<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Jorge A. Caballero, MD on Medium]]></title>
        <description><![CDATA[Stories by Jorge A. Caballero, MD on Medium]]></description>
        <link>https://medium.com/@DataDrivenMD?source=rss-fd0e8cb79ac1------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*38XulOV-henVc8Rsh9ZtMA.png</url>
            <title>Stories by Jorge A. Caballero, MD on Medium</title>
            <link>https://medium.com/@DataDrivenMD?source=rss-fd0e8cb79ac1------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 19:01:42 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@DataDrivenMD/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[The Social Media Landscape 1-year after Elon’s Twitter Takeover]]></title>
            <link>https://medium.com/@DataDrivenMD/the-social-media-landscape-1-year-after-elons-twitter-takeover-c1f7e65ed57e?source=rss-fd0e8cb79ac1------2</link>
            <guid isPermaLink="false">https://medium.com/p/c1f7e65ed57e</guid>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[future]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[meta]]></category>
            <dc:creator><![CDATA[Jorge A. Caballero, MD]]></dc:creator>
            <pubDate>Sun, 22 Oct 2023 23:44:24 GMT</pubDate>
            <atom:updated>2023-10-22T23:44:24.116Z</atom:updated>
            <content:encoded><![CDATA[<p>After months of public posturing seemingly intended to strong-arm Twitter’s Board of Directors into negotiating a lower price, or into rejecting his unsolicited offer outright, Elon Musk’s acquisition of Twitter closed on October 28, 2022. Nearly one year to the day.</p><p>It was a wild ride leading up to the takeover, and it’s been even more tumultuous since. And yet, for as much of a roller coaster ride as the last year has been, Twitter’s product and platform has been on a steep downward trajectory from the moment Elon walked into Twitter headquarters carrying a sink.</p><figure><img alt="Screenshot of Elon Musk’s Tweet from October 26, 2022. The text reads, “Entering Twitter HQ — let that sink in!” and there’s an embedded video focused on Elon Musk as he carries a white sink into Twitter headquarters." src="https://cdn-images-1.medium.com/max/1024/1*33cC4MWOn7K-lBpDVOG2jw.png" /></figure><p>Within hours of finalizing the deal, Elon fired any Twitter executive that wasn’t deemed sympathetic to his cause. In so doing, he permanently wiped more than a decade of institutional memory from the company. And then he made matters worse by haphazardly laying off ~50% of the company’s workforce; this erased operational knowledge about mission-critical projects ranging in scope from data security to regulatory compliance.</p><p>Twitter’s brain drain happened in the first week following Elon’s takeover. By November 5, 2022, the only people remaining at Twitter were those that he and his inner circle deemed to meet 2 criteria: absolutely necessary to advance their agenda and completely bought-in to their “hardcore” approach to realizing their vision for the company.</p><p>Let <em>that</em> sink in.</p><p>Everything that’s happened since — the data breaches, the pay-to-play verification for brands, transforming Twitter Blue into a multi-level marketing scheme, alienating developers by charging exorbitant API fees, intermittent global outages — all of these are examples of self-inflicted wounds directly stemming from Musk’s impetuous dismissal of Twitter executives and employees who knew where the bodies were buried. The product’s demise was so inevitable from that point, that I posted this on November 4, 2022:</p><figure><img alt="Screenshot of a Tweet I posted on November 4, 2022, which reads, “In the not-too-distant future, someone will look back at the ashes of Twitter and point to this day as the tipping point." src="https://cdn-images-1.medium.com/max/944/1*CEr8zvOInetGkdAP9wQW-A.png" /></figure><p>Today, nearly a year later, Twitter no longer exists. Twitter is dead. Twitter has been replaced by X, a zombie that haunts our digital lives cloaked in the carcass of a dead bird. But not all is lost.</p><p>From the ashes of Twitter we are seeing the sprouts of a new era of social media — an era where microblogging feeds are less like Twitter and more like podcasts — channels of content that are accessible through a wide range of products and web services. The new era of text-based social media is Textcasting.</p><p>In this new era, apps like Mastodon and Threads will compete based on the privacy, security, moderation, and discovery features that they offer. The new generation of text-based social media apps and services is already behaving less like content sandboxes, and more like web browsers. Early movers, like Medium and Flipboard, are bridging the old world with the new; they are opening windows into an universe of communities that live on the decentralized social web.</p><p>ActivityPub, the protocol that Mastodon uses to broadcast content is the foundation. WordPress recently announced a plugin that allows publishers a turnkey solution to publish their existing content via ActivityPub — more than a third of the internet’s websites can broadcast their content without straining their existing technical or editorial resources. Mozilla, NuxtLabs, and Fastly have each committed resources to build products, services, and solutions for a decentralized social web based on the ActivityPub protocol. Meta’s Threads, the world’s fastest-growing microblogging service, is actively working toward ActivityPub compatibility.</p><p>This decentralized post-Twitter model creates massive opportunities for publishers and content creators alike, including digital activists and journalists. Creators will gain control of their content and their audiences. Publishers will automatically expand their distribution every time a new industry or vertical joins the decentralized social web. For both, it means they’ll have the flexibility to assume as much (or as little) control over their distribution as they desire — through moderation actions that can serve to broaden or focus the reach of their content. It will empower artists to directly monetize their work. It will expand the reach of digital influencers beyond the walled garden of legacy platforms. It will allow celebrities to directly engage with their audiences and to protect their personal brands and digital likenesses. It will allow for talented reporters to build their reputations while laying the groundwork for a career in independent journalism.