A new paper has researched whether social media algorithms can be used in predicting and diagnosing depression.
The paper uses social media algorithms to assess several factors such as the content of social media posts to assess a consenting user’s mental health, and advocates diagnosing depression in this way in the future.
The study is by University of Pennsylvania and Stony Brook University and the paper has has been published in the Proceedings of the National Academy of Sciences.
How did they use social media posts to assess mental health?
The study used Facebook posts of consenting individuals from patients in an academic emergency department. There were 683 participants in the study, 114 of which had previous diagnoses of depression.The mean age was 29.9 years old, and most participants were female (76.7%) and black (70.1%).
The study used the language in the Facebook posts to see if they could accurately identify people with depression and predict diagnosing depression in the future. They built a social media algorithm using the textual content of the Facebook posts, post length, frequency of posting, temporal posting patterns, and demographics.
The study identified several language markers in the Facebook posts were helpful in diagnosing depression using social media algorithms.
These markers include:
- Preoccupation with the self; and
Diagnosing depression with social media algorithms
The authors concluded that social media algorithms can be used in diagnosing depression. They said: “In this study, we show that the content shared by consenting users on Facebook can predict a future occurrence of depression in their medical records. This study suggests that an analysis of social media data could be used to screen consenting individuals for depression.”
Similarly, the authors added: “Unobtrusive depression assessment through social media of consenting individuals may become feasible as a scalable complement to existing screening and monitoring procedures.”