AI4HealthyCities: a protocol for a mixed-method ethnographic study on the social determinants of cardiovascular health in New York City.

Abstract

Introduction: Cardiovascular disease is a leading cause of morbidity and mortality globally and in New York City. Significant disparities in prevalence and risk factors persist across city neighbourhoods and among populations of varying socio-economic status, racial and ethnic backgrounds. These disparities are shaped and sustained by the complex interplay of social determinants of health, including housing, employment, access to healthcare and structural inequities. This study builds on prior quantitative research conducted under the AI4HealthyCities initiative, which applied machine learning to identify spatial clusters of cardiovascular vulnerability and social disadvantage. This research addresses key gaps in that work by generating disaggregated, qualitative data on underrepresented populations. The study also aims to explore the mechanisms through which specific social determinants may contribute to cardiovascular risk, including the role of behavioural and demographic mediators. By combining lived experiences and system-level perspectives, the research will provide contextualised insights to support local stakeholders in designing more effective, equity-oriented interventions.

Methods and analysis: This mixed-method ethnographic study will collect data in three phases: expert interviews, community roundtables and vulnerability assessments across three New York City boroughs (Brooklyn, the Bronx and Queens). Qualitative data will be analysed using a combination of deductive and inductive approaches. Thematic synthesis will be used to identify patterns across and within boroughs. Study design and interim findings will be reviewed in collaboration with community stakeholders, a research steering group and the AI4HealthyCities Global Expert Council.

Ethics and dissemination: This study has received ethical approval from the Ethics Committee of Weill Cornell Medicine (code number: 23-04025988). The findings of the project will be disseminated via conferences, speaking engagements and peer-reviewed publications.