Machine learning discovery of longitudinal patterns of depression and suicidal ideation

English version

Autoři: Jue Gong ^aff001; Gregory E. Simon ^aff002; Shan Liu ^aff001
Působiště autorů: Department of Industrial and Systems Engineering, University of Washington, Seattle, WA, United States of America ^aff001; Kaiser Permanente Washington Health Research Institute, Seattle, WA, United States of America ^aff002; Psychiatry and Behavioral Sciences, University of Washington, Seattle, WA, United States of America ^aff003
Vyšlo v časopise: PLoS ONE 14(9)
Kategorie: Research Article
prolekare.web.journal.doi_sk: https://doi.org/10.1371/journal.pone.0222665

Souhrn

Background and aim

Depression is often accompanied by thoughts of self-harm, which are a strong predictor of subsequent suicide attempt and suicide death. Few empirical data are available regarding the temporal correlation between depression symptoms and suicidal ideation. We investigated the anecdotal concern that suicidal ideation may increase during a period of depression improvement.

Data

Longitudinal Patient Health Questionnaire (PHQ)-9 is a questionnaire of 9 multiple-choice questions to assess the frequency of depressive symptoms within the previous two weeks. We analyzed a chronic depression treatment population’s electronic health record (EHR) data, containing 610 patients’ longitudinal PHQ-9 scores (62% age 45 and older; 68% female) within 40 weeks.

Methods

The irregular and sparse EHR data were transformed into continuous trajectories using Gaussian process regression. We first estimated the correlations between the symptoms (total score of the first 8 questions; PHQ-8) and suicide ideation (9th question score; Item 9) using the cross-correlation function. We then used an artificial neural network (ANN) to discover subtypes of depression patterns from the fitted depression trajectories. In addition, we conducted a separate analysis using the unfitted raw PHQ scores to examine PHQ-8’s and Item 9’s pattern changes.

Results

Results showed that the majority of patients’ PHQ-8 and Item 9 scores displayed strong temporal correlations. We found five patterns in the PHQ-8 and the Item 9 trajectories. We also found 8% - 13% of the patients have experienced an increase in suicidal ideation during the improvement of their PHQ-8. Using a trajectory-based method for subtype pattern detection in depression progression, we provided a better understanding of temporal correlations between depression symptoms over time.

Klíčová slova:

Biology and life sciences – Computational biology – Research and analysis methods – Neuroscience – Computer and information sciences – Medicine and health sciences – Pharmacology – Research design – Drugs – Survey research – Mental health and psychiatry – Neural networks – Mood disorders – Depression – Questionnaires – Artificial intelligence – Suicide – Self harm – Antidepressants – Computational neuroscience – Artificial neural networks

Zdroje

1. Valuck RJ, Anderson HO, Libby AM, Brandt E, Bryan C, Allen RR, et al. (2012) Enhancing electronic health record measurement of depression severity and suicide ideation: a Distributed Ambulatory Research in Therapeutics Network (DARTNet) study. J Am Board Fam Med 25: 582–593. doi: 10.3122/jabfm.2012.05.110053 22956694

2. World Health Organization website on depression. Accessed at https://www.who.int/news-room/fact-sheets/detail/depression, March 22 2018.

3. Simon GE, Johnson E, Lawrence JM, Rossom RC, Ahmedani B, Lynch FL, et al. (2018) Predicting suicide attempts and suicide deaths following outpatient visits using electronic health records. American Journal of Psychiatry 175: 951–960. doi: 10.1176/appi.ajp.2018.17101167 29792051

4. Simon GE, Rutter CM, Peterson D, Oliver M, Whiteside U, Operskalski B, et al. (2013) Does response on the PHQ-9 Depression Questionnaire predict subsequent suicide attempt or suicide death? Psychiatric Services 64: 1195–1202. doi: 10.1176/appi.ps.201200587 24036589

5. Ribeiro J, Franklin J, Fox KR, Bentley K, Kleiman EM, Chang B, et al. (2016) Self-injurious thoughts and behaviors as risk factors for future suicide ideation, attempts, and death: a meta-analysis of longitudinal studies. Psychological Medicine 46: 225–236. doi: 10.1017/S0033291715001804 26370729

6. Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, et al. (2017) Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin 143: 187. doi: 10.1037/bul0000084 27841450

7. Mittal V, Brown WA, Shorter E (2009) Are patients with depression at heightened risk of suicide as they begin to recover? Psychiatric services 60: 384–386. doi: 10.1176/appi.ps.60.3.384 19252052

8. Ellis TE, Green KL, Allen JG, Jobes DA, Nadorff MR (2012) Collaborative Assessment and Management of Suicidality in an Inpatient Setting: Results of a Pilot Study. Psychotherapy (Chicago, Ill) 49: 72–80.

9. Gunn J, Elliott P, Densley K, Middleton A, Ambresin G, Dowrick C, et al. (2013) A trajectory-based approach to understand the factors associated with persistent depressive symptoms in primary care. Journal of affective disorders 148: 338–346. doi: 10.1016/j.jad.2012.12.021 23375580

10. Kroenke K, Spitzer RL (2002) The PHQ-9: a new depression diagnostic and severity measure. Psychiatric annals 32: 509–515.

11. Bayley KB, Belnap T, Savitz L, Masica AL, Shah N, Fleming NS (2013) Challenges in using electronic health record data for CER: experience of 4 learning organizations and solutions applied. Medical care 51: S80–S86. doi: 10.1097/MLR.0b013e31829b1d48 23774512

12. Kahn MG, Raebel MA, Glanz JM, Riedlinger K, Steiner JF (2012) A pragmatic framework for single-site and multisite data quality assessment in electronic health record-based clinical research. Medical care 50.