</p><p>But there’s still work to do. Addressing the needs of marginalized communities remains an unfulfilled promise. Mastodon and Threads have invested in the development of accessibility features, such as alternative text for images, product-wide scalable fonts, and mindful color themes. The approach to content moderation on both services, however, is failing to create safe spaces for all — albeit for different reasons. Mastodon has, on several occasions, failed to moderate against racially-motivated harassment directed at Black, Native, Hispanic, and Asian persons on the service. Threads has, on several occasions, demonstrated an over-reliance on automated content moderation, which has left users frustrated when the systems issue unwarranted Community Guidelines violations for benign content (e.g. a photo of a gay couple with their child). Threads’ overt censoring of certain topics from its full-text search (e.g. COVID, vaccines) has also drawn criticism, and undermined interest in the product from well-respected scientists that would otherwise be willing to share their domain expertise on the site.</p><p>Limitations notwithstanding, there is growing momentum toward a decentralized future for the social web, and virtually all of that momentum is building on the ActivityPub protocol. A year after the beginning of the end of Twitter, the future of text-based social media is brighter than ever.</p><p><em>About the author:</em> Jorge Caballero is a tech innovator, data scientist, and advocate. He’s <a href="https://www.warren.senate.gov/newsroom/press-releases/warren-smith-question-top-retail-testing-providers-companies-on-lack-of-pediatric-testing-at-their-covid-19-testing-sites">best known for solving COVID-19 data problems that the U.S. Government and global tech companies could not</a> — and for working collaboratively to scale, manage, and evolve those solutions. Every person in the U.S. who searched the web for a COVID testing site in 2020 was served results directly stemming from his work.</p><p>Among the Very Online™️ he’s also known for spearheading the #TwitterMigration movement that fueled Mastodon’s unprecedented growth from November through December 2022. Prior to Elon Musk’s takeover, he served on Twitter’s Content Advisory Team, which was an independent external advisory board focused on platform trust and integrity. He also served on the Advisory Board for Doximity, the largest social network for healthcare professionals in the United States.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c1f7e65ed57e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[COVID-19 Hospitalizations Are on the Rise — What You Should Know (Part 3)]]></title>
            <link>https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-part-3-b06dd66efaf5?source=rss-fd0e8cb79ac1------2</link>
            <guid isPermaLink="false">https://medium.com/p/b06dd66efaf5</guid>
            <dc:creator><![CDATA[Jorge A. Caballero, MD]]></dc:creator>
            <pubDate>Mon, 26 Oct 2020 17:52:03 GMT</pubDate>
            <atom:updated>2020-10-26T19:44:16.999Z</atom:updated>
            <content:encoded><![CDATA[<h3>COVID-19 Hospitalizations Are on the Rise — What You Should Know (Part 3)</h3><h4>How to think about hospital/ICU occupancy</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*1M-3d9bLtt4HfkzS" /><figcaption>Photo by <a href="https://unsplash.com/@coolmilo?utm_source=medium&amp;utm_medium=referral">camilo jimenez</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>This article belongs to a three-part series written for anyone that is looking to understand why experts are expressing concern about rising COVID-19 hospitalizations. <a href="https://coronavirus.medium.com/covid-19-hospitalizations-are-on-the-rise-heres-what-you-should-know-fa2c47d3fcc1">Part one</a> explains what the number of hospitalized patients actually means, <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-series-b05add876050">part two</a> explains why hospital bed distribution is important, and <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-part-3-b06dd66efaf5">part three</a> covers the nuances of hospital and ICU occupancy.</p><h3>What is hospital/ICU occupancy and why does it matter?</h3><p>The concept of hospital/bed occupancy predates the COVID-19 pandemic, it reflects:</p><ul><li>availability of medical resources (the supply),</li><li>number of patients in need of hospitalization (the demand), and</li><li>how quickly patients are treated and discharged from the hospital (patient turnover)</li></ul><blockquote><em>Aside: each of these merit further discussion, but that’s out of scope for this series.</em></blockquote><figure><img alt="An overhead view of a field hospital set up at the state fair ground near Milwaukee, Wisconsin on October 12, 2020" src="https://cdn-images-1.medium.com/max/992/1*YhR_-vYwe99Vw4oXbNmZzQ.jpeg" /><figcaption>A field hospital set up at the state fair ground near Milwaukee, Wisconsin on October 12, 2020 (Photo by Wisconsin Department of Administration via Reuters)</figcaption></figure><blockquote>In the context of the COVID-19 pandemic, occupancy is a metric used to identify regions where the number of patients requiring hospitalization exceeds the available resources.</blockquote><h3>What is the usual/normal occupancy rate?</h3><blockquote>Based on historical data: approximately 64% of hospital beds in the U.S. were occupied at the start of the COVID-19 pandemic.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rbTBXDbLPZQmored12FHAQ.png" /><figcaption>Historical hospital occupancy rates for industrialized countries. Source: <a href="https://www.oecd-ilibrary.org/sites/0d67e02a-en/index.html?itemId=/content/component/0d67e02a-en">OECD Health Statistics 2019</a>.</figcaption></figure><ul><li>U.S. hospital occupancy between 2000 and 2017 hovered around ~64%. (Note: this is a national figure, which means that individual hospitals may have reported higher or lower occupancy rates)</li><li>We <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-series-b05add876050">previously mentioned</a> that there are 792,417 staffed beds in community hospitals (i.e. those that serve the general public).</li><li>Taken together, we estimate that ~285,270 total hospital beds were available across the entire United States at the start of the pandemic.</li></ul><h3>How is hospital/ICU occupancy calculated?</h3><p>Occupancy is calculated by dividing the number of patients (the numerator) by the <em>total number of staffed beds</em> (the denominator).</p><p>As simple as it may seem, this metric requires thoughtful interpretation — especially when the value is plotted over time. That’s because the numerator and denominator change over time — and they do so independent of one-another. We’ve <a href="https://coronavirus.medium.com/covid-19-hospitalizations-are-on-the-rise-heres-what-you-should-know-fa2c47d3fcc1">already covered</a> how the numerator evolves over time; so let’s briefly discuss how the <em>total</em> <em>number of staffed beds</em> (denominator) is subject to change from-day-to-day, week-to-week, or even month-to-month.</p><blockquote>The total number of staffed beds isn’t a basic count of physical beds: it’s an all-inclusive metric that reflects the number of patients that can be <em>safely </em>admitted to the hospital (or ICU).