Here’s something we can all agree on: We’re getting older. Globally, the number of people over 65 will more than double by 2050.1 One in every three people die with some form of dementia.2 The number of people living with the disease doubles every 5 years beyond age 65.3 This means the majority of us know someone living with or caring for someone with Alzheimer’s disease and related dementias (AD/ADRD), Parkinson’s, or other functional impairments from neurodegenerative disease.
Key questions remain: What happens to the aging brain as dementia sets in? How should we best care for people experiencing cognitive decline, and for those caring for their loved ones?
Urgency for answers feeds into the great and growing promise of artificial intelligence (AI). As buzz grows to a fever pitch, patients, caregivers, clinicians, researchers, innovators, investors, and federal officials are all eager to understand the implications: How can today’s panoply of datasets, devices, algorithms, and emerging technologies help improve healthy aging?
On March 19-20, 2024, the a2 Collective, a research program funded by the National Institute on Aging (NIA), part of the National Institutes of Health (NIH), convened its second annual a2 National Symposium to discuss progress, opportunities, and challenges in the design, implementation, evaluation, and commercialization of AI and other emerging technologies for healthy aging and caregiver support. The a2 Collective represents the Artificial Intelligence and Technology Collaboratories (AITC) for Aging Research program, through which NIA plans to award at least $40M over a 5-year period to fund AgeTech pilots with the potential to improve the lives of older adults, including individuals with AD/ADRD, and their caregivers. The symposium was hosted by PennAITech on the University of Pennsylvania campus in Philadelphia, PA, and attended by nearly 200 in-person and more than 150 virtual participants.
Generative AI as a foundation model for innovation
As AI hype reaches new heights, translating potential into reality starts with building context. A trio of keynote speakers and invited panelists with expertise in health technology, venture capital, and entrepreneurship offered their appraisal of today’s excitement about generative AI and its myriad implications for aging.
Joel Dudley, PhD, Innovation Endeavors, described the “super evolution” underway in life sciences, where new potential has rapidly emerged with generative AI. Technology advances gradually, and then suddenly, he observed. GenAI itself can function as a foundation model, akin to an operating system, whose value is realized by the applications and ecosystems built on top of underlying technology.
GenAI platforms are like “big brains” trained on vast data from the internet and beyond, with better understanding of users and the ability to generate human-like text as well as recognize images, spoken words, and more. The added context these systems can absorb will help innovations “increase the probability of translational success.”
Flywheel models that cyclically “design, build, test, learn” can compound advancement, such as by creating rapid learning systems in biology that enable generative chemistry for de novo molecules and protein designs that don’t require invasive, expensive, and timely human screenings or sample collection. GenAI could help devise new solutions in engineering, drug delivery, trial design, study review, compliance in manufacturing, reporting, and more.
Suchi Saria, PhD, Bayesian Health and Johns Hopkins University, offered a grounded perspective, suggesting that healthcare lacks incentives more than technology. Medicine’s complexity, and the “translation gap” between testing new ideas, gaining clinical and community adoption, and successful commercialization should not be underestimated. Saria emphasized the need for not just smart ideas, but smart ideas that actually scale.
GenAI could vastly improve efficiency in circulating validated information among different stakeholders through cloud-based, configurable approaches. Successful innovators will build businesses rooted in pre-existing workflows, datasets, and reimbursement pathways that subsidize technology’s development.
In a panel on GenAI and aging, Oded Nov, PhD, New York University, and Ross Koppel, PhD, University of Pennsylvania, underscored the importance of human agency, trust, and interdisciplinary design in building GPT-style chatbots for clinical advice, decision making, and safety. Others throughout the symposium weighed in on GenAI’s potential to create better caregiving and clinical tools, as well as ease administrative burdens by streamlining reporting, regulatory compliance, prior authorization, education, and training, among other topics.
The closing keynote speaker, Andrew Toy, MS, Clover Health, spoke to GenAI’s ability to accelerate innovations by freeing up healthcare professionals from saying “I don’t have the time to do that,” as well as to glue new tools together in ways that “let us do the human part faster.” Efforts should focus on saving clinicians time and improving early detection of aging’s “greatest hits,” chronic conditions such as congestive heart failure, COPD, type 2 diabetes, and dementia.
