{"id":26278,"date":"2026-09-15T11:42:33","date_gmt":"2026-09-15T03:42:33","guid":{"rendered":"https:\/\/pi-union.com\/?p=26278"},"modified":"2026-09-15T12:40:34","modified_gmt":"2026-09-15T04:40:34","slug":"ai-generated-discharge-dates-how-accurate-are-they-compared-with-case-managers","status":"publish","type":"post","link":"https:\/\/pi-union.com\/en\/2026\/09\/15\/ai-generated-discharge-dates-how-accurate-are-they-compared-with-case-managers\/","title":{"rendered":"AI-Generated Discharge Dates: How Accurate Are They Compared With Case Managers?"},"content":{"rendered":"<h3 class=\"wp-block-heading\">AI-Generated Discharge Dates: How Accurate Are They Compared With Case Managers?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Authors:<\/strong> Havish S. Kantheti, Connor Dolan, Jordan Dale<br><strong>Journal:<\/strong> <em>JAMA Network Open<\/em><br><strong>Published:<\/strong> September 3, 2026<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Part I. Study Overview<\/h3>\n\n\n\n<h3 class=\"wp-block-heading\">Background<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Accurately predicting when hospitalized patients will be discharged can help hospitals optimize <strong>bed planning, patient flow, and post-discharge care coordination<\/strong>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Machine-learning tools can use electronic health record (EHR) data to predict discharge dates, and commercial AI tools are increasingly being adopted in hospitals. However, most previous studies have compared AI predictions with patients\u2019 actual discharge dates rather than with the <strong>case managers<\/strong> who are directly involved in discharge planning.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This study therefore compared:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*<strong>AI-predicted discharge dates<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*<strong>Case manager\u2013predicted discharge dates<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*<strong>Actual discharge dates<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">at three stages of hospitalization: <strong>admission, approximately 48 hours before discharge, and approximately 24 hours before discharge.<\/strong><\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Methods<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">This quality improvement study was conducted at <strong>Houston Methodist Hospital<\/strong> in the United States.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The study included <strong>22,349 hospitalizations involving 17,173 patients<\/strong>, with discharge dates between August 1, 2023, and February 28, 2024.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">A commercial AI clinical decision-support tool integrated with the EHR generated predicted discharge dates. Case managers independently documented their expected discharge dates as part of routine clinical practice.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Prediction accuracy was assessed using:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*<strong>Mean absolute error (MAE):<\/strong> the average difference, in days, between predicted and actual discharge dates. Lower values indicate greater accuracy.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*<strong>Mean difference:<\/strong> predicted date minus actual date. Negative values indicate earlier-than-actual predictions, whereas positive values indicate later-than-actual predictions.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Key Findings<\/h3>\n\n\n\n<h4 class=\"wp-block-heading\">1. At admission: AI and case managers performed similarly<\/h4>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th><\/th><th>AI<\/th><th>Case Managers<\/th><\/tr><\/thead><tbody><tr><td><strong>MAE<\/strong><\/td><td><strong>4.20 days<\/strong><\/td><td><strong>4.27 days<\/strong><\/td><\/tr><tr><td><strong>Within \u00b11 day<\/strong><\/td><td>40.8%<\/td><td><strong>46.3%<\/strong><\/td><\/tr><tr><td><strong>Exact prediction<\/strong><\/td><td>15.3%<\/td><td><strong>23.6%<\/strong><\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">At admission, AI performed approximately as well as case managers overall. However, case managers were more likely to predict the exact discharge date or a date within \u00b11 day.<\/p>\n\n\n\n<h4 class=\"wp-block-heading\">2. Approximately 48 hours before discharge: case managers became more accurate<\/h4>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th><\/th><th>AI<\/th><th>Case Managers<\/th><\/tr><\/thead><tbody><tr><td><strong>MAE<\/strong><\/td><td><strong>1.59 days<\/strong><\/td><td><strong>1.29 days<\/strong><\/td><\/tr><tr><td><strong>Within \u00b11 day<\/strong><\/td><td>41.7%<\/td><td><strong>63.6%<\/strong><\/td><\/tr><tr><td><strong>Exact prediction<\/strong><\/td><td>13.6%<\/td><td><strong>24.7%<\/strong><\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The advantage of case managers became increasingly apparent as discharge approached.<\/p>\n\n\n\n<h4 class=\"wp-block-heading\">3. Approximately 24 hours before discharge: case managers clearly outperformed AI<\/h4>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th><\/th><th>AI<\/th><th>Case Managers<\/th><\/tr><\/thead><tbody><tr><td><strong>MAE<\/strong><\/td><td><strong>1.93 days<\/strong><\/td><td><strong>0.98 days<\/strong><\/td><\/tr><tr><td><strong>Within \u00b11 day<\/strong><\/td><td><strong>37.9%<\/strong><\/td><td><strong>79.5%<\/strong><\/td><\/tr><tr><td><strong>Exact prediction<\/strong><\/td><td>7.8%<\/td><td><strong>33.6%<\/strong><\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">This was the most striking finding.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Nearly <strong>80% of case manager predictions<\/strong> were within \u00b11 day of the actual discharge date, compared with only <strong>37.9% of AI predictions<\/strong>.<\/p>\n\n\n\n<h4 class=\"wp-block-heading\">4. AI prediction bias changed over time<\/h4>\n\n\n\n<p class=\"wp-block-paragraph\">At admission, the AI had a mean difference of <strong>\u22122.74 days<\/strong>, suggesting that it tended to predict discharge too early.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Approximately 24 hours before discharge, the mean difference became <strong>+1.24 days<\/strong>, indicating a tendency to predict discharge too late.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In contrast, case managers had a mean difference of only <strong>+0.04 days<\/strong> near discharge, indicating almost no systematic bias.<\/p>\n\n\n\n<h4 class=\"wp-block-heading\">5. AI performance varied by length of stay and clinical setting<\/h4>\n\n\n\n<p class=\"wp-block-paragraph\">AI did not consistently perform worse than case managers.