Artificial intelligence has many benefits for senior living providers, including streamlining workflows and improving resident care, but it’s not without its drawbacks. That’s according to an expert who warns that technology and AI have the potential to increase inequities and mistrust in senior living and healthcare.
Jay Bhatt, geriatrician and managing director of the Center for Health Solutions and Health Equity Institute at Deloitte, says that tech-based healthcare inequities occur in three major areas: race and ethnicity data, infrastructure (in terms of broadband access and the digital divide) and tech literacy and engagement.
For example, researchers have found some preexisting algorithm biases in the overall healthcare system, such as tests for kidney disease, and racial bias can also be embedded in AI tools for speech and facial recognition, which can exacerbate inequities, Bhatt says. Meanwhile, prediction algorithms can sometimes have inaccurate results that affect healthcare treatment outcomes, he adds. The danger, then, is that senior living and other healthcare providers are making treatment decisions based on inaccurate data.
“[There’s] fear of how it further drives mistrust and misinformation in a world that’s really struggling with that,” Bhatt told HIMSS Media’s MobiHealth News. “We often say that health equity can be impacted by the speed of how you build trust, but also, more importantly, how you sustain trust. When we don’t think through and test the output and it turns out that it might cause an unintended consequence, we still have to be accountable to that.”
His advice for senior living and healthcare providers in solving these challenges is to be agile and strategic about the ways they implement AI tools, with proper training for staff, as well.
Biases have been apparent in AI and healthcare for some time now, experts have pointed out. Potential healthcare iniquities and racial biases were found in a recent study in the journal Nature Medicine. Researchers said that AI models demonstrated biases due to the underrepresentation of certain demographic groups in the datasets that were used in their development.
The cost of health inequities in the US is high, with one recent study estimating the economic burden of racial and ethnic health inequities at $421 billion.
