{"id":7972,"date":"2018-04-09T17:04:15","date_gmt":"2018-04-09T21:04:15","guid":{"rendered":"https:\/\/digital.hbs.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/"},"modified":"2018-04-09T17:04:15","modified_gmt":"2018-04-09T21:04:15","slug":"medaware-using-ai-to-eliminate-prescription-errors","status":"publish","type":"hck-submission","link":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/","title":{"rendered":"MedAware \u2013 Using AI to eliminate prescription errors"},"content":{"rendered":"

Background<\/strong><\/p>\n

A 2016 study from researchers from the Johns Hopkins University\u2019s School of Medicine found that deaths from medical errors may be responsible for more than 250,000 deaths annually, making this the third leading cause of a non-violent death in the US. (1)<\/p>\n

Headquartered in Israel, MedAware was founded in 2012 by a former physician, Dr. Gidi Stein. MedAware’s patent-pending technology uses big data analytics and machine learning algorithms to analyze large scale data of Electronic Medical Records (EMRs), to automatically learn how physicians treat patients in real life scenarios. When a new prescription deviates from the spectrum of typical treatment patterns, it\u2019s flagged as a potential error and prompts the physician to double check.<\/p>\n

\"\"<\/a><\/p>\n

Value Creation<\/strong><\/p>\n

By identifying and preventing prescription errors in real-time, MedAware is creating value to patients, physicians, and hospitals.<\/p>\n

For patients, MedAware solutions can save lives, improve their safety to reduce preventable adverse effects from prescription errors, and prevent unnecessary healthcare costs. MedAware\u2019s trial studies done at major Israeli hospitals showed that patients for whom alerts were generated had statistically significantly higher short-term mortality rates than patients without alerts. (2)<\/p>\n

For physicians, who are often overworked and serving increasing number of patients, MedAware serves as a valuable tool to check against prescription errors and reduce potential lawsuits from their errors.<\/p>\n

For hospitals, MedAware helps reduce the inevitable human error of physicians, and costs associated with those errors. For example, MedAware\u2019s trial studies showed that patients for whom alerts were generated had significantly longer hospital stays and more hospital admissions. (3) (The trial results were statistically significant, p <0.001). \u00a0If those patients received proper prescriptions, the hospital beds and services could be freed up to serve more patients in need.<\/p>\n

\"\"<\/a><\/p>\n

Differentiation from existing CDS systems<\/strong><\/p>\n

A key difference is that existing clinical decision support (CSD) screening systems are not based on screening for outliers, which seems to be an industry first.<\/p>\n

In addition, MedAware is differentiating itself by having alerts with high specificity and accuracy, preventing \u201calert fatigue\u201d. MedAware only flags about 0.2 to 0.5% of all prescriptions, with about 75-80% being true positives, and only 35% are false negatives. (4)<\/p>\n

In contrast, existing CDS screening systems can only detect a small number of actual errors because they are not patient-specific or sufficiently adaptable. Thus, they often result in a high volume of false alarm rates, causing doctors to begin to disregarding these fake alerts. According to MedAware\u2019s research, most electronical medical records (EMR) systems flag 20-30% of prescriptions and have up to 90% false positives for suspected interactions or allergies for medications. (5)<\/p>\n

\"\"<\/a><\/p>\n

Market Validation<\/strong><\/p>\n

In a January 2017 study published in the Journal of American Medical Informatics Association, 性视界 Medical School researchers used the algorithmic software from MedAware, to screen five years of electronic health record data to detect outliers suggestive of potential medication errors and compared with an existing clinical decision support (CDS) screening system. (6)<\/p>\n

This January 2017 study has also been important in validating MedAware\u2019s solutions. Using MedAware, researchers got alerts for over 15,000 charts and in a sample of 300 charts, they found that 75% of alerts were valid in identifying potential medication errors. Of the valid alerts, 75% were for potentially life-threatening prescription errors, providing validation of MedAware\u2019s probabilistic, machine-learning approach. (7)<\/p>\n

\"\"<\/a><\/p>\n

Value Capture<\/strong><\/p>\n

Currently MedAware sells several tools to healthcare providers \u2013 the core alerting system that is also self learning, as well as several decision support tools around improving risk management, quality control, and reducing alert fatigue. (8)<\/p>\n

According to the CEO, the cost to providers is about 5% of the expected cost reduction, which is about the industry standard.<\/p>\n