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For patients hospitalized for <strong>1\u20132 days<\/strong>, MAE was:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*AI: <strong>1.66 days<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Case managers: <strong>0.89 days<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For patients hospitalized for <strong>5\u20137 days<\/strong>, AI actually performed better:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*AI: <strong>1.49 days<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Case managers: <strong>2.24 days<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For hospitalizations lasting <strong>more than 14 days<\/strong>, both approaches performed relatively poorly.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Performance also varied across hospital units. Case managers were more accurate in obstetrics, whereas both AI and case managers performed poorly in the ICU.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Conclusion<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">AI performed similarly to current clinical practice when predicting discharge dates at admission. However, as patients approached actual discharge, <strong>case managers became substantially more accurate than AI<\/strong>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The findings suggest that AI may provide useful early predictions but may not yet capture the real-time clinical and contextual information available to healthcare professionals near discharge.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">Part II. PI-Union Medical Science Commentary<\/h2>\n\n\n\n<h2 class=\"wp-block-heading\">Can AI Really Replace Clinical Judgment?<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The most important message of this study is not simply that <strong>humans performed better than AI<\/strong>. Rather, it raises a more important question:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<h4 class=\"wp-block-heading\"><strong>Should we compare \u201cAI vs. humans,\u201d or should we develop better \u201cAI + humans\u201d systems?<\/strong><\/h4>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">Most AI studies evaluate whether an algorithm can predict an outcome accurately. This study goes one step further by comparing AI directly with the <strong>case managers who are actually involved in discharge planning<\/strong>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This makes the research particularly relevant to real-world clinical practice.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">1. AI has value for early prediction<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">At admission, AI and case managers had similar MAEs: <strong>4.20 versus 4.27 days<\/strong>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This suggests that AI can provide useful early estimates when the future clinical course remains highly uncertain.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For hospital operations, AI could potentially support:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>*Bed planning<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>*Patient flow management<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>*Length-of-stay forecasting<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>*Resource allocation<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>*Post-discharge care planning<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">AI also has the advantage of continuously analyzing large amounts of EHR data and updating predictions.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">2. But near discharge, human judgment becomes much more accurate<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The most important finding is that case managers became increasingly accurate as discharge approached.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">At approximately 24 hours before discharge:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>Case manager MAE: 0.98 days<\/strong><br><strong>AI MAE: 1.93 days<\/strong><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">And:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>79.5% of case manager predictions were within \u00b11 day, compared with 37.9% for AI.<\/strong><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">Why might this happen?<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Clinical professionals may have access to information that is difficult for current AI systems to fully capture, such as:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Current clinical status<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Treatment progress<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Physician decisions<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Family readiness<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Transportation and home-care arrangements<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Rehabilitation needs<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Medication preparation<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Social and caregiving circumstances<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In other words:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>More data do not necessarily mean better clinical judgment.<\/strong><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">Clinical decision-making often requires integrating structured data with real-time, contextual, and social information.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">3. AI is not simply \u201cworse than humans\u201d<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The length-of-stay analysis provides an important nuance.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Among patients hospitalized for <strong>5\u20137 days<\/strong>, AI actually had a lower MAE than case managers (<strong>1.49 vs 2.24 days<\/strong>).<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This suggests that AI performance depends on:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>Who, when, and where.<\/strong><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">In future AI research, overall accuracy alone may therefore be insufficient. We should also ask:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>For whom does AI work best?<\/strong><\/li>\n\n\n\n<li><strong>At what point during hospitalization?<\/strong><\/li>\n\n\n\n<li><strong>In which clinical settings?<\/strong><\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">4. The future may be Human + AI<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">This study does not demonstrate that AI is ineffective.