Artificial intelligence (AI) has taken the healthcare industry by storm with its promises for more accurate, timely predictions, process automation, and precise decision support for clinicians and administrators alike.  

So far, AI is certainly <a href="https://www.dhinsights.org/news/mayo-clinic-is-using-ai-to-solve-the-unstructured-data-problem-for-all">proving its value</a> across a number of important use cases – but it’s also showing critical vulnerabilities that need to be addressed sooner rather than later.  Chief among these potential problems is unintentional bias in the results.  

Bias occurs when the training data sets for AI algorithms are incomplete, low quality, or skewed in a particular direction.  This can lead to questionable outputs and suboptimal decisions for patients and health systems.

<hr />

Related story: <a href="https://www.dhinsights.org/news/mayo-clinic-is-using-ai-to-solve-the-unstructured-data-problem-for-all">Mayo Clinic is using AI to solve the unstructured data problem for all</a>

<hr />

How can healthcare leaders prevent potential biases from influencing decision-making and perpetuating health inequities?
<h2>How does bias creep into AI algorithms?</h2>
All artificial intelligence algorithms are man-made by human computer scientists who develop, train, and optimize their applications for real-world use.  

Humans bring with them a complex universe of personal experiences, opinions, skills, and predispositions that make it impossible to be truly objective or one hundred percent accurate in their assigned tasks.  

These “imperfections” are propagated through the healthcare system, from the clinician who sits down at the keyboard to record patient data to the computer scientist who curates the training dataset to the CIO who decides that an algorithm is ready to deploy in her organization.

Sometimes, undesirable results are just an issue of poor programming and can be fixed with some tweaks to the code.  But other times, the biases go deeper and can start to <a href="https://www.dhinsights.org/news/building-techquity-into-the-digital-healthcare-delivery-system">influence – or exacerbate – health inequities</a> among the patients they are supposed to serve. 

<hr />

Related story: <a href="https://www.dhinsights.org/news/building-techquity-into-the-digital-healthcare-delivery-system">Building techquity into the digital healthcare delivery system</a>

<hr />

<a href="https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0000022" target="_blank" rel="noopener">One recent study</a> in PLOS Digital Health examined more than 7000 academic papers about AI algorithms and found that the 40 percent of databases came from the US and 13 percent from China. Authors were primarily from China (24 percent), the US (18 percent), or other high-income countries. Sixty percent of first authors were statisticians, not clinicians. And close to three-quarters were male. 

When perspectives and datasets are overly limited, AI tools become vulnerable to algorithmic bias, <a href="https://www.hsph.harvard.edu/ecpe/how-to-prevent-algorithmic-bias-in-health-care/" target="_blank" rel="noopener">defined by a team of Harvard researchers</a> as “the application of an algorithm that compounds existing inequities in socioeconomic status, race, ethnic background, religion, gender, disability, or sexual orientation and amplifies inequities in health systems.”

For example, <a href="https://www.biorxiv.org/content/10.1101/826057v1" target="_blank" rel="noopener">a Ugandan research team</a> found that neural networks used to identify skin lesions were mostly trained on images of white patients, with Black individuals comprising just 10 percent of the training data. Since skin lesions often appear very differently on darker skin, the resulting algorithm may not accurately identify issues in non-white patients.
<h2>Solutions for preventing built-in biases</h2>
As health equity takes center stage for the healthcare industry, organizations must be sure that they are using their growing collection of AI algorithms in a positive manner. Fortunately, the vast majority of executive leaders – 94 percent in <a href="https://www.unitedhealthgroup.com/newsroom/2021/2021-12-15-optum-ai-survey-for-better-equitable-affordable-outcomes.html" target="_blank" rel="noopener">a recent Optum poll</a> – acknowledge the duty to use AI responsibly and are already putting plans in place to do so.

Developers and users must acknowledge and address their potential biases when architecting and implementing AI algorithms in the real-world care setting.

An <a href="https://www.sciencedirect.com/science/article/pii/S2666389921002026" target="_blank" rel="noopener">international team of experts</a> suggests a three-pronged approach to creating an equitable and transparent AI ecosystem:

Engaging in patient-centered, participatory development of AI tools: Computer science teams should include perspectives from diverse communities, including patient representatives, clinicians, and other relevant parties, when testing and refining their algorithms.  

