{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T16:30:54Z","timestamp":1787589054348,"version":"build-2736575974"},"reference-count":37,"publisher":"American Medical Association (AMA)","issue":"8","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JAMA Netw Open"],"abstract":"<jats:sec id=\"ab-zoi260829-4\">\n                    <jats:title>Importance<\/jats:title>\n                    <jats:p>Patient portal messages are an increasingly central channel for patient-clinician communication, but there are inequities in how responsive care teams are to messages sent by patients from historically marginalized groups. The reasons driving these disparities are not well understood.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"ab-zoi260829-5\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To quantify factors that could potentially explain lower rates of care team response to messages sent by patients from historically marginalized groups.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"ab-zoi260829-6\">\n                    <jats:title>Design, Setting, and Participants<\/jats:title>\n                    <jats:p>This cross-sectional study used natural language processing to systematically extract message content and writing style features from portal message threads. Messages were from adult patients who sent a medical advice request message from January 1, 2021, to December 31, 2023, to a nontrainee physician, nurse practitioner, or physician assistant at a primary care practice. Data analysis was performed from May 2024 through June 2026.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"ab-zoi260829-7\">\n                    <jats:title>Exposure<\/jats:title>\n                    <jats:p>Message thread characteristics, including patient demographics, message content, and writing style.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"ab-zoi260829-8\">\n                    <jats:title>Main Outcomes and Measures<\/jats:title>\n                    <jats:p>Rates at which messages were responded to by care teams overall and by the message\u2019s originally intended target clinician specifically. Regression analysis quantified the degree to which message content and writing style potentially explain differences in care team response across patient groups.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"ab-zoi260829-9\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The sample included 3\u202f619\u202f390 patient message threads sent by 511\u202f020 patients. Of the threads, 4.0% were sent by Asian patients, 3.7% by Black patients, 5.9% by Hispanic patients, and 83.0% by White patients. Messages sent by patients from historically marginalized groups (eg, Black and Hispanic patients, patients with only a high school education, and Medicaid beneficiaries) received lower levels of care team response. Compared with White patients, Black patients had a 3.7\u2013percentage point (pp) (95% CI, \u22124.1 to \u22123.3 pp;\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      \u2009&amp;amp;lt;\u2009.001) lower response rate from the target clinician, representing an 11.6% reduction. Although message content was associated with target clinician response, it could not explain these differences. On the contrary, writing style accounted for 48.0% of the difference for Black patients, 34.9% for Hispanic patients, 60.5% for patients with only a high school education (relative to patients with a college degree), and 42.8% for Medicaid patients (relative to commercially insured patients).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"ab-zoi260829-10\">\n                    <jats:title>Conclusions and Relevance<\/jats:title>\n                    <jats:p>In this cross-sectional study of patient portal messages, writing style was a potential explanation for approximately half of the observed disparities in care team response, highlighting writing style as an important potential source of bias in message triage and a target for intervention.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1001\/jamanetworkopen.2026.30379","type":"journal-article","created":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T15:43:22Z","timestamp":1787586202000},"page":"e2630379","source":"Crossref","is-referenced-by-count":0,"title":["Care Team Response to Patient Portal Message Content and Writing Style"],"prefix":"10.1001","volume":"9","author":[{"given":"Mitchell","family":"Tang","sequence":"first","affiliation":[{"name":"Harvard Kenneth C. Griffin Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts"},{"name":"Department of Health Policy and Management, Mailman School of Public Health, Columbia University, New York, New York"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lisa S.","family":"Rotenstein","sequence":"additional","affiliation":[{"name":"Brigham and Women\u2019s Hospital, Boston, Massachusetts"},{"name":"Department of Medicine, University of California, San Francisco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ariel D.","family":"Stern","sequence":"additional","affiliation":[{"name":"Hasso Plattner Institute, University of Potsdam, Potsdam, Germany"},{"name":"Windreich Department of Artificial Intelligence & Human Health, Icahn School of Medicine at Mount Sinai, New York, New York"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rebecca G.","family":"Mishuris","sequence":"additional","affiliation":[{"name":"Brigham and Women\u2019s Hospital, Boston, Massachusetts"},{"name":"Mass