In the future, I believe there could also be opportunities to sell its data to drug makers and their marketing teams. MedAware\u2019s aggregated data on patients\u2019 medical records, their prescriptions, and medical journey provide a more complete and specific picture that have generated insights on different patient behaviors, prescription patterns, and off-label opportunities.<\/p>\n

Future Priorities<\/strong><\/p>\n

MedAware is currently seeing positive results with its product live in Israel\u2019s largest hospital, Sheba Medical Center, and has recently established a Stamford, Connecticut office as it looks to expand its US footprint. The company has raised a total of $12M, from Series A venture funding and grants. (9)<\/p>\n

The company plans to develop addition machine learning-enabled decision support solutions as well as increase its data to cover more catastrophic types of errors. To expand into the US, MedAware is also expanding its number of electronic medical records (EMR) integrations.<\/p>\n

Future Challenges<\/strong><\/p>\n

Going forward, I also see several challenges for the company.<\/p>\n

Although initial research results are promising, MedAware is operating in a risk-averse and highly regulated industry so the company must continue to invest in research studies to validate its alerting system to prove that results are not correlation, but causation directly stemming from preventable prescription errors.<\/p>\n

Screening through outliers is a very new way of detection that must be further proved out. These MedAware tools are also very new relative to the tools that hospitals are used to using for many years. Thus, the hospital management must be willing to experiment with such new products and train its staff accordingly for this behavior change.<\/p>\n

There are also many differences in healthcare systems across geographies \u2013 different regulations, patient demographics, behavior, availability of drugs etc so data from one country may not be directly applicable to data for another country, which can make it challenging for MedAware to expand to US or other countries quickly.<\/p>\n

Importantly, the system is only as good as the data and prescription pattern it receives, so integrating well with the electronical medical record (EMR) systems is critical, and ensuring that the data collection process is also reliable. However, these EMR systems come with existing tools and may view MedAware as a competitor.<\/p>\n

Additionally, medical science is constantly evolving and MedAware faces risk that older prescription patterns and former practices will be outdated \u2013 there will perhaps need to be ways to prioritize newer data or certain sources (e.g., leading medical centers).<\/p>\n

Lastly, MedAware\u2019s technology is still patent-pending so it faces copycat solutions from incoming competitors. This can be mitigated if the company can leverage its first mover advantage and build a larger and higher quality dataset first.<\/p>\n

Sources:<\/strong><\/span><\/p>\n

    \n
  1. http:\/\/abcnews.go.com\/Health\/medical-errors-result-200000-deaths-study-finds\/story?id=38840983<\/a><\/li>\n
  2. http:\/\/www.medaware.com\/israeli-pilot-studies\/<\/a><\/li>\n
  3. http:\/\/www.medaware.com\/israeli-pilot-studies\/<\/a><\/li>\n
  4. http:\/\/www.telemedmag.com\/article\/israeli-start-introduces-artificial-intelligence-reduce-prescribing-errors\/<\/a><\/li>\n
  5. http:\/\/www.telemedmag.com\/article\/israeli-start-introduces-artificial-intelligence-reduce-prescribing-errors\/<\/a><\/li>\n
  6. http:\/\/www.mobihealthnews.com\/content\/harvard-study-almost-800k-lives-shows-medaware-technology-reduces-medication-error<\/a><\/li>\n
  7. http:\/\/www.mobihealthnews.com\/content\/harvard-study-almost-800k-lives-shows-medaware-technology-reduces-medication-error<\/a><\/li>\n
  8. http:\/\/www.medaware.com\/our-products\/<\/a><\/li>\n
  9. http:\/\/www.medaware.com\/medaware-raises-8-million-in-series-a-funding-to-eradicate-catastrophic-prescription-errors\/<\/a><\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"

