{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T20:04:52Z","timestamp":1780344292873,"version":"3.54.1"},"reference-count":47,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2018,9,21]],"date-time":"2018-09-21T00:00:00Z","timestamp":1537488000000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>We comment on Eichstaedt et\u00a0al.\u2019s (2015a) claim to have shown that language patterns among Twitter users, aggregated at the level of US counties, predicted county-level mortality rates from atherosclerotic heart disease (AHD), with \u201cnegative\u201d language being associated with higher rates of death from AHD and \u201cpositive\u201d language associated with lower rates. First, we examine some of Eichstaedt et al.\u2019s apparent assumptions about the nature of AHD, as well as some issues related to the secondary analysis of online data and to considering counties as communities. Next, using the data files supplied by Eichstaedt et al., we reproduce their regression- and correlation-based models, substituting mortality from an alternative cause of death\u2014namely, suicide\u2014as the outcome variable, and observe that the purported associations between \u201cnegative\u201d and \u201cpositive\u201d language and mortality are reversed when suicide is used as the outcome variable. We identify numerous other conceptual and methodological limitations that call into question the robustness and generalizability of Eichstaedt et al.\u2019s claims, even when these are based on the results of their ridge regression\/machine learning model. We conclude that there is no good evidence that analyzing Twitter data in bulk in this way can add anything useful to our ability to understand geographical variation in AHD mortality rates.<\/jats:p>","DOI":"10.7717\/peerj.5656","type":"journal-article","created":{"date-parts":[[2018,9,21]],"date-time":"2018-09-21T04:49:01Z","timestamp":1537505341000},"page":"e5656","source":"Crossref","is-referenced-by-count":5,"title":["Does Twitter language reliably predict heart disease? A commentary on Eichstaedt et al. (2015a)"],"prefix":"10.7717","volume":"6","author":[{"given":"Nicholas J.L.","family":"Brown","sequence":"first","affiliation":[{"name":"University Medical Center, University of Groningen, Groningen, Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James C.","family":"Coyne","sequence":"additional","affiliation":[{"name":"University Medical Center, University of Groningen, Groningen, Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2018,9,21]]},"reference":[{"key":"10.7717\/peerj.5656\/ref-1","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1017\/S1049096512000017","article-title":"The Big Sort that wasn\u2019t: A skeptical reexamination","volume":"45","author":"Abrams","year":"2012","journal-title":"Political Science & Politics"},{"key":"10.7717\/peerj.5656\/ref-2","article-title":"Language on Twitter tracks rates of coronary heart disease","author":"Association\u00a0for Psychological\u00a0Science","year":"2015"},{"key":"10.7717\/peerj.5656\/ref-3","volume-title":"Counties as service delivery agents: changing expectations and roles","author":"Benton","year":"2002"},{"key":"10.7717\/peerj.5656\/ref-4","doi-asserted-by":"publisher","first-page":"386","DOI":"10.2307\/2095230","article-title":"An introduction to sample selection bias in sociological data","volume":"48","author":"Berk","year":"1983","journal-title":"American Sociological Review"},{"key":"10.7717\/peerj.5656\/ref-5","first-page":"37","article-title":"Using ZIP\u00ae codes as geocodes in cancer research","volume-title":"Geocoding health data: the use of geographic codes in cancer prevention and control, research, and practice","author":"Beyer","year":"2008"},{"key":"10.7717\/peerj.5656\/ref-6","doi-asserted-by":"publisher","first-page":"712","DOI":"10.1038\/nrcardio.2009.163","article-title":"Socioeconomic status and cardiovascular disease: risks and implications for care","volume":"6","author":"Clark","year":"2009","journal-title":"Nature Reviews