13. Musliner KL, Munk-Olsen T, Eaton WW, Zandi PP (2016) Heterogeneity in long-term trajectories of depressive symptoms: Patterns, predictors and outcomes. Journal of affective disorders 192: 199–211. doi: 10.1016/j.jad.2015.12.030 26745437

14. Lloyd S (1982) Least squares quantization in PCM. IEEE transactions on information theory 28: 129–137.

15. Zhang Z, Murtagh F, Van Poucke S, Lin S, Lan P (2017) Hierarchical cluster analysis in clinical research with heterogeneous study population: highlighting its visualization with R. Ann Transl Med 5: 75. doi: 10.21037/atm.2017.02.05 28275620

16. Twisk J, Hoekstra T (2012) Classifying developmental trajectories over time should be done with great caution: a comparison between methods. J Clin Epidemiol 65: 1078–1087. doi: 10.1016/j.jclinepi.2012.04.010 22818946

17. Kaplan D, Sage Publications. (2004) The Sage handbook of quantitative methodology for the social sciences. Thousand Oaks, Calif.: Sage. xiii, 511 pages p.

18. Berlin KS, Parra GR, Williams NA (2014) An introduction to latent variable mixture modeling (part 2): longitudinal latent class growth analysis and growth mixture models. J Pediatr Psychol 39: 188–203. doi: 10.1093/jpepsy/jst085 24277770

19. Lin Y, Liu K, Byon E, Qian X, Liu S, Huang S (2018) A Collaborative Learning Framework for Estimating Many Individualized Regression Models in a Heterogeneous Population. IEEE Transactions on Reliability 67: 328–341.

20. Lin Y, Huang S, Simon GE, Liu S (2016) Analysis of depression trajectory patterns using collaborative learning. Mathematical Biosciences 282: 191–203. doi: 10.1016/j.mbs.2016.10.008 27789353

21. Lasko TA, Denny JC, Levy MA (2013) Computational phenotype discovery using unsupervised feature learning over noisy, sparse, and irregular clinical data. PloS one 8: e66341. doi: 10.1371/journal.pone.0066341 23826094

22. Jacobson O, Dalianis H (2016) Applying deep learning on electronic health records in Swedish to predict healthcare-associated infections. ACL 2016: 191.

23. Van Loo HM, De Jonge P, Romeijn J-W, Kessler RC, Schoevers RA (2012) Data-driven subtypes of major depressive disorder: a systematic review. BMC medicine 10: 156. doi: 10.1186/1741-7015-10-156 23210727

24. Lin Y, Liu S, Huang S (2018) Selective sensing of a heterogeneous population of units with dynamic health conditions. IISE Transactions 50: 1076–1088.

25. Rehfeld K, Marwan N, Heitzig J, Kurths J (2011) Comparison of correlation analysis techniques for irregularly sampled time series. Nonlinear Processes in Geophysics 18: 389–404.

26. Goodfellow I, Bengio Y, Courville A (2016) Deep learning: MIT Press.

27. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. (2011) Scikit-learn: Machine learning in Python. JMLR 12: 2825–2830.

28. Team TTD (2016) Theano: A Python framework for fast computation of mathematical expressions. arXiv e-prints abs/1605.02688.

29. Swingler K (1996) Applying neural networks: a practical guide: Morgan Kaufmann.

30. Lvd Maaten, Hinton G (2008) Visualizing data using t-SNE. Journal of Machine Learning Research 9: 2579–2605.

31. Snippe E, Bos EH, van der Ploeg KM, Sanderman R, Fleer J, Schroevers MJ (2015) Time-series analysis of daily changes in mindfulness, repetitive thinking, and depressive symptoms during mindfulness-based treatment. Mindfulness 6: 1053–1062.

32. Anderson HD, Pace WD, Brandt E, Nielsen RD, Allen RR, Libby AM, et al. (2015) Monitoring suicidal patients in primary care using electronic health records. J Am Board Fam Med 28: 65–71. doi: 10.3122/jabfm.2015.01.140181 25567824

33. Zhong Q-Y, Karlson EW, Gelaye B, Finan S, Avillach P, Smoller JW, et al. (2018) Screening pregnant women for suicidal behavior in electronic medical records: diagnostic codes vs. clinical notes processed by natural language processing. BMC medical informatics and decision making 18: 30. doi: 10.1186/s12911-018-0617-7 29843698

34. Ahmedani BK, Peterson EL, Hu Y, Rossom RC, Lynch F, Lu CY, et al. (2017) Major Physical Health Conditions and Risk of Suicide. American Journal of Preventive Medicine.

35. Coupland C, Hill T, Morriss R, Arthur A, Moore M, Hippisley-Cox J (2015) Antidepressant use and risk of suicide and attempted suicide or self harm in people aged 20 to 64: cohort study using a primary care database. BMJ 350: h517. doi: 10.1136/bmj.h517 25693810

36. Dong G, Li J. Efficient mining of emerging patterns: discovering trends and differences. In KDD '99 Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Diego, CA, USA (pp. 43 e52). 1999.

37. Somaraki V, Broadbent D, Coenen F, Harding S. Finding Temporal Patterns in Noisy Longitudinal Data: A Study in Diabetic Retinopathy. In: Perner P. (eds) Advances in Data Mining. Applications and Theoretical Aspects. ICDM 2010. Lecture Notes in Computer Science, vol 6171. Springer, Berlin, Heidelberg; 2010.