Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 49 of the 49 references it cites were found in a registry or an index.

Forecasting the onset and course of mental illness with Twitter data

Open access

Abstract

We developed computational models to predict the emergence of depression and Post-Traumatic Stress Disorder in Twitter users. Twitter data and details of depression history were collected from 204 individuals (105 depressed, 99 healthy). We extracted predictive features measuring affect, linguistic style, and context from participant tweets (N = 279,951) and built models using these features with supervised learning algorithms. Resulting models successfully discriminated between depressed and healthy content, and compared favorably to general practitioners’ average success rates in diagnosing depression, albeit in a separate population. Results held even when the analysis was restricted to content posted before first depression diagnosis. State-space temporal analysis suggests that onset of depression may be detectable from Twitter data several months prior to diagnosis. Predictive results were replicated with a separate sample of individuals diagnosed with PTSD (N users = 174, N tweets = 243,775). A state-space time series model revealed indicators of PTSD almost immediately post-trauma, often many months prior to clinical diagnosis. These methods suggest a data-driven, predictive approach for early screening and detection of mental illness.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41598-017-12961-9

Cites

49 of 49 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 ROC curve and top predictors for Random Forests algorithm, for depression and PTSD samples (Ndepr = 74, 990, Nptsd = 54,197). Predictor names ending in “_happy” are happiness measures; LIWC predictors37 refer to the occurrence of semantic categories (eg. LIWC_ingest refers to food and eating words, LIWC_swear refers to profanity).
Figure 1 ROC curve and top predictors for Random Forests algorithm, for depression and PTSD samples (Ndepr = 74, 990, Nptsd = 54,197). Predictor names ending in “_happy” are happiness measures; LIWC predictors37 refer to the occurrence of semantic categories (eg. LIWC_ingest refers to food and eating words, LIWC_swear refers to profanity).CC BY · PubMed Central
Figure 2 Hidden Markov Model showing probability of depression (N = 74,990). X-axis represents days from diagnosis. Healthy data are plotted from a consecutive time span of equivalent length. Trend lines represent cubic polynomial regression fits with 95% CI bands, points are aggregations of 14 day periods, with error bars indicating 95% CI on central tendency of daily values.
Figure 2 Hidden Markov Model showing probability of depression (N = 74,990). X-axis represents days from diagnosis. Healthy data are plotted from a consecutive time span of equivalent length. Trend lines represent cubic polynomial regression fits with 95% CI bands, points are aggregations of 14 day periods, with error bars indicating 95% CI on central tendency of daily values.CC BY · PubMed Central
Figure 3 Hidden Markov Model showing probability of PTSD (N = 54,197). X-axis represents days from trauma event. Healthy data are plotted from a consecutive time span of equivalent length. The purple vertical line indicates mean number of days to PTSD diagnosis, post-trauma, and the purple shaded region shows the average period between trauma and diagnosis. Trend lines represent cubic polynomial regression fits with 95% CI bands, points are aggregations of 30 day periods, with error bars indicating 95% CI on central tendency of daily values.
Figure 3 Hidden Markov Model showing probability of PTSD (N = 54,197). X-axis represents days from trauma event. Healthy data are plotted from a consecutive time span of equivalent length. The purple vertical line indicates mean number of days to PTSD diagnosis, post-trauma, and the purple shaded region shows the average period between trauma and diagnosis. Trend lines represent cubic polynomial regression fits with 95% CI bands, points are aggregations of 30 day periods, with error bars indicating 95% CI on central tendency of daily values.CC BY · PubMed Central
Figure 4 Depression word-shift graph revealing contributions to difference in Twitter happiness observed between depressed (5.98) and healthy (6.11) participants. In column 3, (−) indicates a relatively negative word, and (+) indicates a relatively positive word, both with respect to the average happiness of all healthy tweets. An up (down) arrow indicates that word was used more (less) by the depressed class. Words on the left (right) contribute to a decrease (increase) in happiness in the depressed class. See Appendix III for PTSD word-shift graph.
Figure 4 Depression word-shift graph revealing contributions to difference in Twitter happiness observed between depressed (5.98) and healthy (6.11) participants. In column 3, (−) indicates a relatively negative word, and (+) indicates a relatively positive word, both with respect to the average happiness of all healthy tweets. An up (down) arrow indicates that word was used more (less) by the depressed class. Words on the left (right) contribute to a decrease (increase) in happiness in the depressed class. See Appendix III for PTSD word-shift graph.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 49 references

Research works printed but not linked: 2

  1. De Choudhury, M., Gamon, M., Counts, S. & Horvitz, E. Predicting depression via social media. In Seventh International AAAI Conference on Weblogs and Social Media (2013).
  2. Park, M., Cha, C. & Cha, M. Depressive moods of users portrayed in Twitter. In Proceedings of the ACM SIGKDD Workshop on healthcare informatics (HI-KDD) (pp. 1–8) (2012).

Cited by

See all 401

Metadata

Authors
Reece, A. G., Reagan, A. J., Lix, K. L. M., Dodds, P. S., Danforth, C. M., & Langer, E. J.
Affiliations
  • Harvard University (US)
  • University of Vermont (US)
  • Stanford University (US)
Corresponding author
  • Reece, Andrew G.
  • Danforth, Christopher M.
Venue
Scientific Reports
Published
2017-10-11
Volume
7
Issue
1
Pages
13006-13006
ISSN
2045-2322 (ISSN-L)
OpenAlex
W2512937600
Cited by
401
References
49
Citations by year
2026: 21 · 2025: 48 · 2024: 44 · 2023: 57 · 2022: 65 · 2021: 68 · 2020: 41 · 2019: 36 · 2018: 15 · 2017: 6
FWCI
38.39
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • Digital Access to Scholarship at Harvard (DASH) (Harvard University) (submittedVersion, cc-by)
  • ScholarWorks -A service of University of Vermont Libraries (University of Vermont) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • arXiv (Cornell University)
Funders
  • National Science Foundation (1447634)
  • Harvard University
Topics
  • Social Sciences › Psychology › Social Psychology › Mental Health via Writing
  • Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Mental Health Research Topics
Keywords
depression, post-traumatic stress disorder, Twitter, social media, early detection, language style, affect, computational psychiatry, early screening
MeSH terms
Depression, Humans, Stress Disorders, Post-Traumatic, Disease Progression, Social Media

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Forecasting the onset and course of mental illness with Twitter data. https://app.cochat.ai/papers/10.1038/s41598-017-12961-9
The paper
Reece, A. G., Reagan, A. J., Lix, K. L. M., Dodds, P. S., Danforth, C. M., & Langer, E. J. (2017). Forecasting the onset and course of mental illness with Twitter data. Scientific Reports. https://doi.org/10.1038/s41598-017-12961-9

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.