Cardiology"},{"key":"10.7717\/peerj.5656\/ref-7","volume-title":"Understanding psychology as a science","author":"Dienes","year":"2008"},{"key":"10.7717\/peerj.5656\/ref-8","doi-asserted-by":"crossref","DOI":"10.31234\/osf.io\/p75ku","article-title":"More evidence that Twitter language predicts heart disease: a response and replication","author":"Eichstaedt","year":"2018"},{"key":"10.7717\/peerj.5656\/ref-9","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1177\/0956797614557867","article-title":"Psychological language on Twitter predicts county-level heart disease mortality","volume":"26","author":"Eichstaedt","year":"2015","journal-title":"Psychological Science"},{"key":"10.7717\/peerj.5656\/ref-10","article-title":"Supplemental method","author":"Eichstaedt","year":"2015"},{"key":"10.7717\/peerj.5656\/ref-11","article-title":"Supplemental tables","author":"Eichstaedt","year":"2015"},{"key":"10.7717\/peerj.5656\/ref-12","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1037\/bul0000084","article-title":"Risk factors for suicidal thoughts and behaviors: a meta-analysis of 50 years of research","volume":"143","author":"Franklin","year":"2017","journal-title":"Psychological Bulletin"},{"key":"10.7717\/peerj.5656\/ref-13","doi-asserted-by":"publisher","first-page":"1286","DOI":"10.1001\/jama.1959.03000290012005","article-title":"Association of specific overt behaviour pattern with blood and cardiovascular findings","volume":"169","author":"Friedman","year":"1959","journal-title":"Journal of the American Medical Association"},{"key":"10.7717\/peerj.5656\/ref-14","doi-asserted-by":"publisher","first-page":"569654","DOI":"10.1155\/2012\/569654","article-title":"Hyperglycemia and endothelial dysfunction in atherosclerosis: lessons from type 1 diabetes","volume":"2012","author":"Funk","year":"2012","journal-title":"International Journal of Vascular Medicine"},{"key":"10.7717\/peerj.5656\/ref-15","first-page":"209","article-title":"Alternative spelling and censorship: the treatment of profanities in virtual communities","volume-title":"Aspects of linguistic impoliteness","author":"Goudet","year":"2013"},{"key":"10.7717\/peerj.5656\/ref-16","doi-asserted-by":"crossref","DOI":"10.4135\/9781452240152","volume-title":"Political encyclopedia of US states and regions","author":"Haider-Merkel","year":"2009"},{"key":"10.7717\/peerj.5656\/ref-17","article-title":"The changing profile of autopsied deaths in the United States, 1972\u20132007 (NCHS data brief no. 67)","author":"Hoyert","year":"2011"},{"key":"10.7717\/peerj.5656\/ref-18","article-title":"Tweets can better predict heart disease rates than income, smoking and diabetes, study finds. The Washington Post","author":"Izadi","year":"2015"},{"key":"10.7717\/peerj.5656\/ref-19","article-title":"Happier tweets, healthier communities. Pacific Standard","author":"Jacobs","year":"2015"},{"issue":"9","key":"10.7717\/peerj.5656\/ref-20","doi-asserted-by":"publisher","first-page":"e0180080","DOI":"10.1371\/journal.pone.0180080","article-title":"Putting the methodological brakes on claims to measure national happiness through Twitter: methodological limitations in social media analytics","volume":"12","author":"Jensen","year":"2017","journal-title":"PLOS ONE"},{"key":"10.7717\/peerj.5656\/ref-21","article-title":"Counties as communities. Washington monthly","author":"Kilgore","year":"2012"},{"key":"10.7717\/peerj.5656\/ref-22","doi-asserted-by":"publisher","first-page":"267","DOI":"10.1055\/s-2002-35401","article-title":"Systematic review of prospective cohort studies of psychosocial factors in the aetiology and prognosis of coronary heart disease","volume":"2","author":"Kuper","year":"2002","journal-title":"Seminars in Vascular Medicine"},{"key":"10.7717\/peerj.5656\/ref-23","doi-asserted-by":"publisher","first-page":"278","DOI":"10.1177\/1745691614528520","article-title":"Sailing from the seas of chaos into the corridor of stability: practical recommendations to increase the informational value of studies","volume":"9","author":"Lakens","year":"2014","journal-title":"Perspectives on Psychological