Predicting state level suicide fatalities in the united states with realtime data and machine learning.
Journal
Npj mental health research
ISSN: 2731-4251
Titre abrégé: Npj Ment Health Res
Pays: England
ID NLM: 9918592488906676
Informations de publication
Date de publication:
16 Jan 2024
16 Jan 2024
Historique:
received:
02
05
2023
accepted:
20
11
2023
medline:
13
4
2024
pubmed:
13
4
2024
entrez:
12
4
2024
Statut:
epublish
Résumé
Digital trace data and machine learning techniques are increasingly being adopted to predict suicide-related outcomes at the individual level; however, there is also considerable public health need for timely data about suicide trends at the population level. Although significant geographic variation in suicide rates exist by state within the United States, national systems for reporting state suicide trends typically lag by one or more years. We developed and validated a deep learning based approach to utilize real-time, state-level online (Mental Health America web-based depression screenings; Google and YouTube Search Trends), social media (Twitter), and health administrative data (National Syndromic Surveillance Program emergency department visits) to estimate weekly suicide counts in four participating states. Specifically, per state, we built a long short-term memory (LSTM) neural network model to combine signals from the real-time data sources and compared predicted values of suicide deaths from our model to observed values in the same state. Our LSTM model produced accurate estimates of state-specific suicide rates in all four states (percentage error in suicide rate of -2.768% for Utah, -2.823% for Louisiana, -3.449% for New York, and -5.323% for Colorado). Furthermore, our deep learning based approach outperformed current gold-standard baseline autoregressive models that use historical death data alone. We demonstrate an approach to incorporate signals from multiple proxy real-time data sources that can potentially provide more timely estimates of suicide trends at the state level. Timely suicide data at the state level has the potential to improve suicide prevention planning and response tailored to the needs of specific geographic communities.
Identifiants
pubmed: 38609512
doi: 10.1038/s44184-023-00045-8
pii: 10.1038/s44184-023-00045-8
doi:
Types de publication
Journal Article
Langues
eng
Pagination
3Informations de copyright
© 2024. The Author(s).
Références
National vital statistics system, underlying cause of death 1999-2019 on cdc wonder online database. http://wonder.cdc.gov/ucd-icd10.html (2020).
Centers for Disease Control, Prevention. et al. Regional variations in suicide rates–united states, 1990-1994. Morb. Mortal. Wkly. Rep. 46, 789–793 (1997).
Walker, J. T. County level suicide rates and social integration: urbanicity and its role in the relationship. Sociol. Spectr. 29, 101–135 (2008).
doi: 10.1080/02732170802480568
Durkheim E. Suicide: A Study In Sociology. (Routledge, 2005).
Baller R. D. & Richardson K. K. Social integration, imitation, and the geographic patterning of suicide. Am. Sociol. Rev. 67, 873–888 (2002).
Barkan, S. E., Rocque, M. & Houle, J. State and regional suicide rates: a new look at an old puzzle. Sociol. Perspect. 56, 287–297 (2013).
doi: 10.1525/sop.2013.56.2.287
Kunce, M. & Anderson, A. L. The impact of socioeconomic factors on state suicide rates: a methodological note. Urban Stud. 39, 155–162 (2002).
doi: 10.1080/00420980220099131
Giles-Sims, J. & Lockhart, C. Explaining cross-state differences in elderly suicide rates and identifying state-level public policy responses that reduce rates. Suicide Life Threat. Behav. 36, 694–708 (2006).
doi: 10.1521/suli.2006.36.6.694
pubmed: 17250474
Ivey-Stephenson, A. Z., Crosby, A. E., Jack, S. P. D., Haileyesus, T. & Kresnow-Sedacca, M. J. Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death—United States, 2001–2015. MMWR Surveill. Summ. 66, 1 (2017).
doi: 10.15585/mmwr.ss6618a1
pubmed: 28981481
pmcid: 5829833
Stone, D. M. et al. Vital signs: Trends in state suicide rates—United States, 1999–2016 and circumstances contributing to suicide—27 states, 2015. Morb. Mortal. Wkly Rep. 67, 617–624 (2018).
Spencer M. & Ahmad F. Timeliness of death certificate data for mortality surveillance and provisional estimates. Technical report, Division of Vital Statistics, National Center for Health Statistics (2016).
Fatal injury reports, national, regional and state, 1981–2019 on cdc wisqars online database. https://webappa.cdc.gov/sasweb/ncipc/mortrate.html (2019).
Ikeda, R. et al. Improving national data systems for surveillance of suicide-related events. Am. J. Prevent. Med. 47, S122 (2014).
doi: 10.1016/j.amepre.2014.05.026
Ahmad, F. B., Dokpesi, P., Escobedo, L. & Rossen L. Timeliness of death certificate data by sex, age, and geography. (National Center for Health Statistics, CDC, 2020).
