A Social Media Study on the Effects of Psychiatric Medication Use

By Koustuv Saha, Benjamin Sugar, John B. Torous, Bruno D. Abrahao, Emre Kıcıman, M. de Choudhury. 2019

110 citations recorded by the source corpus.

Understanding the effects of psychiatric medications during mental health treatment constitutes an active area of inquiry. While clinical trials help evaluate the effects of these medications, many trials suffer from a lack of generalizability to broader populations. We leverage social media data to examine psychopathological effects subject to self-reported usage of psychiatric medication. Using a list of common approved and regulated psychiatric drugs and a Twitter dataset of 300M posts from 30K individuals, we develop machine learning models to first assess effects relating to mood, cognition, depression, anxiety, psychosis, and suicidal ideation. Then, based on a stratified propensity score based causal analysis, we observe that use of specific drugs are associated with characteristic changes in an individual’s psychopathology. We situate these observations in the psychiatry literature, with a deeper analysis of pre-treatment cues that predict treatment outcomes. Our work bears potential to inspire novel clinical investigations and to build tools for digital therapeutics.

Subfields: Applied Psychology, Experimental and Cognitive Psychology, Social Psychology, Medical & Health AI, Social Sciences & Humanities

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Applied Psychology

1. J. Firth
2. Matcheri S. Keshavan
3. John B. Torous

In Experimental and Cognitive Psychology

1. Nadav Oved
2. Matcheri S. Keshavan
3. Brendan T. O'Connor
4. Roi Reichart
5. John B. Torous
6. Amir Feder

In Social Psychology

1. Nadav Oved
2. Matcheri S. Keshavan
3. Roi Reichart
4. John B. Torous
5. Amir Feder

In Medical & Health AI

1. Roi Reichart

Citations in the corpus

• CausaLM: Causal Model Explanation Through Counterfactual Language Models (2020, methodology): “Specifically, researchers have been looking into methods of measuring other confounders via text (Pennebaker, Francis, and Booth 2001; Saha et al. 2019), or using text as confounders (Johansson, Shalit, and Sontag 2016; De Choudhury et al. 2016; Roberts, Stewart, and Nielsen 2018).”
• LibRA (2019, unknown): “Then, drawing on natural language analysis methods, we use word-embeddings, particularly pre-trained GloVe vectors [113, 123] to project the role descriptions of individuals and companies in a 50-dimensional word-vector space, so as to obtain rich lexico-semantic context surrounding the hand-curated”
• The growing field of digital psychiatry: current evidence and the future of apps, social media, chatbots, and virtual reality (2021, methodology): “These methods have also been employed in studies exploring psycho­ social reactions to the COVID­19 pandemic 64,65 , as well as the ef­ fects of psychiatric medications 66 .”
• Text and Causal Inference: A Review of Using Text to Remove Confounding from Causal Estimates (2020, methodology): “For example, Saha et al. (2019) use categories of the Linguistic Inquiry and Word Count (LIWC) lexicon (Pennebaker et al., 2001) such as tentativeness, inhibition, and negative affect, and use indicators of these categories in the text as confounders.”