Socioeconomic determinants of health (SDH) — such as gender, race, and household income — have an impact on who gets infected by COVID-19, who becomes very sick from the disease, and who requires the highest level of life-saving interventions. Predicting risk levels based on SDH allows providers to get ahead of the game in their prevention and treatment strategies. A new study suggests that AI and machine learning can accurately predict the need for inpatient or emergency medical services based on publicly available SDH data.
The study results were published in January in the American Journal of Managed Care; they could have a broad influence in providing targeted treatment and potentially reduce health care costs. Without interacting with patients and using only the patient’s age, gender, race, and address, a provider can utilize these AI algorithms to predict the way COVID-19 will affect a specific patient or even a neighborhood. The algorithms also help providers assess the health risks for patients who miss multiple routine care visits due to stay-at-home orders or concerns about contracting and spreading the disease.
To help people avoid hospitalization and other costly services, providers, and caregivers could use the data to drive outreach programs in at-risk communities, encouraging individuals and families to participate in preventive care practices. The information could also help providers effectively treat a particular patient in the early stages of a disease. And medical administrators might even rely on the data to identify uninsured patients so they can facilitate Medicare or Medicaid applications.
Numerous studies conducted over several decades have demonstrated that poverty, race, ethnicity, gender, and other socioeconomic factors increase the risk of chronic disease and mental health conditions, lower life expectancy, and contribute to additional medical concerns. Ultimately, access to quality predictive SDH information has the potential to offset the burden born by people of color, low-income families, and other marginalized groups. That potential could make a big difference–not only in the wake of the COVID-19 crisis but also across the spectrum of debilitating illnesses that take an immense financial and personal toll on many Americans.