Spencer, M. R. & Ahmad, F. Timeliness of death certificate data for mortality surveillance and provisional estimates. Technical report, National Center for Health Statistics, vol. 1 (2017).
Ramchand, R. et al. Prioritizing improved data and surveillance for suicide in the united states in response to covid-19. Am. J. Public Health 111, S84–S88 (2021).
Barros, J. et al. The validity of google trends search volumes for behavioral forecasting of national suicide rates in ireland. Int. J. Environ. Res. Public Health 16, 3201 (2019).
doi: 10.3390/ijerph16173201
pubmed: 31480718
pmcid: 6747463
Jashinsky, J. et al. Tracking suicide risk factors through twitter in the US. Crisis 35, 51–59 (2014).
Homan, C. et al. Toward macro-insights for suicide prevention: analyzing fine-grained distress at scale. In Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality, pp. 107–117 (2014).
Zhang, L. et al. Using linguistic features to estimate suicide probability of chinese microblog users. In International Conference on Human Centered Computing, pp. 549–559. (Springer, 2014).
O’dea, B. et al. Detecting suicidality on twitter. Internet Interv. 2, 183–188 (2015).
doi: 10.1016/j.invent.2015.03.005
Burnap, P., Colombo, W. & Scourfield J. Machine classification and analysis of suicide-related communication on twitter. In Proceedings of the 26th ACM conference on hypertext & social media, pp. 75–84, vol. 8 (2015).
De Choudhury, M. et al. Discovering shifts to suicidal ideation from mental health content in social media. In Proceedings of the 2016 CHI conference on human factors in computing systems, pp. 2098–2110 (2016).
Braithwaite, S. R., Giraud-Carrier, C., West, J., Barnes, M. D. & Hanson, C. L. Validating machine learning algorithms for twitter data against established measures of suicidality. JMIR Ment. health 3, e21 (2016).
doi: 10.2196/mental.4822
pubmed: 27185366
pmcid: 4886102
Vioules, M. J., Moulahi, B., Azé, J. & Bringay, S. Detection of suicide-related posts in twitter data streams. IBM J. Res. Dev. 62, 7–1 (2018).
doi: 10.1147/JRD.2017.2768678
Du, J. et al. Extracting psychiatric stressors for suicide from social media using deep learning. BMC Med. Inform. Decis. Mak. 18, 77–87 (2018).
doi: 10.1186/s12911-018-0632-8
Bryan, C. J. et al. Predictors of emerging suicide death among military personnel on social media networks. Suicide Life Threat. Behav. 48, 413–430 (2018).
doi: 10.1111/sltb.12370
pubmed: 28752655
Robinson, J. et al. Social media and suicide prevention: a systematic review. Early interv. Psychiatry 10, 103–121 (2016).
doi: 10.1111/eip.12229
pubmed: 25702826
Won, H.-H. et al. Predicting national suicide numbers with social media data. PloS ONE 8, e61809 (2013).
doi: 10.1371/journal.pone.0061809
pubmed: 23630615
pmcid: 3632511
Choi, D. et al. Development of a machine learning model using multiple, heterogeneous data sources to estimate weekly us suicide fatalities. JAMA Netw. Open 3, e2030932 (2020).
doi: 10.1001/jamanetworkopen.2020.30932
pubmed: 33355678
pmcid: 7758810
Hargittai, E. Potential biases in big data: Omitted voices on social media. Soc. Sci. Comput. Rev. 38, 089443931878832 (2018).
Tufekci, Z. Big questions for social media big data: representativeness, validity and other methodological pitfalls. Proceedings of the 8th International Conference on Weblogs and Social Media, ICWSM 2014, vol. 3 (2014).
De Choudhury, M., Morris, M. & White, R. Seeking and sharing health information online: comparing search engines and social media. Conference on Human Factors in Computing Systems—Proceedings, vol. 4 (2014).