General Brigham Digital, Somerville, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael L.","family":"Barnett","sequence":"additional","affiliation":[{"name":"Brigham and Women\u2019s Hospital, Boston, Massachusetts"},{"name":"Department of Health Services Policy and Practice, Brown University School of Public Health, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"10","published-online":{"date-parts":[[2026,8,24]]},"reference":[{"issue":"11","key":"zoi260829r2","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2022.44363","article-title":"Association of electronic health record inbasket message characteristics with physician burnout.","volume":"5","author":"Baxter","year":"2022","journal-title":"JAMA Netw Open"},{"issue":"3","key":"zoi260829r4","doi-asserted-by":"publisher","first-page":"453","DOI":"10.1093\/jamia\/ocab268","article-title":"Assessing the impact of the COVID-19 pandemic on clinician ambulatory electronic health record use.","volume":"29","author":"Holmgren","year":"2022","journal-title":"J Am Med Inform Assoc"},{"issue":"4","key":"zoi260829r5","doi-asserted-by":"publisher","first-page":"461","DOI":"10.1001\/jamainternmed.2024.8138","article-title":"Trends in physician electronic health record time and message volume.","volume":"185","author":"Holmgren","year":"2025","journal-title":"JAMA Intern Med"},{"issue":"8-9","key":"zoi260829r6","doi-asserted-by":"publisher","first-page":"855","DOI":"10.1093\/jamia\/ocz023","article-title":"Interventions to increase patient portal use in vulnerable populations: a systematic review.","volume":"26","author":"Grossman","year":"2019","journal-title":"J Am Med Inform Assoc"},{"issue":"5","key":"zoi260829r7","doi-asserted-by":"publisher","first-page":"1053","DOI":"10.1055\/a-1951-3153","article-title":"Characteristics of adult primary care patients who use the patient portal: a cross-sectional analysis.","volume":"13","author":"Casacchia","year":"2022","journal-title":"Appl Clin Inform"},{"issue":"2","key":"zoi260829r8","doi-asserted-by":"publisher","first-page":"308","DOI":"10.1093\/jamia\/ocac227","article-title":"Disparities in patient portal access and the role of providers in encouraging access and use.","volume":"30","author":"Richwine","year":"2023","journal-title":"J Am Med Inform Assoc"},{"issue":"10","key":"zoi260829r9","doi-asserted-by":"publisher","first-page":"1475","DOI":"10.1377\/hlthaff.2024.00406","article-title":"Unsustainable: why i left primary care.","volume":"43","author":"Hahn","year":"2024","journal-title":"Health Aff (Millwood)"},{"issue":"12","key":"zoi260829r10","doi-asserted-by":"publisher","first-page":"5461","DOI":"10.1287\/mnsc.2017.2900","article-title":"The impact of e-visits on visit frequencies and patient health: evidence from primary care.","volume":"64","author":"Bavafa","year":"2018","journal-title":"Manage Sci"},{"issue":"3","key":"zoi260829r11","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.2618","article-title":"Differences in care team response to patient portal messages by patient race and ethnicity.","volume":"7","author":"Tang","year":"2024","journal-title":"JAMA Netw Open"},{"issue":"10","key":"zoi260829r12","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2025.34549","article-title":"Patient-physician messaging by race, ethnicity, insurance type, and preferred language.","volume":"8","author":"Rotenstein","year":"2025","journal-title":"JAMA Netw Open"},{"issue":"7","key":"zoi260829r14","doi-asserted-by":"publisher","first-page":"698","DOI":"10.1097\/MLR.0000000000000735","article-title":"Identifying increased risk of readmission and in-hospital mortality using hospital administrative data: the AHRQ Elixhauser Comorbidity Index.","volume":"55","author":"Moore","year":"2017","journal-title":"Med Care"},{"issue":"4","key":"zoi260829r16","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.4867","article-title":"Content of patient electronic messages to physicians in a large integrated system.","volume":"7","author":"Liu","year":"2024","journal-title":"JAMA Netw Open"},{"issue":"1","key":"zoi260829r17","doi-asserted-by":"publisher","first-page":"e24","DOI":"10.37765\/ajmc.2022.88817","article-title":"Understanding physicians\u2019 work via text analytics on EHR inbox messages.","volume":"28","author":"Escribe","year":"2022","journal-title":"Am J Manag Care"},{"key":"zoi260829r18","doi-asserted-by":"publisher","first-page":"110","DOI":"10.1016\/j.ijmedinf.2017.06.004","article-title":"A comparison of rule-based and machine learning approaches for classifying patient portal messages.","volume":"105","author":"Cronin","year":"2017","journal-title":"Int J Med Inform"},{"issue":"4","key":"zoi260829r21","doi-asserted-by":"publisher","first-page":"471","DOI":"10.1080\/00224545.2021.1921681","article-title":"Please and no, thank you: politeness norms alter compliance more when refusing than when making or acquiescing to a request.","volume":"162","author":"Gasper","year":"2022","journal-title":"J Soc Psychol"},{"issue":"2","key":"zoi260829r22","doi-asserted-by":"publisher","first-page":"300","DOI":"10.1038\/s41562-023-01761-z","article-title":"The formality effect.","volume":"8","author":"Linos","year":"2024","journal-title":"Nat Hum Behav"},{"issue":"2","key":"zoi260829r26","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1086\/683668","article-title":"When do covariates matter? and which ones, and how much?","volume":"34","author":"Gelbach","year":"2016","journal-title":"J Labor