Ensuring responsible data sharing and inclusive data standards: The industry should work together to develop common metrics for AI reliability, consider launching clinical trials to objectively test AI tools, and establish minimum standards for the diversity and inclusivity of data 

Sharing code and prioritizing explainability: AI code doesn’t always have to be open source, but developers should consider integrating outside perspectives into the design and training process. AI algorithms should also be explainable, as much as possible, and have clear mechanisms for tracing data provenance and usage.

By integrating more inclusive data into AI training sets and establishing industry-wide protocols for transparency and reliability, the healthcare industry can avoid unwanted bias, close gaps in health equity, and ensure that all AI tools provide optimal decision support when put into practice. 

&nbsp;

<hr />

Jennifer Bresnick is a journalist and freelance content creator with a decade of experience in the health IT industry. Her work has focused on leveraging innovative technology tools to create value, improve health equity, and achieve the promises of the learning health system.

AI algorithms show a race problem

There's more than just money on the line when healthcare organizations choose their technology partners. New health IT tools frequently bring disruption to workflows and experiences – and it's a CIO's job to ensure that the disruption is beneficial instead of otherwise.

In a panel discussion at <a href="https://www.viveevent.com/" target="_blank" rel="noopener">ViVE 2022</a> in Miami Beach, CIOs and venture experts shared their strategies for identifying promising health IT solutions, measuring the value of potential partners, and fitting new tools into existing health IT ecosystems.
<h2>Build or buy tools?</h2>
To identify the right opportunities, leaders must look at their organization's existing strengths, weaknesses, and workflow gaps.

"Even if you have a large platform EHR, there are typically nooks and crannies where systems don't function well or can't talk to each other," said Dr. Neal Patel, Chief Information Officer at Vanderbilt University Medical Center. "We try to solve those problems by finding an interesting startup or developing our own tools, if that's a viable path to finding the right solution."

Aaron Martin, EVP and Chief Digital Officer at Providence Health in Washington State, takes a similar approach. "We start off with a prioritized set of opportunities to move the needle. Then, we ask if we already own a potential solution. If not, we scan the marketplace for companies that are working on the problem. If we don't have it and can't find it out there, we will develop it ourselves in our Venture Studio." 

"Most health systems are fundamentally fighting the same battles and looking for similar solutions," he continued. "That creates opportunities for these companies to establish themselves in the larger market, which is valuable for all of us."  
<h2>Ensuring shared value in a crowded marketplace of ideas</h2>
Venture investors and product purchasers alike need to be clear about any new tool's purpose and potential value, said Tressa Springmann, Senior Vice President and Chief Information and Digital Officer at LifeBridge Health.

"When looking for potential companies to work with, both parties must be transparent and upfront about what the partnership will look like, how it will work, and who has control over intellectual property," she asserted. "Getting that on the table early is critical, because not knowing everything is a risk that healthcare systems can't afford to take."

Hard data can mitigate those risks to some extent, commented Martin. "I won't let a team start a pilot unless they have a clear line of sight into key performance metrics," he said. "Just saying 'we talked to some people and it sounds like it's working' isn't good enough. Meaningful data helps us course correct before it's too late." 

Healthcare organizations must also look at the idea of "value" from multiple perspectives, Patel added. "As a CIO, I'm not just looking for something that will make workflows easier. I need to invest in products that will reduce our overall technical debt."

"Is there something else I can turn off because this is better? Better yet, are there two things I can get rid of? We can't just keep accumulating things – we have to reduce underlying burdens and ongoing maintenance costs. The best investments check all of those boxes."
<h2>Managing disruption to achieve long-term ROI</h2>
Disruption is a frequent byproduct of <a href="https://dhinsights.org/news/overcoming-3-common-digital-transformation-pitfalls/" target="_blank" rel="noopener">implementing new solutions</a>, even when that new tool is intended to improve workflows and experiences. When discontent is a possibility, a champion for change can be key for a successful implementation that leads to lasting ROI, said Springmann.