Birnbaum, M. L., Rizvi, A. F., Correll, C. U., Kane, J. M. & Confino, J. Role of social media and the i nternet in pathways to care for adolescents and young adults with psychotic disorders and non-psychotic mood disorders. Early Interv. Psychiatry 11, 290–295 (2017).
doi: 10.1111/eip.12237
pubmed: 25808317
Thackeray, R., Crookston, B. T. & West, J. H. Correlates of health-related social media use among adults. J. Med. Internet Res. 15, e21 (2013).
doi: 10.2196/jmir.2297
pubmed: 23367505
pmcid: 3636287
Zhao, Y. & Zhang, J. Consumer health information seeking in social media: a literature review. Health Inf. Libr. J. 34, 268–283 (2017).
doi: 10.1111/hir.12192
Tran, U. et al. Low validity of google trends for behavioral forecasting of national suicide rates. PLoS ONE 12, e0183149 (2017).
doi: 10.1371/journal.pone.0183149
pubmed: 28813490
pmcid: 5558943
Jagodic, H. K., Agius, M. & Pregelj, P. Inter-regional variations in suicide rates. Psychiatr. Danub 24, S82–S85 (2012).
pubmed: 22945194
Planalp, C. & Hest, R. Suicide rates on the rise: State trends and variation in suicide deaths from 2000 to 2017. Technical report, Robert Wood Johnson Foundation, vol. 10 (2019).
Kessler, R., Mickelson, K. & Williams, D. The prevalence, distribution, and mental health correlates of perceived discrimination in the united states. J. Health Soc. Behav. 40, 208–230 (1999).
doi: 10.2307/2676349
pubmed: 10513145
Reeves, A. et al. Increase in state suicide rates in the usa during economic recession. Lancet 380, 11 (2012).
doi: 10.1016/S0140-6736(12)61910-2
Chou, W.-Y., Hunt, Y., Beckjord, E., Moser, R. & Hesse, B. Social media use in the united states: Implications for health communication. J. Med. Internet Res. 11, e48 (2009).
doi: 10.2196/jmir.1249
pubmed: 19945947
pmcid: 2802563
Hecht, B., Hong, L., Suh, B. & Ed, H Chi. Tweets from justin bieber’s heart: the dynamics of the location field in user profiles. In Proceedings of the SIGCHI conference on human factors in computing systems, pp. 237–246 (2011).
Saha, K. et al. A social media study on demographic differences in perceived job satisfaction. Proc. ACM Hum. Comput. Interact. 5, 1–29 (2021).
doi: 10.1145/3449241
Roesslein, J. Tweepy: Twitter for python! https://github.com/tweepy/tweepy (2020).
HERE Developer API. https://developer.here.com (2017).
OpenStreetMap contributors. Planet dump retrieved from https://planet.osm.org . https://www.openstreetmap.org (2017).
Topol, E. J. High-performance medicine: the convergence of human and artificial intelligence. Nat. Med. 25, 44–56 (2019).
doi: 10.1038/s41591-018-0300-7
pubmed: 30617339
Siami-Namini, S., Tavakoli N. & Namin, A. S. A comparison of arima and lstm in forecasting time series. In 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 1394–1401. (IEEE, 2018).
Xingjian, S. et al. Convolutional lstm network: a machine learning approach for precipitation nowcasting. In Advances in neural information processing systems, pp. 802–810 (2015).
Chimmula, V. K. R. & Zhang, L. Time series forecasting of covid-19 transmission in canada using lstm networks. Chaos Solitons Fractals 135, 109864 (2020).
doi: 10.1016/j.chaos.2020.109864
pubmed: 32390691
pmcid: 7205623
Salganik, M. J. Bit by bit: Social Research in the Digital Age. (Princeton University Press, 2019.
Lazer, D., Kennedy, R., King, G. & Vespignani, A. The parable of google flu: traps in big data analysis. Science 343, 1203–1205 (2014).
doi: 10.1126/science.1248506
pubmed: 24626916
Abdi, H. & Williams, L. J. Principal component analysis. Wiley Interdiscip. Rev. 2, 433–459 (2010).
doi: 10.1002/wics.101
Tondo, L. et al. Suicide rates in relation to health care access in the united states: an ecological study. J. Clin. Psychiatry 67, 517–523 (2006).
doi: 10.4088/JCP.v67n0402
pubmed: 16669716
Investing in America’s Health: A state-by-state look at public health funding and key health facts. Technical report, Trust for America’s Health, vol. 4 (2016).
Duggan, M. & Smith A. Cell iNternet Use 2013 (2013).
Auxier, B. & Anderson, M. Social media use in 2021. (Pew Research Center, 2021).
Shrira, I. & Christenfeld, N. Disentangling the person and the place as explanations for regional differences in suicide. Suicide Life Threat. Behav. 40, 287–297 (2010).
doi: 10.1521/suli.2010.40.3.287
pubmed: 20560750
Marian, E. et al. Elevated suicide rates at high altitude: sociodemo-graphic and health issues may be to blame. Suicide Life Threat. Behav. 41, 562–573 (2011).
doi: 10.1111/j.1943-278X.2011.00054.x
CDC. Suicide prevention resource for action: a compilation of the best available evidence. (National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, 2022).
Stone, D. M. et al. Preventing Suicide: A Technical Package of Policies, Programs, and Practice (2017).