Science"},{"key":"10.7717\/peerj.5656\/ref-24","doi-asserted-by":"publisher","first-page":"1203","DOI":"10.1126\/science.1248506","article-title":"The parable of Google Flu: traps in big data analysis","volume":"343","author":"Lazer","year":"2014","journal-title":"Science"},{"key":"10.7717\/peerj.5656\/ref-25","volume-title":"Detroit: an American autopsy","author":"LeDuff","year":"2014"},{"key":"10.7717\/peerj.5656\/ref-26","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1136\/jech.2005.034017","article-title":"Socioeconomic gradients in the prevalence of cardiovascular disease in Scotland: the roles of composition and context","volume":"59","author":"Leyland","year":"2005","journal-title":"Journal of Epidemiology & Community Health"},{"key":"10.7717\/peerj.5656\/ref-27","doi-asserted-by":"publisher","first-page":"862","DOI":"10.1016\/j.jclinepi.2005.11.021","article-title":"Clinicians didn\u2019t reliably distinguish between different causes of cardiac death using case histories","volume":"59","author":"Mant","year":"2006","journal-title":"Journal of Clinical Epidemiology"},{"key":"10.7717\/peerj.5656\/ref-28","first-page":"677","article-title":"Death certification and doctors\u2019 dilemmas: a qualitative study of GPs\u2019 perspectives","volume":"55","author":"McAllum","year":"2005","journal-title":"The British Journal of General Practice"},{"key":"10.7717\/peerj.5656\/ref-29","doi-asserted-by":"publisher","first-page":"195","DOI":"10.2466\/pr0.1990.66.1.195","article-title":"Why summaries of research on psychological theories are often uninterpretable","volume":"66","author":"Meehl","year":"1990","journal-title":"Psychological Reports"},{"key":"10.7717\/peerj.5656\/ref-30","doi-asserted-by":"publisher","first-page":"794","DOI":"10.1001\/jama.1996.03530340058030","article-title":"Accuracy of death certificate completion: the need for formalized physician training","volume":"275","author":"Messite","year":"1996","journal-title":"Journal of the American Medical Association"},{"issue":"5","key":"10.7717\/peerj.5656\/ref-31","doi-asserted-by":"publisher","first-page":"e64417","DOI":"10.1371\/journal.pone.0064417","article-title":"The geography of happiness: connecting Twitter sentiment and expression, demographics, and objective characteristics of place","volume":"8","author":"Mitchell","year":"2013","journal-title":"PLOS ONE"},{"key":"10.7717\/peerj.5656\/ref-32","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1016\/S2215-0366(14)70222-6","article-title":"The psychology of suicidal behaviour","volume":"1","author":"O\u2019Connor","year":"2014","journal-title":"The Lancet Psychiatry"},{"key":"10.7717\/peerj.5656\/ref-33","doi-asserted-by":"publisher","first-page":"906","DOI":"10.1007\/s00127-004-0873-y","article-title":"Neuroticism: a non-informative marker of vulnerability to psychopathology","volume":"39","author":"Ormel","year":"2004","journal-title":"Social Psychiatry and Psychiatric Epidemiology"},{"key":"10.7717\/peerj.5656\/ref-34","article-title":"Are you a type A, B, or D personality? Psychology today","author":"Riggio","year":"2014"},{"key":"10.7717\/peerj.5656\/ref-35","doi-asserted-by":"publisher","first-page":"853","DOI":"10.1136\/jech.2006.050740","article-title":"Toward the next generation of research into small area effects on health: a synthesis of multilevel investigations published since 1998","volume":"61","author":"Riva","year":"2007","journal-title":"Journal of Epidemiology & Community Health"},{"issue":"8","key":"10.7717\/peerj.5656\/ref-36","doi-asserted-by":"publisher","first-page":"e0183551","DOI":"10.1371\/journal.pone.0183551","article-title":"The unbearable emptiness of tweeting\u2014about journal articles","volume":"12","author":"Robinson-Garcia","year":"2017","journal-title":"PLOS ONE"},{"key":"10.7717\/peerj.5656\/ref-37","doi-asserted-by":"publisher","first-page":"1976","DOI":"10.1001\/jama.2017.4150","article-title":"Trends and patterns of geographic variation in cardiovascular mortality among US counties, 1980\u20132014","volume":"317","author":"Roth","year":"2017","journal-title":"Journal