</blockquote><p>In this context, <em>safely</em> means that all of the following are available (and at-hand):</p><ul><li>A physical bed</li><li>medical supplies and equipment</li><li>clinical staffing</li></ul><p>Each of these these factors can impact <em>the total number of </em><strong><em>staffed</em></strong><em> beds</em>, as I’ve outlined in the following table:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5Ub_cwPBwzObOieAa-ce2A.png" /><figcaption>Attribution: <a href="https://medium.com/u/fd0e8cb79ac1">Jorge A. Caballero, MD</a></figcaption></figure><blockquote>When hospital (or ICU) occupancy is greater than 100%, it simply means that physical beds and staff were added.</blockquote><h3>How does this fit into the big picture?</h3><p>In the table (above) I highlighted 3 boxes that speak to the big picture:</p><ul><li>COVID-19 can decrease the number of staff available to care for patients. Getting replacements/substitutes for doctors and nurses is easier said than done — that’s why it’s imperative that we <a href="https://getusppe.org/">provide them with personal protective equipment (PPE)</a>: to keep them on the job.</li><li>The way to increase PPE supply is to increase manufacturing <strong><em>and</em></strong> accelerate its delivery to hospitals across the country.</li><li>Therefore, in order to protect our most valuable asset (healthcare workers) we need a whole-of-government approach that leverages the <a href="https://en.wikipedia.org/wiki/Defense_Production_Act_of_1950"><strong>Defense Production Act</strong></a><strong> </strong>to the greatest extent possible and we need a supply chain that is coordinated at the federal level to ensure that supplies are distributed as soon as they are manufactured— neither is currently happening on a consistent basis.</li></ul><h3>What does success look like?</h3><p>Texas Medical Center (TMC) serves as an example of a well-managed COVID-19 response — albeit at a hyperlocal level. Through a lot of hard work and coordination across state and local agencies, TMC is able to publish a <a href="https://www.tmc.edu/coronavirus-updates/overview-of-tmc-icu-bed-capacity-and-occupancy/">near real-time dashboard of ICU capacity and occupancy</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DxJztte0dbwhiU3hBX3ozQ.png" /><figcaption>Source: <a href="https://www.tmc.edu/coronavirus-updates/overview-of-tmc-icu-bed-capacity-and-occupancy/">Texas Medical Center</a></figcaption></figure><p>TMC illustrates what is possible, and what we should expect from every hospital across the country.</p><h3>Key takeaways</h3><ul><li>Hospital/ICU occupancy is a metric used to identify regions where COVID-19 is a strain on medical resources.</li><li>The denominator is the <em>total number of staffed beds — </em>this value changes on a daily, weekly, and monthly basis</li><li>Interpreting plots of hospital/ICU occupancy over time requires that we bear in mind that both the numerator and denominator change over time.</li><li>When everything is working in concert, it is possible to know exactly which hospitals have sufficient capacity, and which hospitals are under-resourced.</li></ul><h3>Another way to say it</h3><p>As it relates to COVID-19, <strong>occupancy</strong> is the most valuable metric available because the numerator and denominator are mission-critical metrics in their own right. We need access to the underlying data in order to tease apart all of the different factors that come into play.</p><p>I’m considering adding another part to this series that explains how the underlying data are collected and reported. If you’re interested in seeing this article, please “clap” or leave a comment.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b06dd66efaf5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[COVID-19 Hospitalizations Are on the Rise — What You Should Know (Series)]]></title>
            <link>https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-series-b05add876050?source=rss-fd0e8cb79ac1------2</link>
            <guid isPermaLink="false">https://medium.com/p/b05add876050</guid>
            <category><![CDATA[coronavirus]]></category>
            <category><![CDATA[covid-19-data]]></category>
            <category><![CDATA[hospital]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[covid19]]></category>
            <dc:creator><![CDATA[Jorge A. Caballero, MD]]></dc:creator>
            <pubDate>Sun, 25 Oct 2020 14:52:55 GMT</pubDate>
            <atom:updated>2020-10-26T19:28:16.044Z</atom:updated>
            <content:encoded><![CDATA[<h3>COVID-19 Hospitalizations Are on the Rise — What You Should Know (Part 2)</h3><h4>Hospital haves and have nots</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*-90JUYJotixxvuJu" /><figcaption>Photo by <a href="https://unsplash.com/@coolmilo?utm_source=medium&amp;utm_medium=referral">camilo jimenez</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>This article belongs to a three-part series written for anyone that is looking to understand why experts are expressing concern about rising COVID-19 hospitalizations. <a href="https://coronavirus.medium.com/covid-19-hospitalizations-are-on-the-rise-heres-what-you-should-know-fa2c47d3fcc1">Part one</a> explains what the number of hospitalized patients actually means, <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-series-b05add876050">part two</a> explains why hospital bed distribution is important, and <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-part-3-b06dd66efaf5">part three</a> covers the nuances of hospital and ICU occupancy.</p><p>According to the latest figures from the <a href="https://www.aha.org/statistics/fast-facts-us-hospitals">American Hospital Association</a>, there are 792,417 staffed beds in community hospitals serving the general public (i.e. excludes military, prison, and college infirmaries) — this includes beds in all kinds of intensive care units:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3Xkn9NQf_1jyKMMTOT0ISg.png" /><figcaption><strong>Number of intensive care beds in the United States by type. </strong>Medical-surgical and Cardiac intensive care provides patient care of a more intensive nature than the usual medical and surgical care to adults. Pediatric intensive care provides care to pediatric patients that is of a more intensive nature than that usually provided to pediatric patients. Neonatal intensive care is a unit that must be separate from the newborn nursery providing intensive care to all sick infants. Burn care is a unit that provides care to severely burned patients. Other intensive care is a specially staffed, specialty equipped, separate section of a hospital dedicated to the observation, care, and treatment of patients with life-threatening illnesses, injuries, or complications from which recovery is possible. Source: <a href="https://www.aha.org/statistics/fast-facts-us-hospitals">American Hospital Association Fast Facts 2020</a>.