Dartmouth Health physicians: AI can transform rural healthcare, but not without significant technology, access improvements

A newly published article by three Dartmouth Health physicians highlights the potential of artificial intelligence, coupled with improved broadband internet access, to improve healthcare access and outcomes for patients living in rural communities. “From Bandwidth to Bedside — Bringing AI-Enabled Care to Rural America,” co-authored by Nathan E. Goldstein, MD, chair of the department of medicine and professor of medicine at the Geisel School of Medicine at Dartmouth; Aretha D. Davis, MD, JD, strategic liaison for the department of medicine; and Angelo E. Volandes, MD, MPH, vice chair of research, professor of medicine and the Anna Gundlach Huber Professor in Medicine at Geisel; was published in the most recent issue of the New England Journal of Medicine.

In their article, Volandes, Goldstein and Davis describe a plausible scenario involving an elderly patient who lives alone, two hours’ drive from the nearest hospital. When the patient’s AI-enabled scale and smartwatch flagged weight gain and slowed gait, the alert reached her daughter by text; however, false alarms, unreliable broadband, and years of fragmented care had eroded the daughter’s trust in these technologies, leading her to wait to check on her mother until the next day, finding her in significant respiratory distress from pulmonary edema. The system had been right, technically, but the message lacked context."

While this situation is fictional, it represents similar real-life scenarios caused by inadequate technology for people in rural areas, like where many Dartmouth Health patients live. “Rural healthcare has always relied more on relationships than on devices—neighbors who stop by unasked, clinicians who make house calls after hours, grown children who drive in when they can. All are part of the extended social support network of rural America, an informal but resilient web of human connection that enables older adults to age in place,” Davis, Volandes and Goldstein write. “AI can bolster—but also fracture—the informal support network that is rural medicine’s lifeline. Whether it helps depends on design choices that respect limited bandwidth, honor local culture, and maintain fragile community trust.”

The authors also point to the success of an AI-powered triage system developed by Dartmouth Health that parses thousands of patient messages each week to help clinical teams identify and prioritize urgent concerns. Such predictive models can flag early signs of clinical deterioration hours before symptoms are obvious, but the factors that make rural populations vulnerable—uneven broadband coverage, fragmented care, overextended caregiver networks—make them significantly less effective for these patients.

“In our rural health system, we have seen technically sound models stumble once they collide with the lived realities of the communities they are meant to serve,” Goldstein, Volandes and Davis write. “…For AI to synergize with and not disrupt rural care delivery, it must be integrated into the social support network that sustains patients in their homes and communities. That means systems must offer actionable insights in plain language, accommodate local caregiving norms and infrastructure gaps, and support hybrid models that combine digital tools with human follow-up—whether in the form of a neighbor’s knock or a paramedic’s visit.”

Click here to read the full article.