of the American Medical Association"},{"key":"10.7717\/peerj.5656\/ref-38","article-title":"Can angry tweets predict heart-disease rates? New York Magazine","author":"Singal","year":"2015"},{"key":"10.7717\/peerj.5656\/ref-39","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1111\/j.1467-8721.2006.00441.x","article-title":"Personality as risk and resilience in physical health","volume":"15","author":"Smith","year":"2006","journal-title":"Current Directions in Psychological Science"},{"key":"10.7717\/peerj.5656\/ref-40","doi-asserted-by":"publisher","first-page":"528","DOI":"10.1097\/PSY.0b013e31822e095e","article-title":"Toward a more systematic, cumulative, and applicable science of personality and health: lessons from type D personality","volume":"73","author":"Smith","year":"2011","journal-title":"Psychosomatic Medicine"},{"key":"10.7717\/peerj.5656\/ref-41","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1016\/j.ijmedinf.2006.11.005","article-title":"Reliability of diagnoses coding with ICD-10","volume":"75","author":"Stausberg","year":"2006","journal-title":"International Journal of Medical Informatics"},{"key":"10.7717\/peerj.5656\/ref-42","doi-asserted-by":"publisher","first-page":"342","DOI":"10.1093\/ije\/dyn359","article-title":"Revisiting Robinson: the perils of individualistic and ecologic fallacy","volume":"38","author":"Subramanian","year":"2008","journal-title":"International Journal of Epidemiology"},{"key":"10.7717\/peerj.5656\/ref-43","article-title":"Geographical mobility: 2009 to 2010","author":"US\u00a0Census\u00a0Bureau","year":"2011"},{"key":"10.7717\/peerj.5656\/ref-44","doi-asserted-by":"crossref","DOI":"10.1609\/icwsm.v11i1.14871","article-title":"Online human-bot interactions: detection, estimation, and characterization","author":"Varol","year":"2017"},{"key":"10.7717\/peerj.5656\/ref-45","doi-asserted-by":"publisher","first-page":"1084","DOI":"10.1136\/jech-2017-209535","article-title":"Psychological distress and ischaemic heart disease: cause or consequence? Evidence from a large prospective cohort study","volume":"71","author":"Welsh","year":"2017","journal-title":"Journal of Epidemiological and Community Health"},{"key":"10.7717\/peerj.5656\/ref-46","doi-asserted-by":"publisher","first-page":"e0152719","DOI":"10.1371\/journal.pone.0152719","article-title":"Statistically controlling for confounding constructs is harder than you think","volume":"11","author":"Westfall","year":"2016","journal-title":"PLOS ONE"},{"key":"10.7717\/peerj.5656\/ref-47","article-title":"Type A and B personalities. The Guardian, March 7","author":"Wilson","year":"2009"}],"container-title":["PeerJ"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/5656.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/5656.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/5656.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/5656.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,5]],"date-time":"2023-09-05T04:27:17Z","timestamp":1693888037000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/5656"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,9,21]]},"references-count":47,"alternative-id":["10.7717\/peerj.5656"],"URL":"https:\/\/doi.org\/10.7717\/peerj.5656","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{"has-review":[{"id-type":"doi","id":"10.7287\/peerj.5656v0.1\/reviews\/4","asserted-by":"object"},{"id-type":"doi","id":"10.7287\/peerj.5656v0.1\/reviews\/2","asserted-by":"object"},{"id-type":"doi","id":"10.7287\/peerj.5656v0.1\/reviews\/3","asserted-by":"object"},{"id-type":"doi","id":"10.7287\/peerj.5656v0.1\/reviews\/1","asserted-by":"object"},{"id-type":"doi","id":"10.7287\/peerj.5656v0.2\/reviews\/4","asserted-by":"object"},{"id-type":"doi","id":"10.7287\/peerj.5656v0.2\/reviews\/3","asserted-by":"object"},{"id-type":"doi","id":"10.7287\/peerj.5656v0.2\/reviews\/1","asserted-by":"object"}]},"ISSN":["2167-8359"],"issn-type":[{"value":"2167-8359","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,9,21]]},"article-number":"e5656"}}