By helping move medicine away from the traditional approach to the “diagnosis moment” and using longitudinal data to treat aging as a chronic condition, Toy argued that emergent AI-driven early detection may soon be functionally equivalent to a cure.
Enhancing understanding of aging and disease trajectories
As AI enables prediction and enhanced understanding of the progression of specific aging-related diagnoses, it is also opening a lens into how we think about the experience of disease from a human perspective. For researchers, clinicians, and caregivers alike, technology can offer a more granular understanding of how both aging and disease manifest over time, as well as new opportunities to re-examine fundamental questions about treatment and care across the aging community.
Rui Zhang, PhD, University of Minnesota, spoke of the opportunity to explore non-pharmacological interventions for Alzheimer's disease, such as sleep, diet, exercise, non-Western treatments, aromatherapy, light therapy, and other approaches. By leveraging AI and various real-world data sets from EHR and clinical trials, researchers can better understand which combinations of drugs and non-drug treatments produce the best outcomes in patients, and why.
Jason Karlawish, MD, and Emily Largent, PhD, JD, RN, University of Pennsylvania, led a panel discussion exploring an under-discussed aspect of dementia: temporary moments of lucidity that occur in neurodegenerative diseases. For example, as, or even before, a person starts experiencing symptoms of dementia, cognitive health data could add transformative caregiving and biological insights. Kishore Kuchibhotla, PhD, Johns Hopkins University, suggested that today’s standard measures of cognitive health “are not wrong; they’re noisy.” He called for mobile diagnostic tools, assessed via gamified tests, smartphones, tablets, devices.
Ipsit V. Vahia, MD, Harvard Medical School, echoed the role of wearable sensors to detect and differentiate between symptoms of disease and side effects of medications. This research into sensor-guided psychopharmacology has also raised challenging questions about clinical liability and accountability in responding to remotely captured data. Justin Clapp, PhD, University of Pennsylvania, underscored the need for greater partnership with caregivers to observe factors such as humor, situational awareness, eye contact, and other nonlinguistic attributes. All panelists agreed on both the need for better tools for caregivers to contribute to this research, the value of employing patient-caregiver dyad models in clinical studies, and the need for clinical research teams to think about the impact of their work beyond publishing papers.
Inclusive communication and care in aging and dementia
Over the course of the symposium, nuanced perspectives emerged on how to best engage older adults and caregivers in the development and implementation of technologies that can serve their needs. Speakers also gestured toward a central thematic question: Should healthcare leaders avoid medicalizing aging, or treat it as a chronic disease?
Amanda Lazar, PhD, University of Maryland, spoke about the explicit and unconscious impacts of ageism, “a process of systematic stereotyping and discrimination against people because they are old.” Stereotypes and pre-built assumptions about willingness to use technology hinder effective clinical research and product design, as well as undermine trust.
Reuben Ng, PhD, National University of Singapore, offered a global lens on communication about aging, underscoring the importance of avoiding reinforcement of negative stereotypes in public messaging and developing an emotional route to persuading the public to engage in their health.
"There are only four kinds of people in the world. Those who have been caregivers. Those who are currently caregivers. Those who will be caregivers, and those who will need a caregiver."
— Former First Lady Rosalynn Carter
In a panel on stakeholder engagement and lived experience, caregivers-turned-advocates Dee Fowlkes, Susan Jackewicz, Lily Liu, and Marie Maloney engaged in discussion about specific opportunities to improve care for individuals living with diseases of aging, including AD/ADRD. As panelists shared stories of caring for loved ones through, and losing them to, aging-related diseases, several themes emerged:
● Medication Side Effects: Drugs used to treat dementia can produce panic, sedateness, dyskinesia, suicidal ideation, and more. While well-documented clinically, these represent a chronic, pervasive, and under-addressed challenge for patients and caregivers alike.
● Inclusive Design: Technology design tends not to consider factors such as ethnicity, language, cultural norms, non-Western traditions, deafness, legal blindness, accessibility, and more. Caregivers have deep perspectives to offer developers of new technologies.