"Changing the status quo usually comes with a cost," she acknowledged. "But if you make a change that people can get behind, the cost isn't as high. If you have an incredibly enthusiastic internal stakeholder and you believe in their ability to successfully champion change, it can make the process must smoother so you can see faster, better returns."

"Whether you're building a tool from scratch or partnering with a developer, it's always important to engage your stakeholders and get their feedback so that you can be confident in your decision making and achieve the ultimate goals you have in mind."

Read more of our ViVE event coverage:
<ul class="ul1">
 	<li class="li1"><a href="https://www.dhinsights.org/news/empowering-nurses-to-lead-healthcares-digital-revolution">Empowering nurses to lead healthcare’s digital revolution</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/oncs-2022-agenda-focuses-on-interoperability-to-enable-the-greater-good">ONC’s 2022 agenda focuses on interoperability to enable the greater good</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/how-to-move-beyond-cybersecurity-compliance-in-healthcare-focus-on-clinician-workflows">How to move beyond cybersecurity compliance in healthcare? Focus on clinician workflows</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/in-a-covid-19-world-digital-transformation-of-the-workforce-goes-beyond-clinical-tools">In a COVID-19 world, digital transformation of the workforce goes beyond clinical tools</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/embedding-dei-initiatives-into-healthcare-talent-recruitment">Embedding DEI initiatives into healthcare talent recruitment</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/building-virtual-care-infrastructure-begins-with-removal-of-perceived-barriers">Building virtual care infrastructure begins with removal of perceived barriers</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/sutter-health-staying-agile-to-improve-patient-experience-with-digital-automation">Sutter Health staying agile to improve patient experience with digital automation</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/mayo-clinic-is-using-ai-to-solve-the-unstructured-data-problem-for-all">Mayo Clinic is using AI to solve the unstructured data problem for all</a></li>
</ul>
&nbsp;

<hr />
Jennifer Bresnick is a journalist and freelance content creator with a decade of experience in the health IT industry.  Her work has focused on leveraging innovative technology tools to create value, improve health equity, and achieve the promises of the learning health system.

Maximizing ROI in digital health tools

At <a href="https://www.viveevent.com/agenda" target="_blank" rel="noopener">ViVE 2022</a> and the CHIME Spring Forum in Miami Beach this week, thousands of health IT professionals are roaming the convention center’s halls. With the vaccinated and tested attendees largely going mask-free, the event has the same energy and vibrancy as a pre-pandemic gathering.

But no one is at risk of forgetting the enormous toll of COVID-19 or how it has irrevocably changed the way healthcare organizations operate. The pandemic is at the forefront of everyone’s thoughts, especially as CIOs and others gather to envision <a href="https://www.dhinsights.org/categories/digital-transformation" target="_blank" rel="noopener">the future of healthcare</a> in 2022 and beyond.

The opening keynote session, moderated by Andy Crowder CHCIO, SVP Chief Information and Analytics Officer at Atrium Health, began with a moment of reflection on the past two years and continued with a deep exploration of how the pandemic will shape the digital transformation of the healthcare workforce for the foreseeable future.

“Digital transformation is well underway in all areas of care, but it’s also happening behind the scenes,” said Stephanie Lahr, MD, CHCIO, CIO and CMIO at Monument Health. “The reality is that for many years, technology was seen as adding tasks that didn’t really make people’s lives easier. But now, we’re approaching a place where all our tools are becoming more mature and exciting.”

“Our <a href="https://www.dhinsights.org/news/mayo-clinic-is-using-ai-to-solve-the-unstructured-data-problem-for-all" target="_blank" rel="noopener">workforce</a> is in the middle of a huge transition, so we need to use these tools to reduce friction and make it easier for people to do their work,” she continued. “Having people work from home is one thing, but creativity in digital transformation is going to be about a lot more than where you work. It’ll be about when, how long, and what kind of work people are doing.”

Avoiding burnout in this new environment is a top priority. With the ongoing stresses of the pandemic, burnout is just around the corner for too many employees – and senior leaders are no exception, added Angela Diop, ND Vice President of Information Systems for Unity Health Care.  