</figcaption></figure><p>As a nation, the U.S. reached 60,000 COVID-19 hospitalizations on two separate occasions. There are currently 42,000 patients hospitalized with COVID-19 across the U.S.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cKFhMW5ZsYXDTsD4e6l5Wg.png" /><figcaption>The total number of patients in the United States hospitalized with COVID-19. Source: <a href="https://covidtracking.com/data/charts/us-currently-hospitalized">The COVID Tracking Project</a></figcaption></figure><p>On paper, it seems like the U.S. should have plenty of hospital beds to handle a major surge of COVID-19 cases but that’s not actually the case — why? There are two big reasons:</p><ul><li>Not all of the beds are actually available (we’ll explore this further in the next part in this series).</li><li>There is a supply-and-demand mismatch: the beds are not always physically located where they are needed</li></ul><p>We can visualize the mismatch by comparing two maps. Here’s a map showing the number of hospital beds per 1000 population. The darker the state, the more hospital beds there are available. We can see that Louisiana, Mississipi, West Virginia, Alabama, and the Dakotas are among the top 5 states in the country by this particular measure:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eY9VhcNxfOOpGeAgL577PQ.jpeg" /><figcaption>Source: <a href="https://www.caliper.com/featured-maps/maptitude-hospital-beds-and-largest-hospital-state-map.html">Maptitude</a>.</figcaption></figure><p>We can also see that Florida, New York, and Texas are in the top 3 nationally when we measure in terms of beds within a single hospital. In other words, these three states are home to the largest hospitals in the country. Note is that the two lists are very different.</p><p>Now, compare that to the geographic distribution of cases as of October 24:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g0d7PX28PehrbrXCMLzTEQ.png" /><figcaption>Source: <a href="https://covidtracking.com/data/charts/currently-hospitalized-by-state">The COVID Tracking Project</a>.</figcaption></figure><p>This simple comparison sheds light on the need for field hospitals in <a href="https://www.samaritanspurse.org/article/field-hospital-opens-in-new-york-citys-central-park/">New York City</a> (first wave) near <a href="https://abcnews.go.com/Health/wisconsin-admits-1st-patient-field-hospital-reports-record/story?id=73759874">Milwakee, Wisconsin</a> (third wave). In both cases, the number of patients needing hospitalization for COVID-19 exceeded the number of available beds.</p><blockquote>A state with a few massive hospitals isn’t necessarily better off than a state with many smaller hospitals — the inverse is also true. What matters most is that the hospitals are physically close to the people that need them.</blockquote><h4>Key takeaways</h4><ul><li>Although the U.S. boasts nearly 800,000 hospital beds, neither the hospitals nor the beds are evenly distributed.</li><li>When the number of COVID-19 patients needing hospitalization in a geographic area exceeds the number of available hospital beds: patients need to be moved to another hospital, the government needs to set up field hospitals, or both.</li></ul><h4>Another way to say it</h4><p>To understand where COVID-19 threatens to strain medical resources, we need to know where COVID-19 cases are surging in relation to where hospital beds are physically located.</p><p>Next, we’ll go beyond raw numbers to discuss the % of beds occupied.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b05add876050" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[COVID-19 Hospitalizations Are on the Rise — Here’s What You Should Know]]></title>
            <link>https://coronavirus.medium.com/covid-19-hospitalizations-are-on-the-rise-heres-what-you-should-know-fa2c47d3fcc1?source=rss-fd0e8cb79ac1------2</link>
            <guid isPermaLink="false">https://medium.com/p/fa2c47d3fcc1</guid>
            <category><![CDATA[covid-19-data]]></category>
            <category><![CDATA[hospital]]></category>
            <category><![CDATA[covid]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[coronavirus]]></category>
            <dc:creator><![CDATA[Jorge A. Caballero, MD]]></dc:creator>
            <pubDate>Sat, 24 Oct 2020 11:03:50 GMT</pubDate>
            <atom:updated>2020-10-26T19:28:28.199Z</atom:updated>
            <content:encoded><![CDATA[<h3>Covid-19 Hospitalizations Are on the Rise — Here’s What You Should Know (Part 1)</h3><h4>How to break down the raw numbers like a pro</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*-90JUYJotixxvuJu" /><figcaption>Photo by <a href="https://unsplash.com/@coolmilo?utm_source=medium&amp;utm_medium=referral">camilo jimenez</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>This article belongs to a three-part series written for anyone that is looking to understand why experts are expressing concern about rising COVID-19 hospitalizations. <a href="https://coronavirus.medium.com/covid-19-hospitalizations-are-on-the-rise-heres-what-you-should-know-fa2c47d3fcc1">Part one</a> explains what the number of hospitalized patients actually means, <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-series-b05add876050">part two</a> explains why hospital bed distribution is important, and <a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-part-3-b06dd66efaf5">part three</a> covers the nuances of hospital and ICU occupancy.</p><p>On September 20th, U.S. Covid-19 hospitalizations reached a nadir not seen since mid-June.</p><p>Just one month later, the trend is moving in the opposite direction at an alarming rate, threatening to overwhelm resources in <a href="https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patients.html">communities with limited hospital capacity</a>.</p><figure><img alt="A chart showing the total number of U.S. persons hospitalized with COVID-19 by date." src="https://cdn-images-1.medium.com/max/1024/1*vEv_Uom-rpXpWK7bvmLWpQ.png" /><figcaption>A chart showing the total number of U.S. persons hospitalized with COVID-19 by date. (Chart by The COVID Tracking Project)</figcaption></figure><p>Visualizing data at the national level is helpful to identify <em>national</em> trends, but mobilizing resources requires far more granular data (and a lot more context). A number of people involved in the Covid-19 response effort at the federal level recognized these data gaps as early as March, and had taken concrete steps to address the issue by the time I was looped into a related meeting in late April. The meeting was an opportunity to brief the data architects responsible for building the Covid-19 Data Hub on the technical details of the Covid-19 testing site dataset produced by our volunteer effort. During that meeting, the data hub project leads shared with us their broader vision, which included near-real time hospital capacity and even weekly snapshots of PPE inventory.