Demographic and Socioeconomic Factors Associated with Fitbit Ownership in the NIH All of Us Cohort

Wearable fitness trackers are increasingly popular for monitoring health-related metrics, yet their ownership patterns across socioeconomic, demographic, and gender-diverse populations remain underexplored at a population level. This study utilized data from the NIH All of Us Research Program to investigate how area-level socioeconomic status, race, and gender identity influence wearable device ownership. Methods. Data were analyzed from 633,547 participants from the All of Us Dataset. Fitbit ownership was modeled with four binary logistic regression models: a demographics-only model, a ZIP3-level socioeconomic indicators model, and a combined model incorporating four demographic × median household income interactions (race, gender, age, and Hispanic/Latino ethnicity), and an intersectional model adding a race x gender interaction. Continuous socioeconomic predictors were rescaled for interpretability (median income per USD 10,000; area-level fractions per 10 percentage points). Socioeconomic-adjusted models were restricted to 606,414 participants with available ZIP3-linked data. Fitbit ownership was defined as having a Fitbit record in the database. Results. Fitbit ownership was observed in 8.34% of the study population. Logistic regression analyses revealed significant demographic disparities: female participants and gender-diverse identities had significantly higher odds of ownership than males (OR = 1.25–2.2). Black or African American (OR = 0.38) and NHPI/MENA (OR = 0.82) participants had lower odds compared to White participants, while Asian (OR = 1.13), more than one race (OR = 1.25), and Hispanic or Latino (OR = 1.25) participants had higher odds. Each USD 10,000 increase in ZIP3 median household income was associated with 12.5% lower odds of ownership overall (OR = 0.875), but this gradient varied significantly by race. For Black or African American participants, the relationship reversed direction (OR = 1.08 per $10,000). A race x gender interaction further showed that female ownership was not uniform across race, being the largest among Black or African American participants (OR = 2.27) and reversed among Asian participants (OR = 0.87). ZIP3 socioeconomic data were structurally unavailable for all American Indian or Alaska Native participants due to the All of Us program’s small-population ZIP3 aggregation policy, precluding their inclusion in socioeconomic-adjusted models. Conclusions. This analysis demonstrates significant gender, racial, and socioeconomic disparities in wearable fitness tracker ownership, showing significantly higher device usage among females and gender-diverse individuals, but lower usage among certain racial groups and a seemingly contradictory negative ownership rates among higher socioeconomic levels. Ownership patterns nonetheless appear more equitable than in consumer cohorts, likely reflecting the device-provision programs undertaken by the NIH.