    MedAware’s machine-learning algorithms mine data gathered from millions of electronic medical records to detect outliers in prescription behavior and flag them in real-time to healthcare providers<\/p>\n","protected":false},"author":2227,"featured_media":7973,"comment_status":"open","ping_status":"closed","template":"","categories":[877,24,366],"class_list":["post-7972","hck-submission","type-hck-submission","status-publish","has-post-thumbnail","hentry","category-ai","category-healthcare","category-machine-learning","hck-taxonomy-organization-medaware","hck-taxonomy-industry-health","hck-taxonomy-country-israel"],"connected_submission_link":"https:\/\/d3.harvard.edu\/platform-digit\/assignment\/competing-with-data-challenge\/","yoast_head":"\nMedAware \u2013 Using AI to eliminate prescription errors - Digital Innovation and Transformation<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"MedAware \u2013 Using AI to eliminate prescription errors - Digital Innovation and Transformation\" \/>\n<meta property=\"og:description\" content=\"MedAware's machine-learning algorithms mine data gathered from millions of electronic medical records to detect outliers in prescription behavior and flag them in real-time to healthcare providers\" \/>\n<meta property=\"og:url\" content=\"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/\" \/>\n<meta property=\"og:site_name\" content=\"Digital Innovation and Transformation\" \/>\n<meta property=\"og:image\" content=\"https:\/\/d3.harvard.edu\/platform-digit\/wp-content\/uploads\/sites\/2\/2018\/04\/medaware-logo-1.png\" \/>\n\t<meta property=\"og:image:width\" content=\"423\" \/>\n\t<meta property=\"og:image:height\" content=\"139\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/\",\"url\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/\",\"name\":\"MedAware \u2013 Using AI to eliminate prescription errors - Digital Innovation and Transformation\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2018\\\/04\\\/medaware-logo-1.png\",\"datePublished\":\"2018-04-09T21:04:15+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/#primaryimage\",\"url\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2018\\\/04\\\/medaware-logo-1.png\",\"contentUrl\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2018\\\/04\\\/medaware-logo-1.png\",\"width\":423,\"height\":139},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/medaware-using-ai-to-eliminate-prescription-errors\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Submissions\",\"item\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/submission\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"MedAware \u2013 Using AI to eliminate prescription errors\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/#website\",\"url\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/\",\"name\":\"Digital Innovation and Transformation\",\"description\":\"MBA Student Perspectives\",\"potentialAction\":[{\"@type\":\"性视界Action\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-digit\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"MedAware \u2013 Using AI to eliminate prescription errors - Digital Innovation and Transformation","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/","og_locale":"en_US","og_type":"article","og_title":"MedAware \u2013 Using AI to eliminate prescription errors - Digital Innovation and Transformation","og_description":"MedAware's machine-learning algorithms mine data gathered from millions of electronic medical records to detect outliers in prescription behavior and flag them in real-time to healthcare providers","og_url":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/","og_site_name":"Digital Innovation and Transformation","og_image":[{"width":423,"height":139,"url":"https:\/\/d3.harvard.edu\/platform-digit\/wp-content\/uploads\/sites\/2\/2018\/04\/medaware-logo-1.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/","url":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/","name":"MedAware \u2013 Using AI to eliminate prescription errors - Digital Innovation and Transformation","isPartOf":{"@id":"https:\/\/d3.harvard.edu\/platform-digit\/#website"},"primaryImageOfPage":{"@id":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/#primaryimage"},"image":{"@id":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/#primaryimage"},"thumbnailUrl":"https:\/\/d3.harvard.edu\/platform-digit\/wp-content\/uploads\/sites\/2\/2018\/04\/medaware-logo-1.png","datePublished":"2018-04-09T21:04:15+00:00","breadcrumb":{"@id":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/#primaryimage","url":"https:\/\/d3.harvard.edu\/platform-digit\/wp-content\/uploads\/sites\/2\/2018\/04\/medaware-logo-1.png","contentUrl":"https:\/\/d3.harvard.edu\/platform-digit\/wp-content\/uploads\/sites\/2\/2018\/04\/medaware-logo-1.png","width":423,"height":139},{"@type":"BreadcrumbList","@id":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/medaware-using-ai-to-eliminate-prescription-errors\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/d3.harvard.edu\/platform-digit\/"},{"@type":"ListItem","position":2,"name":"Submissions","item":"https:\/\/d3.harvard.edu\/platform-digit\/submission\/"},{"@type":"ListItem","position":3,"name":"MedAware \u2013 Using AI to eliminate prescription errors"}]},{"@type":"WebSite","@id":"https:\/\/d3.harvard.edu\/platform-digit\/#website","url":"https:\/\/d3.harvard.edu\/platform-digit\/","name":"Digital Innovation and Transformation","description":"MBA Student Perspectives","potentialAction":[{"@type":"性视界Action","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/d3.harvard.edu\/platform-digit\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"_links":{"self":[{"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/hck-submission\/7972","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/hck-submission"}],"about":[{"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/types\/hck-submission"}],"author":[{"embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/users\/2227"}],"replies":[{"embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/comments?post=7972"}],"version-history":[{"count":0,"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/hck-submission\/7972\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/media\/7973"}],"wp:attachment":[{"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/media?parent=7972"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-digit\/wp-json\/wp\/v2\/categories?post=7972"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}