● Practical Solutions: From their expertise in “hacking” quick fixes for patients to tracking numerous data, caregivers represent a valuable and under-engaged perspective in today’s innovation ideation phase.
● Caring for Caregivers: Caregiving in dementia is typically an unpaid, do-it-yourself journey that can be made less arduous with special training, education, and support. AI can help caregivers with hands-on duties (e.g., data tracking, administrative compliance, filing taxes), as well as provide broader support (e.g., enabling bathroom breaks, emotional support, virtual entertainment or pre-recorded video tools to share with loved ones).
● Policy Reforms Needed: Health insurance designed for acute disease does not reflect the lived reality for families and communities facing dementia. Coverage skews toward narrow clinical outcomes, data access rules are counterproductive, and economic assumptions about new care models and technologies miss the real costs and benefits they can have in patients’ and families’ lives.
Forward with four words
In her opening remarks, Antonia M. Villarruel, PhD, RN, University of Pennsylvania, helped chart next steps by offering a rubric of four “H” words linked to key considerations to guide us forward through the coming era.
● Hype: While AI seems capable of answering every problem, how can we cut through hyperbole to understand both the limitations and unintended consequences of its use?
● Hope: Where does AI hold most potential for discovery, diagnosis, and supportive care?
● Harmony: How can emerging technology enhance, support, and amplify research and clinical care? How can it support people who need help right now?
● Humility: Setting aside our individual and institutional goals, how can we tap into our collective interests and make the best decisions for individuals and families?
These questions and the multifaceted challenges they represent were echoed by the other two opening speakers: George Demiris, PhD, University of Pennsylvania, who serves as PennAITech co-principal investigator, noted the breadth and depth of speakers’ expertise in areas ranging from industry to healthcare systems to family caregiving; Partha Bhattacharyya, PhD, NIA, who serves as an NIA AITC program official, underscored the importance of collaboration and its centrality to the AITC program’s mission. The four Hs served to permeate the symposium as an apt refrain embedded in perspectives shared by presenters, reinforcing the importance of fostering continued opportunities for varied stakeholders to engage and collaborate toward solutions for collective benefit.
For many pilot awardees, a next step toward delivering those solutions to real-world application in people’s lives will be to tap NIA’s Small Business Innovation Research and Small Business Technology Transfer programs, which make available approximately $150 million annually in nondilutive seed funding to help accelerate commercialization of aging-related technology to support older adults and caregivers. Todd Haim, PhD, NIA, shared in a panel on AgeTech funding and entrepreneurship that as members of the NIH innovator community, the a2 Collective’s funded pilot awardees can access benefits including expert consultations with industry veterans and are eligible to apply for NIH entrepreneurial bootcamps that can help position them well to secure new rounds of funding.
The a2 Collective is accepting applications for the fourth annual a2 Pilot Awards competition from March 1 to April 30, 2024 (5 p.m. ET). Learn more at a2PilotAwards.ai.
NIA is one of 27 Institutes and Centers of the National Institutes of Health at the U.S. Department of Health and Human Services. The a2 Collective is funded through NIA grants U24AG073094 (the a2 Collective Coordinating Center), P30AG073104 (JH AITC), P30AG073105 (PennAITech), and P30AG073107 (MassAITC).
The full 2024 a2 National Symposium agenda, along with upcoming events, additional resources, and other information, is available online. You can also follow the a2 Collective’s work on LinkedIn and X.
The a2 Collective thanks Naveen Rao, MHS, Patchwise Labs, for contributions to this piece.
References
1. World Social Report 2023: Leaving No One Behind in an Ageing World. United Nations. https://www.un.org/development/desa/dspd/wp-content/uploads/sites/22/2023/01/WSR_2023_Chapter_Key_Messages.pdf
2. 2024 Alzheimer’s Disease Facts and Figures. Alzheimer’s Association. https://www.alz.org/media/Documents/alzheimers-facts-and-figures-infographic.pdf
3. Alzheimer’s Disease and Related Dementias, CDC: https://www.cdc.gov/aging/aginginfo/alzheimers.htm
