Inequitable protections exist for residents as notably, some southern states did not expand Medicaid, stripping important physical and mental healthcare access for young adults. Understanding multilevel factors is important to develop interventions that change conditions and environments, not solely individual behaviors. The aim of this study is to understand the multilevel predictors of mental health among ages 18-24 in the US South.

Methods:

Data were collected using a cross-sectional survey with young adults (18-24; N=563) attending a college/university between 2015-2018 in Houston as part of a SAMHSA-funded study to promote sexual and mental health awareness and engagement in HIV testing. A convenience sample of young adults (28.9% African American, 21% Asian, 41.7% White, 38.3% Hispanic) was recruited through peer referrals and outreach events. Fifty-four percent were male (46% female) and the majority of participants attended high school in the US South (82.4%). Participants’ mean age was 22.25 and 67.6% had an annual household income below national average ($57,617).

Feelings of discrimination, health coverage, and difficulty getting medical care were examined as predictors. Descriptive statistics captured multi-level factors: gender, income, and various health seeking behaviors. Chi-squares tested for associations between demographic variables, multi-level factors and perceptions of mental health. Variables supported by empirical literature and significant Chi-square results were considered for prediction modeling. Logistic regression analyses were performed to predict perception of mental health. Three predictor variables with statistically significant associations were included in the model: feelings of discrimination, health coverage, and difficulty getting medical care.

Results:

Results of the binary logistic regression analysis indicated the full model, consisting of all multi-level predictors, was statistically significant (X²=20.034, df=3, N=563, p<.000). The odds of having a poor perception of mental health increased when the participant did not have health insurance (OR=.648), experienced feelings of discrimination (OR=.505) and difficulty getting medical care (OR=.577). The model accounts for 63.6% of the variance and there were no statistically significant interaction effects. Key findings suggest structural barriers or experiences are important correlates of poor perceptions of mental health.

Implications:

Key Findings suggest that multilevel frameworks should be used to change how mental health services are designed and implemented with diverse young adults. Feelings of discrimination, health coverage, and difficulty getting medical care predicted individuals’ perceptions of mental health. These findings are critical to informing advocacy for mental health intervention planning. Future studies could examine not only multi-level factors that predict perceptions of mental health, but also take into account cultural characteristics for a given ethnic or racial group. Analyzing predictors of mental health are key to adapting clinical and community-based interventions designed to reduce barriers to accessing care while increasing successful treatment.