December Insights: The AI Adoption in Healthcare and the Challenge of Equitable Care

👋Hello Readers

Welcome to the December edition of the E&E newsletter. To make it easier for you to find us, we have taken our newsletter to LinkedIn. You can now easily read every post on our website in your browser or mobile device using the app. This month, we’re excited to share our first feature insight article. 

As always, continue to engage with us and subscribe to the newsletter.

🌟Feature Insight: Gen AI adoption in healthcare

As the holidays fast approach we’re setting aside time from thinking about holiday menus and shopping to start a conversation with you about some of the implications of Generative AI (Artificial Intelligence) usage in patient care delivery. Did you know that AI has been explored as a tool in healthcare since the 1950s? It started with an introduction to the concept of machine learning. Then in the 1960s and 1970s basic exploratory efforts began which have now culminated to the rapid advancements we’ve seen in the 21st century. 

📈 Healthcare AI Revolution

In the last three years since OpenAI launched ChatGPT,  the healthcare industry has been leading trends in generative AI adoption. Recent studies have confirmed that the shift is quickly occurring. 

• 85% of healthcare leaders surveyed by McKinsey in 2024 were already implementing or had plans to implement AI in their operations. 

• 71% of surveyed facilities are using some form of AI for patient visits, according to an August MGMA Stat poll 

AI boasts a future of more personalized and proactive care, and its ability to analyze vast amounts of data has positioned it to be implemented throughout healthcare delivery and medical operations. 

🤔The Crucial Balance: Opportunity and Challenges

In August of this year, Epic announced that it is creating AI resources targeted towards patients and clinicians at its Users Group Meeting. These tools will offer patients the ability to schedule appointments via text message, answer billing questions, and interpret lab results all within their patient portal. As the industry celebrates the advent of the AI scribe, we should not forget who remains at the core of this technology. Advances in AI have shown great success in improving diagnostic capabilities, aiding clinician decision making, reducing administrative burden and most importantly it is believed that it can improve patient engagement and outcomes. These applications place AI at a critical point in healthcare delivery; primary care. This is an essential setting that is very often the gateway to care delivery and at the same time can exacerbate health disparities. Research consistently shows that despite having “equal” access to primary care, patients from ethnic minority backgrounds, lower SES groups and living in rural areas experience worse health outcomes. Incorporating AI tools into the primary care setting creates an opportunity to democratize health care access and address some of the root causes of inequality in the system. The AI scribe is just one avenue, AI powered tools such as telemedicine platforms, diagnostic tools, and even NLP based tools like chat bots have already begun to improve patient engagement and in turn patients’ experience within the system. 

While we have the unique opportunity to address some of the root causes of inequality using AI-powered tools, there exist significant challenges that could actually exacerbate existing health disparities:

• Algorithmic bias through lack of representation of ethnic minorities in data powering AI development

• Slow technology uptake, digital divides, and lack of technological literacy

• Community knowledge and beliefs, lack of trust in the system, and privacy concerns

Healthcare organizations must factor these challenges into the research and development of AI-powered tools. There are already several examples of tools being deployed in the health system that later lead to unintended consequences such as misdiagnosing or under-diagnosing individuals from racial minority groups (Obermeyer et al, 2019). This is just one example that highlights the need for model bias mitigation, better training sets, and ongoing monitoring. 

This is a critical and extremely exciting moment in our history and we believe that it is crucial to bring our readers along. We will be turning this topic into a series where we continue to explore the ethical concerns of AI-powered tools, address some of the challenges, present and review real life cases of implementation and share our thoughts on what ethical implementation could look like. So stay tuned and make sure you subscribe so that you get every issue straight to your inbox.

💡Moving Forward 

AI has the potential to transform healthcare delivery, especially through primary care implementation. We have seen great success with Telehealth as a vessel for increased healthcare access across the US, and AI can easily serve the same purpose. As these tools advance, it is important for developers and health care administrators to ensure that they are equitable to all patients who use them. We’re looking forward to exploring this exciting topic with you.

Trends to watch

• The Alarming Reversal in Childhood Mortality

• Controversial Change to the Infant Hepatitis B Vaccine Schedule

  
  

💬Engage With Us

What has been your experience with AI in healthcare? 

Join the conversation on LinkedIn 

Share this newsletter with a friend or colleague who’d find it valuable

Subscribe here

Explore our website

Until then, stay connected!