“COVID-19 has changed the way we do business,” she stated. “When I look at 2022, we all need to refresh ourselves. That includes CIOs and IT leaders. We need to remember the trauma we’ve been through and do what we can to revitalize ourselves and our staff through more meaningful, engaging connections.”  

The key to success with a new remote workforce is ensuring employees know when and how to ask for help, asserted Zafar Chaudry, Senior Vice President and Chief Information Officer at Seattle Children's.

“We have invested in all of these digital tools to enable productivity and communicate with each other, but people are still feeling overworked and overwhelmed,” he said. “We can’t forget that not everyone has a dedicated home office and the luxury of an uninterrupted workday. Some people are working from dining rooms or closets, homeschooling their kids, and taking care of their dogs. There’s an inequity there that we don’t talk about enough.”

“Employees are very good at saying they have too much work, but they’re not very good at asking for specifics to help them with it,” he said. “Leaders have to take the first step by exposing their own feelings and being open to sharing where they are with burnout, otherwise employees won’t do the same.”

To successfully navigate the new hybrid working environment, organizational leaders will need to make a concerted <a href="https://www.dhinsights.org/news/4-ways-to-manage-provider-burnout-during-the-covid-19-pandemic" target="_blank" rel="noopener">effort to change the culture around burnout</a>, leveraging a combination of innovative health IT tools and personal connections.

“Our direct reports might not be used to talking with us about burnout, but we have to change that,” said Lehr. “We have to have those conversations with our colleagues and partners every day so that we can move forward. If we can merge workforce engagement strategies with automation and decision support tools, we will be able to meet the challenges of 2022 for our employees while prioritizing top-quality patient care.”

Read more of our ViVE event coverage:
<ul class="ul1">
 	<li class="li1"><a href="https://www.dhinsights.org/news/empowering-nurses-to-lead-healthcares-digital-revolution">Empowering nurses to lead healthcare’s digital revolution</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/oncs-2022-agenda-focuses-on-interoperability-to-enable-the-greater-good">ONC’s 2022 agenda focuses on interoperability to enable the greater good</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/how-to-move-beyond-cybersecurity-compliance-in-healthcare-focus-on-clinician-workflows">How to move beyond cybersecurity compliance in healthcare? Focus on clinician workflows</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/maximizing-the-roi-of-investments-in-healthcare-digital-transformation-tools">Maximizing the ROI of investments in healthcare digital transformation tools</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/embedding-dei-initiatives-into-healthcare-talent-recruitment">Embedding DEI initiatives into healthcare talent recruitment</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/building-virtual-care-infrastructure-begins-with-removal-of-perceived-barriers">Building virtual care infrastructure begins with removal of perceived barriers</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/sutter-health-staying-agile-to-improve-patient-experience-with-digital-automation">Sutter Health staying agile to improve patient experience with digital automation</a></li>
 	<li class="li1"><a href="https://www.dhinsights.org/news/mayo-clinic-is-using-ai-to-solve-the-unstructured-data-problem-for-all">Mayo Clinic is using AI to solve the unstructured data problem for all</a></li>
</ul>
&nbsp;

<hr />

Jennifer Bresnick is a journalist and freelance content creator with a decade of experience in the health IT industry. Her work has focused on leveraging innovative technology tools to create value, improve health equity, and achieve the promises of the learning health system.

Digital transformation in a COVID-19 world

“If data is inaccurate, or incomplete, it could lead to significant issues that affect your organization on every level. The provider won’t be able to trust the chart – and proper documentation, downstream business processes and ultimately patient care will be affected.”

Download this report to learn the 3 critical steps you need to take during a data migration in order to minimize disruptions to provider workflow and maximize business continuity.

3 steps to maintaining data integrity during EHR migrations

As healthcare organizations move toward the future of health, unlocking enterprise innovation and transformation through cloud migration will be the key to success. Moving to the cloud’s resilient, scalable technology platform can open the door to new growth pathways and expand clinical and business value in the process.
<div class="page" title="Page 6">
<div class="layoutArea">
<div class="column">

Starting with a cloud strategy and blueprint enables a cohesive migration approach that aligns with the organization’s strategic priorities. Implementing a cloud without a clearly defined path forward may result in a lack of organizational buy-in, an inability to maximize efficiencies/cost optimization opportunities, and functions treating the cloud as merely another data center instead of embracing it as a catalyst for change.