</p><p>We know from public reporting that the initial phase of the HHS Covid-19 Data Hub project reached a milestone in late July, <a href="https://www.aha.org/news/headline/2020-08-03-hhs-updates-covid-19-guidance-and-faqs-hospitals-daily-data-reporting">but key details remain opaque</a> and the public still lacks access to key details that are necessary to fully grasp the extent to which the risk of Covid-19 transmission did/does/will shape our day-to-day lives.</p><p>The next few sections offer a (mostly) jargon-free break down of hospitalization data that highlights what we know, and what we don’t know — and explains why it’s important to avoid drawing conclusions or drafting policy based solely on the data that is currently made available to the public. The goal is to teach non-experts how to read and interpret hospitalization data, and to illustrate the value of increased data transparency.</p><h3>Looking at raw data: how health data experts think about the number of Covid-19 hospitalizations</h3><p>The first thing to know about hospitalization data is that the raw number — that is, the number of patients hospitalized with Covid-19 — is not a meaningful measure <em>on its own </em>— it needs proper context. That’s because the risk of being hospitalized for Covid-19 can vary due to any number of factors, including some that we may not yet recognize. Age is a well-known factor:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*S7_DLPnYsWSVrWTijVOZCA.jpeg" /><figcaption>Risk of hospitalization for COVID-19 patients varies by age (Chart by <a href="https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigations-discovery/hospitalization-death-by-age.html">CDC</a>)</figcaption></figure><p>This chart tells us that if you were to take two regions of the country with comparable-sized populations, similar rates of new cases, and similar % test positivity — but one region was comprised entirely of 18–29 year olds and the other was entirely 75–84 year olds, we could expect the number of hospitalizations in the older group to be 8X higher. The real world doesn’t work that way, but it does help to explain why Covid-19 hotspots in Tampa, Florida are more likely to increase hospitalizations than a hotspot in Miami. It also underscores why we can’t just look at the raw number of hospitalizations.</p><blockquote>The number of patients hospitalized with Covid-19 is not a meaningful measure on its own — it needs context.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/786/1*-KTGDFM8AbnYMxFYFUYHcg.png" /><figcaption>Average number of new cases and hospitalizations for the state of Florida. (source: <a href="https://twitter.com/jasonsalemi?s=21">Jason L. Salemi, PhD</a> of <a href="http://www.covid19florida.strikingly.com/"><strong>www.COVID19Florida.strikingly.com</strong></a><strong>)</strong></figcaption></figure><p>Based on the information we’ve covered thus far, we can expect the number of hospitalizations to increase (or decrease) as confirmed Covid-19 cases rise and fall in the general community — as illustrated here using data for the state of Florida. This chart also illustrates that the <a href="https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patients.html">number of patients hospitalized</a> tends to lag the onset of symptoms by 8 to 12 days. This explains why the curve for hospitalizations (bottom) appears to be slightly shifted to the right as compared to the curve for new cases (top).</p><p>But each point on the curve is a just a snapshot in time — it doesn’t tell us anything about the number of days a specific person has been in the hospital. The best available data we have to measure the typical length of stay (LOS) for persons hospitalized with Covid-19 was collected very early in the pandemic; it tells us to expect a Covid-19 patient to spend <a href="https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patients.html">10 to 13 days</a> in the hospital. But the treatment of Covid-19 has evolved over time.</p><p>Assuming that this evolution translates to clinically meaningful improvements, then we can expect to see measurable improvements in hospital LOS (i.e., a statistically significant decrease) — and we do. There are two treatments that are proven to be helpful in managing Covid-19 in hospitalized patients: remdesivir and dexamethasone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bkfTBl-zqkfwFex7DyCniA.png" /><figcaption>Hospital length of stay (LOS) decreased between March and July across the state of Colorado. (Source: <a href="https://cha.com/wp-content/uploads/2020/10/COVID-19-Wave-1-Colorado-Update-9-11-20-Slides.pdf">Colorado Hospital Association</a>)</figcaption></figure><ul><li><strong>Remdesivir </strong>has been shown to reduce the typical (median) hospital LOS from <a href="https://www.nejm.org/doi/full/10.1056/NEJMoa2007764">15 to 10 days</a>.</li><li><strong>Dexamethasone</strong> was shown to <a href="https://www.nejm.org/doi/full/10.1056/NEJMoa2021436">decreased the amount of time</a> Covid-19 patients on mechanical ventilators remain intubated. There was also a slight reduction in hospital LOS from 12 to 13 days.</li></ul><p>Using the state of Colorado as example, we can see that hospital LOS decreased between March and July; as did <a href="https://www.cpr.org/2020/10/16/new-hospital-data-shows-ventilator-use-fell-over-time-and-coronavirus-patient-stays-grew-shorter/">ventilator use</a>.</p><blockquote>If the way doctors manage hospitalized Covid-19 patients has improved over time, then the data should reflect that — and it does.</blockquote><h4>Key takeaways</h4><p>When looking at the number of Covid-19 hospitalizations over time, bear in mind that:</p><ul><li>Daily hospitalization numbers are a summary snapshot for the day — the value doesn’t say much about the patients themselves</li><li>The way doctors treat hospitalized Covid-19 patients has improved, and that’s translated into shorter hospital stays.</li><li><em>Assuming nothing about the disease or the patient population has changed</em>, then the number of patients hospitalized with Covid-19 <em>on any given day </em>should be lower today, than it was over the summer.</li></ul><h4>Another way to say it</h4><p>Saying that you have 100 Covid-19 patients in a hospital today is (roughly) the same as saying that you had 70–90 patients back in March.</p><h3>What’s Next</h3><p><a href="https://medium.com/@DataDrivenMD/covid-19-hospitalizations-are-on-the-rise-what-you-should-know-series-b05add876050">Next, we’ll explain why knowing the geographic distribution of hospital beds is so critical as we move forward.</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fa2c47d3fcc1" width="1" height="1" alt=""><hr><p><a href="https://coronavirus.medium.com/covid-19-hospitalizations-are-on-the-rise-heres-what-you-should-know-fa2c47d3fcc1">COVID-19 Hospitalizations Are on the Rise — Here’s What You Should Know</a> was originally published in <a href="https://coronavirus.medium.com/">Medium Coronavirus Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Health Tech Economy]]></title>
