A new analysis has found that, for African Americans with major depressive disorder (MDD), friendship’s mitigating benefits can vary based on education levels.

“These findings underscore the complexity of social support as a possible intervening process in depression,” wrote Ann W. Nguyen, PhD, of Case Western Reserve University in Cleveland, and her coauthors. The study was published in the Journal of Affective Disorders.

To determine how certain elements of friendship affect MDD, the researchers analyzed 3,434 responses from African Americans to the National Survey of American Life. They assessed variables such as “subjective closeness to friends,” “frequency of contact with friends,” “receipt of support from friends,” and “provision of support to friends” via responses to related questions, along with factoring in the impact of education level on 12-month MDD. Analysis was performed via logistic regression.

Among all respondents, the 12-month prevalence of MDD was 6.7%. Overall, subjective closeness and frequency of contact with friends were negatively associated with 12-month MDD. Those with higher levels of education saw their probability of meeting MDD criteria decrease as frequency of contact increased; those with lower levels of education saw no association between frequency of contact and 12-month MDD. There was a similar association between receipt of support or provision of support and education: The high education group saw their probability of MDD decrease as receipt or provision of support increased.

The authors acknowledged their study’s limitations, including the impossibility of causal inferences because of its cross-sectional design. In addition, the survey only included community-dwelling adults, meaning its findings cannot be extended to institutionalized and homeless individuals. Also, each friendship variable was assessed through a single question. “Future research,” they noted, “should assess these relationship dimensions using multi-item scales, as they tend to be more stable, reliable, and precise.”

No conflicts of interest were reported.

SOURCE: Nguyen AW et al. J Affect Disord. 2019 Jun 15. doi: 10.1016/j.jad.2019.04.013