Econ"},{"issue":"15","key":"zoi260829r27","doi-asserted-by":"publisher","first-page":"4002","DOI":"10.1007\/s11606-022-07766-0","article-title":"The electronic health record inbox: recommendations for relief.","volume":"37","author":"Sinsky","year":"2022","journal-title":"J Gen Intern Med"},{"key":"zoi260829r28","doi-asserted-by":"publisher","DOI":"10.1056\/CAT.22.0438","article-title":"In-basket reduction: a multiyear pragmatic approach to lessen the work burden of primary care physicians.","author":"Fogg","year":"2023","journal-title":"NEJM Catalyst"},{"issue":"4","key":"zoi260829r29","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.6565","article-title":"AI-generated draft replies integrated into health records and physicians\u2019 electronic communication.","volume":"7","author":"Tai-Seale","year":"2024","journal-title":"JAMA Netw Open"},{"issue":"2","key":"zoi260829r30","doi-asserted-by":"publisher","DOI":"10.1093\/jamiaopen\/ooae028","article-title":"Generative artificial intelligence responses to patient messages in the electronic health record: early lessons learned.","volume":"7","author":"Baxter","year":"2024","journal-title":"JAMIA Open"},{"issue":"3","key":"zoi260829r31","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.3201","article-title":"Artificial intelligence-generated draft replies to patient inbox messages.","volume":"7","author":"Garcia","year":"2024","journal-title":"JAMA Netw Open"},{"issue":"10","key":"zoi260829r32","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.38573","article-title":"Utility of artificial intelligence-generative draft replies to patient messages.","volume":"7","author":"English","year":"2024","journal-title":"JAMA Netw Open"},{"issue":"6","key":"zoi260829r33","doi-asserted-by":"publisher","first-page":"589","DOI":"10.1001\/jamainternmed.2023.1838","article-title":"Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum.","volume":"183","author":"Ayers","year":"2023","journal-title":"JAMA Intern Med"},{"issue":"1","key":"zoi260829r34","doi-asserted-by":"publisher","first-page":"e12","DOI":"10.1016\/S2589-7500(23)00225-X","article-title":"Assessing the potential of GPT-4 to perpetuate racial and gender biases in health care: a model evaluation study.","volume":"6","author":"Zack","year":"2024","journal-title":"Lancet Digit Health"},{"issue":"8028","key":"zoi260829r35","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1038\/s41586-024-07856-5","article-title":"AI generates covertly racist decisions about people based on their dialect.","volume":"633","author":"Hofmann","year":"2024","journal-title":"Nature"},{"issue":"6464","key":"zoi260829r36","doi-asserted-by":"publisher","first-page":"447","DOI":"10.1126\/science.aax2342","article-title":"Dissecting racial bias in an algorithm used to manage the health of populations.","volume":"366","author":"Obermeyer","year":"2019","journal-title":"Science"},{"key":"zoi260829r23","author":"Xu","year":"2025"},{"key":"zoi260829r37","author":"Institute of Medicine","year":"2003"},{"key":"zoi260829r1","unstructured":"Ryan? B. Emailing your doctor may carry a fee. New York Times. January 24, 2023. Accessed February 20, 2023. https:\/\/www.nytimes.com\/2023\/01\/24\/health\/emails-billing-doctors-patients.html"},{"key":"zoi260829r3","unstructured":"Noguchi? Y. \u201cHi, Doc!\u201d DM\u2019ing the doctor could cost you (or your insurance plan). NPR. July 21, 2023. Accessed July 31, 2023. https:\/\/www.npr.org\/sections\/health-shots\/2023\/07\/21\/1188739567\/direct-message-doctor-patient-portal-charge"},{"key":"zoi260829r13","unstructured":"Freitas? W. bizdays: Business Days Calculations and Utilities. Published online January 8, 2025. Accessed April 8, 2025. https:\/\/cran.r-project.org\/web\/packages\/bizdays\/index.html"},{"key":"zoi260829r15","unstructured":"langdetect: language detection library ported from Google\u2019s language-detection. Accessed April 8, 2025. https:\/\/github.com\/Mimino666\/langdetect"},{"key":"zoi260829r19","doi-asserted-by":"crossref","unstructured":"Lewis? M, Liu? Y, Goyal? N, . BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension.? arXiv. Preprint posted online October 29, 2019. doi:10.18653\/v1\/2020.acl-main.703","DOI":"10.18653\/v1\/2020.acl-main.703"},{"key":"zoi260829r20","unstructured":"Facebook. bart-large-mnli. Hugging Face. Published online January 4, 2024. Accessed April 8, 2025. https:\/\/huggingface.co\/facebook\/bart-large-mnli"},{"key":"zoi260829r24","unstructured":"Twitter. cardiffnlp\/twitter-roberta-base-sentiment. Hugging Face. January 4, 2024. Accessed April 8, 2025. https:\/\/huggingface.co\/cardiffnlp\/twitter-roberta-base-sentiment"},{"key":"zoi260829r25","unstructured":"textstat: calculate statistical features from text. Accessed April 8, 2025. https:\/\/github.com\/textstat\/textstat"}],"container-title":["JAMA Network Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/articlepdf\/2853209\/tang_2026_oi_260829_1786974533.96781.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T15:43:23Z","timestamp":1787586203000},"score":1,"resource":{"primary":{"URL":"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/fullarticle\/2853209"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,24]]},"references-count":37,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2026,8,3]]}},"URL":"https:\/\/doi.org\/10.1001\/jamanetworkopen.2026.30379","relation":{},"ISSN":["2574-3805"],"issn-type":[{"value":"2574-3805","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,24]]}}}