Multidimensional poverty and risk of atherosclerotic cardiovascular disease: A U.S. National Study

Butt, Sara Ayaz; Tano Retamales, Mauricio; Javed, Zulqarnain; Hagan, Kobina; Arshad, Hassaan Bin; Khan, Safi; Acquah, Isaac; Nikoloski, Zlatko; Mossialos, Elias; Cainzos-Achirica, Miguel; Nasir, Khurram

doi:http://dx.doi.org/10.1016/j.jacadv.2024.100928

Multidimensional poverty and risk of atherosclerotic cardiovascular disease: A U.S. National Study

License

Citation

Butt SA, Retamales MT, Javed Z, Hagan K, Arshad HB, Khan S, et al. Multidimensional poverty and risk of atherosclerotic cardiovascular disease: A U.S. National Study. JACC Adv. 2024 May 15;3(7):100928. DOI: 10.1016/j.jacadv.2024.100928

Abstract

Background: Poverty is associated with atherosclerotic cardiovascular disease (ASCVD). While poverty can be evaluated using income, a unidimensional poverty metric inadequately captures socioeconomic adversity. Objectives: The aim of the study was to examine the association between a multidimensional poverty measure and ASCVD. Methods: Survey data from the National Health Interview Survey was analyzed. Four poverty dimensions were used: income, education, self-reported health, and health insurance status. A weighted deprivation score (c i ) was calculated for each person. The multidimensional poverty index was computed for various cutoffs, k, for total population, and by ASCVD status. The association between multidimensional poverty and ASCVD was examined using Poisson regression. Area under receiver operator characteristics curve analysis was performed to compare the multidimensional poverty measure with the income poverty measure as a classification tool for ASCVD. Results: Among the 328,164 participants, 55.0% were females, the mean age was 46.3 years, 63.1% were non-Hispanic Whites, and 14.1% were non-Hispanic Blacks. Participants with ASCVD (7.95%) experienced greater deprivation at each multidimensional poverty cutoff, k, compared to those without ASCVD. In adjusted models, higher burden of multidimensional poverty was associated with up to 2.4-fold increased prevalence of ASCVD (c i = 0.25, adjusted prevalence ratio [aPR] = 1.66, P < 0.001; c i = 0.50, aPR = 1.99; c i = 0.75, aPR = 2.29; P < 0.001; c i = 1.00, aPR = 2.38, P < 0.001). Multidimensional poverty exhibited modestly higher discriminant validity, compared to income poverty (area under receiver operator characteristics = 0.62 vs 0.58). Conclusions: There is an association between the multidimensional poverty and ASCVD. Multidimensional poverty index demonstrates slightly better discriminatory power than income alone. Future validation studies are warranted to redefine poverty's role in health outcomes.