            <link>https://medium.com/@DataDrivenMD/the-health-tech-economy-aab5bbcdb9c?source=rss-fd0e8cb79ac1------2</link>
            <guid isPermaLink="false">https://medium.com/p/aab5bbcdb9c</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[health-technology]]></category>
            <category><![CDATA[healthcare]]></category>
            <category><![CDATA[strategy]]></category>
            <dc:creator><![CDATA[Jorge A. Caballero, MD]]></dc:creator>
            <pubDate>Thu, 05 Jan 2017 14:02:17 GMT</pubDate>
            <atom:updated>2017-01-05T14:02:17.776Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://medium.com/@DataDrivenMD/3-things-all-health-tech-entrepreneurs-need-to-know-f81f783b9c01">In an earlier post</a> I introduced the idea that information gaps created by EHRs threaten to undermine the strategic position of many health tech companies. In this post I expand on this idea by illustrating the current state of the health tech economy, and highlight the different ways in which companies depend on access to EHR-derived data. In the interest of giving credit where credit is due: I’d like to acknowledge that this post was inspired by <a href="https://medium.com/u/a8e3741de9e2">Chris Dixon</a>’s excellent overview of <a href="https://medium.com/@cdixon/the-internet-economy-fc43f3eff58a#.dp41gkvnd">the Internet Economy</a>.</p><p>As with internet businesses, understanding a health tech company’s strategic position requires an understanding of 1) how the company generates revenue and 2) how those products relate to other products along the entirety of the loop. The following diagram illustrates a typical clinical encounter, which begins when a patient decides to seek medical attention and ends when:</p><ol><li>the patient and clinician mutually agree that the issue has been addressed,</li><li>the patient abstains from further treatment and/or evaluation, or</li><li>the responsible clinician ceases to offer the patient further treatment and/or evaluation.</li></ol><p>Until one of these three events occurs, the entire cycle may repeat itself an infinite number of times.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qQNNC1pekML1kv0p." /></figure><h3>A few notes about this diagram:</h3><ul><li>The diagram includes a sampling of companies operating in the health tech economy. It is not meant to be a comprehensive listing of all players in the space — <a href="https://rockhealth.com/research/">Rock Health</a> and <a href="https://www.crunchbase.com/">Crunchbase</a> offer a far more comprehensive overview of the competitive landscape.</li><li>A company’s position along the loop is a reflection of information available through public outlets (e.g. company websites, press releases, blog posts). Some companies, such as EHR vendors, have multiple product lines spanning multiple segments of the loop (not shown).</li><li>I’ve collapsed many ancillary loops under the heading “function-specific loops.” The billing and claims loop is the most mature — and most important — of the ancillary loops. It translates clinical and administrative data (from EHRs and Revenue Cycle Management software, respectively) into <a href="http://www.x12.org/">ANSI X12</a> transactions — the mechanism by which health insurance claims are submitted, eligibility is verified, and pre-approvals are obtained.</li><li>Conventional last-mile fulfillment (LMF) differs from tech-enabled LMF in that the latter relies on software as the primary means to scale the delivery of health care goods and services. By comparison, conventional LMF scales through hiring and uses software primarily as a means to manage human resources. Another key difference is that, for the most part, tech-enabled LMF companies do not rely on the billing and claims loop to collect payment for goods and services. Instead, they collect payment directly from patients or employers.</li></ul><h3>Key observations</h3><ul><li><strong>Consumer wearables and digital health companies live on islands.</strong> The good thing about being on an island is that these companies do not depend on clinically-derived data to build a great product. The drawback is that they lack access to a patient’s clinical record, which means that it’s exceedingly difficult to build features and apps that are personalized, actionable, and accurate. One way to overcome this obstacle is to ask individuals to enter the information directly into the device/app, but this strategy <a href="https://googleblog.blogspot.com/2011/06/update-on-google-health-and-google.html">has not worked well in the past</a>.</li><li><strong>The notion that the patient is “the” health care consumer is a myth.</strong> That’s not to say that patients don’t have a say in purchasing decisions — we can clearly see that patients decide when and where they seek care. Even so, employers influence where <a href="http://kff.org/private-insurance/issue-brief/trends-in-employer-sponsored-insurance-offer-and-coverage-rates-1999-2014/">66% of non-elderly adults</a> get their care through the types of health insurance they offer their employees. Health insurers, in turn, steer patients toward certain clinicians (“in network”) and away from others (“out of network”). In all cases, clinicians ultimately control much of what gets purchased in all layers downstream of where they sit in the loop. This is because most health insurance will not pay for goods and services unless they are ordered or rendered by a trained and licensed doctor, nurse practitioner, or physician assistant. As a result, there is no such thing as a health care “consumer” — there are only stakeholders with varying degrees of purchasing authority.</li><li><strong>Clinicians are the most influential players in the loop. </strong>This follows from the observation that health benefits managers, insurance companies, and comparison-shopping websites derive their insights from claims data. As illustrated above, health care claims are derived from EHR data, which means that clinicians can exert control over this segment through a feedback loop. This is noteworthy because it effectively gives clinicians strategic leverage over the entire health tech economy.</li><li><strong>EHR companies are a direct threat to companies that sell function-specific products.</strong> Indeed, the top EHR companies make no secret of their intent to develop products that compete with those downstream. Some are already making their way onto the market, including care coordination products from <a href="http://www.cerner.com/solutions/extended_care_providers/">Cerner</a>, <a href="http://www.athenahealth.com/enterprise/athenacoordinator/patient-access">athenahealth</a>, and <a href="http://www.healthcareitleaders.com/blog/what-is-epic-healthy-planet/">Epic</a> just to name a few.</li><li><strong>Tech-enabled LMF companies need to develop strategic partnerships with clinicians in order to maximize growth.</strong> In the current environment, clinicians have the potential to function like lead gen engines for tech-enabled LMF companies. Just food for thought.</li></ul><p>The landscape is always shifting, and this is just a snapshot. In my next post, I’ll explore how data engineering decisions influence a company’s ability to adapt to shifts in the health tech economy.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aab5bbcdb9c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[3 things all health tech entrepreneurs need to know]]></title>
