PITTSBURGH, June 7, 2018 – Negative experiences on social media carry more weight than positive interactions when it comes to the likelihood of young adults reporting depressive symptoms, according to a new University of Pittsburgh analysis.

 

The finding, reported today in the journal Depression and Anxiety, may be useful for designing interventions and clinical recommendations to reduce the risk of depression.

 

“We found that positive experiences on social media were not related or only very slightly linked to lower depressive symptoms. However, negative experiences were strongly and consistently associated with higher depressive symptoms,” said lead author Brian Primack, M.D., Ph.D., dean of the Honors College and director of the Center for Research on Media, Technology and Health at Pitt. “Our findings may encourage people to pay closer attention to their online exchanges. Moving forward, these results could assist scientists in developing ways to intervene and counter the negative effects while strengthening the positive ones.”

 

In August 2016, Primack and his team surveyed 1,179 full-time students ages 18 to 30 at the University of West Virginia about their social media use and experiences. The participants also completed a questionnaire to assess their depressive symptoms. 

 

Each 10 percent increase in positive experiences on social media was associated with a 4 percent decrease in odds of depressive symptoms, but those results were not statistically significant, meaning that the finding could be due to random chance. However, each 10 percent increase in negative experiences was associated with a 20 percent increase in the odds of depressive symptoms, a statistically significant finding.

 

“It is valuable to know that positive and negative experiences are very differently related to depression,” said Primack. “But we don’t know from our study whether the negative social media interactions actually caused the depressive symptoms or whether depressed individuals are more likely to seek out negative online interactions. As with many things in social science, the answer is probably some combination of the two, but more research will be needed to disentangle cause and effect.”

 

Other characteristics also were linked to the participants having depressive symptoms. For example, compared with men, women had 50 percent higher odds of having depressive symptoms. Identifying as non-white and having only completed “some college,” rather than completing a degree, also were associated with higher odds of depressive symptoms. All of these characteristics have previously been shown to increase a person’s likelihood of depression.  

 

While the findings still need to be replicated, Primack said public health practitioners could start using them to educate the public of the risks of negative social media interactions. He also points out that cyberbullying occurs not only among adolescents, but also among adults. Universities, workplaces and community spaces could use the findings to increase awareness around positive and negative social media experiences.

 

Primack noted that health care professionals working with depressed patients could suggest strategies to improve the quality of online experiences, such as restricting time spent on social media to reduce the number of negative interactions and “unfriending” people or groups that tend to enable negative experiences.

 

Although the finding was not statistically significant, Primack said that increasing the opportunities for positive experiences on social media is still likely to be worthwhile.

 

“In other studies, engaging in certain forms of social media use has been shown to enhance communication and social connection,” he said. “Certainly, there are many situations in which connecting with others in this way might actually lower depressive symptoms. That just wasn’t the primary finding in this particular study.”

 

Additional authors on this research are Meghan A. Bisbey, B.A., Ariel Shensa, M.A., Sabrina A. Karim, B.A., and Jaime E. Sidani, Ph.D., all of Pitt; and Nicholas D. Bowman, Ph.D., and Jennifer M. Knight, M.A., both of West Virginia University. 

 

This research was funded by the Fine Foundation.