Using Natural Language Processing to Identify Depression in Patients with HIV

Using Natural Language Processing to Identify Depression in Patients with HIV

Brief description of study

The purpose of this study is to develop algorithms that may help detect depressive symptoms in social media posts, specifically among men who have sex with men and who have HIV.

Detailed description of study

If you agree to join the study, you will be asked to complete the following research procedures:

● Several short self-report surveys. These surveys will ask you basic demographic questions (e.g., your age, gender, race/ethnicity) and will ask you questions about how you’ve been feeling lately, related to depression. You will also be asked sensitive questions regarding your drug use and sexual practices.
● A diagnostic interview with a trained clinical evaluator. This interview will be audio-recorded for training purposes. The diagnostic interview will not be conducted for clinical purposes and the results will not be entered into your medical record.
● You will also share your social media posts (from Twitter, Facebook, and Reddit) with the research team. Using an application, we will collect information on your social media posts (content, date, and time), and other information that you’ve shared on your profile/account such as self-descriptions and information about your friend network. We will not analyze any photographs you might have posted. We will not look at your social media after you leave today. Keep in mind, you will need to agree to separate online consent forms during the course of the study in order to share your Facebook, Twitter, and Reddit data.

Your participation will last for approximately 3 hours.

Eligibility of study

You may be eligible for this study if you meet the following criteria:

  • Conditions:
    Depression,HIV
  • Age: Between 18 Years - 99 Years
  • Gender: Male

Participants must:

- Have HIV
- Be a man who has sex with men
- Use social media frequently (Facebook, Twitter, and/or Reddit)

Updated on 10 Mar 2023. Study ID: 833076

Interested in the study

Select a study center that’s convenient for you, and get in touch with the study team.

Contact a study center