            <link>https://medium.com/@DataDrivenMD/3-things-all-health-tech-entrepreneurs-need-to-know-f81f783b9c01?source=rss-fd0e8cb79ac1------2</link>
            <guid isPermaLink="false">https://medium.com/p/f81f783b9c01</guid>
            <category><![CDATA[health-technology]]></category>
            <category><![CDATA[healthcare]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[startup-lessons]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Jorge A. Caballero, MD]]></dc:creator>
            <pubDate>Tue, 03 Jan 2017 20:34:09 GMT</pubDate>
            <atom:updated>2017-01-05T18:57:11.405Z</atom:updated>
            <content:encoded><![CDATA[<h3>The Great Recession sparked the health tech revolution</h3><p>Until very recently, paper was the primary storage medium for our nation’s healthcare system. In 2009 the American Reinvestment and Recovery Act (ARRA) earmarked $30B in federal funds to incentivize health care providers to adopt electronic health records (EHRs) in an effort to modernize our nation’s healthcare system. By most measures, this program was wildly successful: in 2008, only 9.4% of health care providers used EHRs, by 2015 EHR adoption jumped to 83.8%. Today, 96% of community hospitals rely on EHR technology.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*bEvczvAvVn0TqtDq." /><figcaption>Source: <a href="https://dashboard.healthit.gov/evaluations/data-briefs/non-federal-acute-care-hospital-ehr-adoption-2008-2015.php"><em>Adoption of Electronic Health Record Systems among U.S. Non-Federal Acute Care Hospitals: 2008–2015</em>.</a></figcaption></figure><p>From an insider’s perspective, EHR adoption felt like it progressed at breakneck speed. In 2010 I was a clinical intern at Santa Clara Valley Medical Center (SCVMC) where I spent hours handwriting patient notes and relied on a pneumatic tube system to send orders from one wing of the hospital to another. In 2011 — then an anesthesia resident at Stanford Health Care (SHC) — I used an EHR to place orders for patients in the ICU but spent the bulk of my time in the operating room keeping meticulous records of my patients’ heart rate, blood pressure, and respiratory rate using pen and paper. By the time I finished my training in 2014, the housestaff at SCVMC and SHC were relying almost exclusively on EHRs to fulfill their clinical duties.</p><p>EHR transition teams were lead by “clinical champions” and “super users” — physicians and nurses with backgrounds in clinical informatics and/or EHR-specific training and certification. Once EHRs were in the picture, these hybrid clinical-technical leaders became increasingly influential figures within their respective health care organizations, and many went on to serve as <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1561791/">Chief Medical Information Officers</a>.</p><p>Today, clinicians are more involved than ever in all matters health IT. CMIOs are business leaders that play a key role in decisions involving software licensing and data governance. This means that companies whose software generates, processes, or consumes clinical data need to adopt sales, marketing, and engineering strategy that addresses the broad set of concerns raised by clinical leaders. Doing this effectively requires a deep understanding of how policy translates not only to health IT needs, but also into practice. It also underscores the significance of the ARRA’s EHR incentive as a quintessential case study for all health tech entrepreneurs.</p><h3>Meaningful Use is a clinician’s enemy, and an entrepreneur’s best friend</h3><p>In order to protect the government’s investment in health IT, lawmakers require health care providers to use federal incentives toward EHR software that “offers the necessary technological capability, functionality, and security to help them meet <a href="https://www.healthit.gov/policy-researchers-implementers/meaningful-use-regulations">Meaningful Use criteria</a>.” EHR software that meet these criteria is known as <a href="https://www.cms.gov/regulations-and-guidance/legislation/ehrincentiveprograms/certification.html">certified EHR technology</a> (CEHRT). The Office of the National Coordinator for Health IT (ONC) is the entity that is responsible for managing the EHR incentive and CEHRT programs on behalf of federal lawmakers. MU criteria are the knobs and levers of authority granted to the ONC to achieve their legislative mandate.</p><p>Since the EHR program’s inception, MU requirements have drawn criticism from health care providers. They argue that ongoing changes make compliance into a moving target. They also criticize requirements for being exceedingly specific, painfully vague, or costly to implement. The following examples illustrate these points:</p><ul><li>“For 2016, a [health care provider]<a href="https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Downloads/2016EP_9SecureElectronicMessagingObjective.pdf"> must send a secure message</a> using their [EHR’s] electronic messaging function to at least one patient (or the patient-authorized representative) the [health care provider] sees during the EHR reporting period, or must respond to a secure message that the patient (or the patient-authorized representative) has sent during the EHR reporting period.”</li><li>“The [health care provider] is in active engagement with a public health agency to submit immunization data”</li></ul><p>In both cases, the first challenge to achieving compliance is ensuring that one’s existing IT stack is able to capture and transmit the necessary data. The second challenge is training clinical staff to use new software and/or features — a process that often involves altering clinical workflows to accommodate the need for proper documentation. <a href="http://annals.org/aim/article/2546704/allocation-physician-time-ambulatory-practice-time-motion-study-4-specialties">Given that doctors already spend twice as long entering data into EHRs as they do seeing patients</a>, it’s no wonder that health care providers have grown weary of ever-evolving MU criteria. But it is precisely because MU requirements create pain points for clinicians that they are an entrepreneur’s best friend.</p><p><strong>Simply put: Meaningful Use is a category creator.</strong> Nearly every doctor in America has to comply with MU, this is a fact that all-but-guarantees a large market for a broad spectrum of health tech products. The challenge for entrepreneurs is therefore to deliver software solutions that help clinicians achieve MU compliance without a) disrupting clinical workflows or b) creating administrative overhead.