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Instead, it suggests that AI may be more valuable as a <strong>clinical augmentation tool<\/strong> rather than as a replacement for clinical judgment.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">A future discharge management system could allow:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>AI to:<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Continuously analyze EHR data<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Update discharge predictions<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Identify potential discharge delays<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Support bed management<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Flag high-risk cases<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Clinical professionals to:<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Interpret real-time clinical status<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Integrate family and social circumstances<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Evaluate discharge readiness<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Address exceptions that AI cannot recognize<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The goal would therefore be:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>AI prediction + Clinical judgment = Better discharge planning<\/strong><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">rather than:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>AI replaces clinical judgment<\/strong><\/p>\n<\/blockquote>\n\n\n\n<h3 class=\"wp-block-heading\">5. The next question is not \u201cIs AI accurate?\u201d but \u201cDoes AI improve care?\u201d<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The study primarily evaluated <strong>prediction accuracy<\/strong>, not patient outcomes.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Future prospective studies should determine whether AI-supported discharge prediction can actually:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Reduce length of stay<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Improve bed utilization<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Reduce avoidable discharge delays<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Improve transitions of care<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Reduce healthcare costs<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Decrease clinicians\u2019 workload<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">*Improve patient outcomes<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">PI-Union Medical Science Take-Home Message<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The key lesson from this study is:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<h3 class=\"wp-block-heading\"><strong>AI can predict, but clinical professionals still need to judge.<\/strong><\/h3>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">AI performed reasonably well at admission, but case managers became substantially more accurate as patients approached discharge.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Therefore, the future of healthcare AI may not be:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\"><strong>Human vs. AI<\/strong><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">but rather:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<h3 class=\"wp-block-heading\"><strong>Human + AI<\/strong><\/h3>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">AI can process large amounts of data and generate continuous predictions, while clinical professionals contribute real-time judgment, contextual understanding, and patient-centered decision-making.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>The most mature form of clinical AI may not be AI replacing clinicians, but AI and clinicians working together to make better decisions.<\/strong><\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/youtube.com\/shorts\/8W5XKqjL8iE\">https:\/\/youtube.com\/shorts\/8W5XKqjL8iE<\/a><\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Reference: <\/strong><a href=\"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/fullarticle\/2853623?guestAccessKey=0db5b2cb-fca9-4548-8400-9277a252ae6d&amp;utm_medium=email&amp;utm_source=postup_jn&amp;utm_campaign=article_alert-jamanetworkopen&amp;utm_content=new_this_week_&amp;utm_term=090426\">Artificial Intelligence\u2013Generated Discharge Dates and Estimation Accuracy in Hospitalized Patients<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Reviewer:&nbsp;<a href=\"https:\/\/pi-union.com\/en\/\" target=\"_blank\" rel=\"noopener\">PI-Union Medical Science Ltd.<\/a><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">* E-mail:&nbsp;piunion@pi-union.com<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">* Official Website:&nbsp;<a href=\"https:\/\/pi-union.com\/en\/\" target=\"_blank\" rel=\"noopener\">https:\/\/pi-union.com\/<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">* Facebook:&nbsp;<a href=\"https:\/\/www.facebook.com\/piunion2020\/\" target=\"_blank\" rel=\"noopener\">www.facebook.com\/piunion2020<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">* Youtube:&nbsp;<a href=\"https:\/\/www.youtube.com\/@pi-union\">www.youtube.com\/@pi-union<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">* Instagram:&nbsp;<a href=\"https:\/\/www.instagram.com\/piunion2020\/\">www.instagram.com\/piunion202<\/a>0<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">* LINE: @654eukag<\/p>","protected":false},"excerpt":{"rendered":"<p>AI-Generated Discharge Dates: How Accurate Are They Com [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":26279,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[71,1],"tags":[430,155,432,431,429],"class_list":["post-26278","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-holistic-health","category-medical-news","tag-accuracy","tag-ai","tag-case-managers","tag-clinicians","tag-discharge"],"_links":{"self":[{"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/posts\/26278","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/comments?post=26278"}],"version-history":[{"count":2,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/posts\/26278\/revisions"}],"predecessor-version":[{"id":26281,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/posts\/26278\/revisions\/26281"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/media\/26279"}],"wp:attachment":[{"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/media?parent=26278"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/categories?post=26278"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/pi-union.com\/en\/wp-json\/wp\/v2\/tags?post=26278"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}