</div>
</div>
</div>
Explore strategy and blueprint for healthcare organizations in this new report.

Unlocking enterprise innovation in the cloud

Over the past decade, sectors across the health care industry have invested considerably in virtual and digital health. The area is poised for tremendous growth, as COVID-19 accelerated adoption and incubated vast opportunities for virtual health to improve access to services while delivering safe and convenient care.

In this report, you'll meet the equity imperative in virtual health care through intentional design and deployment.

Using virtual health to push equity in healthcare

ViVE 2023 will be March 26-29 at The Music City Center in Nashville to continue our mission of transforming the business of healthcare.

ViVE 2023

ViVE 2023 - Digital Health Insights

Home

Digital Health Insights is the healthcare industry’s most trusted source for data, analytics, and executive knowledge.

Digital Health Insights

Houston Methodist is betting big on the combination of automation technologies and person-centered strategies to create the smartest hospital yet.

What does a smart hospital of the future look like?

What does a smart hospital of the future look like? - Digital Health Insights

https://secure.gravatar.com/avatar/a63df913c73bc7ed7bc3489063c85afb?s=96&d=mm&r=g

Jennifer Bresnick

The “hospital at home” model is gaining traction with telehealth technology. How can providers deliver personalized care at-home?

The “hospital at home” model is gaining traction as telehealth technologies reach maturity. How can providers use remote high-acuity care to deliver personalized care when and where it’s needed?

Bringing the power of hospitals to homes with high-acuity telehealth

Bringing hospital to homes with telehealth - Digital Health Insights

Bringing hospital to homes with telehealth

The “hospital at home” model is gaining traction as telehealth technologies reach maturity.

The digital ecosystem needs nurses so it’s time they take a more prominent leadership role in the health IT revolution.

Nurses lead healthcare’s digital revolution

Nurses lead healthcare’s digital revolution - Digital Health Insights

Security leaders say cybersecurity compliance programs should center around clinician workflows and focus on efficiency and patient care.

Clinicians are highly resourceful, making decisions based on efficiency and what’s best for patient care. At ViVE on Monday, security leaders note security programs should center around those important workflows.

Clinician workflow is key to cybersecurity

Clinician workflow is key to cybersecurity - Digital Health Insights

https://secure.gravatar.com/avatar/6b68171fb2632207139948ce8fddc65b?s=96&d=mm&r=g

Jessica Davis

The Office of the National Coordinator shepherds the post-pandemic healthcare industry toward greater interoperability.

The Office of the National Coordinator continues to shepherd the industry toward greater interoperability with a focus on shaping an equitable, accessible post-pandemic healthcare system.

ONC 2022 agenda: Interoperability

ONC 2022 agenda: Interoperability - Digital Health Insights

How can healthcare organizations make the most of limited resources when investing in tools to further their digital transformation efforts?

Maximizing ROI in digital health tools - Digital Health Insights

Facing healthcare post-pandemic, CIOs must consider digital transformation for all of their staff members, not just clinicians.

Digital transformation in a COVID-19 world - Digital Health Insights

In a COVID-19 world, digital transformation goes beyond clinical tools

ViVE preview: Ascension’s Dr. Tania Elliott champions a care strategy that develops a virtual infrastructure across health systems.

Building virtual care infrastructure

Building virtual care infrastructure - Digital Health Insights

https://secure.gravatar.com/avatar/cea3570f0aaa87a8d474685e0ac55f1e?s=96&d=mm&r=g

Jacqueline Renfrow

Building virtual care infrastructure must remove 'perceived' barriers

A mix of high-touch patient experience with strategic chatbot implementation is a winning recipe to improve care delivery.

ViVE preview: A mix of high-touch patient experience with strategic chatbot implementation is a winning recipe to improve care delivery.

ViVE 2023

The Latest

Subscribe

Resources

Explore

REACH OUR AUDIENCE

Featured Topics