</p><p>I should note that <a href="http://www.healthcareitnews.com/news/meaningful-use-will-likely-end-2016-cms-chief-andy-slavitt-says">some in the industry speculate that Meaningful Use is on its way out</a> while <a href="http://www.healthcare-informatics.com/article/demise-meaningful-use-not-so-fast-say-industry-leaders">others argue that it is here to stay</a>. The reality is likely somewhere in between. Even if the MU program ceases to exist in name, it is likely to persist in practice. There are simply too many stakeholders that benefit from the role that MU plays in driving health IT policy, not the least of which is Congress. In the last two years alone, Congress has leveraged the information gleaned through the MU program to craft two major health care laws: MACRA (Apr 2015) and the 21st Century Cures Act (Dec 2016). Both laws were passed with overwhelming bipartisan support. So, if recent history is any indication, MU (or a functionally-similar program by a different name) will continue to play a key role in health IT policy and category creation well into the future.</p><h3>Electronic health records are the solution to (and cause of) many problems</h3><p>The transition to EHRs was not easy. Data entry proved to be a particularly stressful pain point as clinicians struggled to write patient notes and place orders. EHR vendors helped to ease the transition by adding features that allowed clinicians to create note templates and to bundle related orders into “order sets.” These features made data entry easier than ever, but they also gave rise to a wave of problems related to information retrieval.</p><p>While note templates streamlined clinical documentation, they did so at the expense of clarity in communication. If you were to review the daily notes written by a single doctor for a given hospital patient, you’d discover that the notes differed by no more than a few sentences here and there. Moreover, if you reviewed all the notes written by a single doctor for all of their patients, you’d find that 70–90% of the content is nearly identical. The problem with this practice is that it drowns salient points in a sea of redundant content. This makes it increasingly difficult for anyone other than the note writer to answer simple questions such as:</p><ul><li>Why was <em>this</em> patient seen by <em>this</em> clinician? Was it at the request of another clinician? If so, what was the specific clinical question that was beyond the scope of training/experience of the referring doctor?</li><li>What did <em>this</em> clinician do for <em>this</em> patient on <em>this</em> occasion? How much of the copy-and-paste content relates to this interaction vs. yesterday’s vs. last week’s vs. last month’s?</li><li>Is the author of the note the clinician best suited to oversee this patient’s care from this point forward? Are they actively involved in the patient’s care <em>today</em>, or has primary responsibility for this patient been transferred to another clinician since the note was written?</li></ul><p>These are examples of real-world information gaps that reflect a systemic problem rooted in clinical data systems that are poorly interoperable. <a href="http://www.nationalacademies.org/hmd/Reports/2012/Best-Care-at-Lower-Cost-The-Path-to-Continuously-Learning-Health-Care-in-America.aspx">A report commissioned by the Institute of Medicine</a> concluded that “30 percent of health spending in 2009 — roughly $750 billion — was wasted on unnecessary services, excessive administrative costs, fraud, and other problems”. <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3651728/">Related studies</a> pinpoint the source of waste to technical barriers limiting clinicians’ access to a patient’s complete medical history. The importance of these findings cannot be understated: <strong>not only are information gaps exposing patients to unnecessary harm, they contribute substantially to wasteful health care spending</strong>.</p><p>Along the same line, order sets solve one problem while creating another. In theory, order sets are a convenient way of ensuring that clinicians adhere to well-established guidelines when managing the care of patients with conditions such as shock, chest pain, and stroke. In practice, the data that speaks to the intent of each order is separated from data about the order itself. Lacking the appropriate clinical context, it’s difficult if not impossible to safely interpret study results, and to measure clinical outcomes in a meaningful way.</p><p>A common example of the problem created by order sets involves the use of Heparin — a blood thinning medication — in conjunction with a lab measurement known as “PTT”, which is used to monitor whether the patient is receiving the appropriate dose of the drug. Heparin is indicated in patients suffering from a heart attack, certain types of stroke, and other conditions. It is <a href="http://circ.ahajournals.org/content/107/23/2884">critically important to maintain the patient’s PTT within a desired range</a>, but the <a href="http://www.vchca.org/docs/hospitals/heparin-protocol-vault-(1).pdf?sfvrsn=0">target range varies according to the drug’s indication and the patient’s condition</a>. A PTT value below the desired range indicates that the patient is at risk for ongoing damage to vital organs and/or death. When PTT values are above the safe threshold, patients are at increased risk for developing internal bleeding. Although leading hospital EHRs ship with pre-built order sets to help manage Heparin dosing, they make it incredibly difficult to understand <em>why</em> a patient is receiving heparin and/or undergoing PTT testing. This is problematic when multiple doctors are involved in the care of the same patient (a very common scenario) because each doctor is making clinical decisions based on their interpretation of the patient’s condition.</p><p>Barriers to information retrieval caused by EHRs affect other products in the health tech economy. For example, digital and consumer health companies will need to find ways to surface their data through existing EHR systems. Meanwhile, companies that are building enterprise software — especially those involving precision medicine, population health, and care coordination — will need to engineer their products to be compatible with the dizzying matrix of clinical software that comprise health IT stacks.</p><p><strong>Companies that fail to mitigate the impact of clinical information gaps risk losing their strategic position in the market.</strong> I expand on this idea in another post: <a href="https://medium.com/@DataDrivenMD/the-health-tech-economy-aab5bbcdb9c#.bvbrjsyzq"><em>The Health Tech Economy</em></a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f81f783b9c01" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>