Social Determinants of Health and Their Association With Parkinson's Disease Prevalence in US Adults: Insights From NHANES 2001-2020.

BACKGROUND

Social determinants of health (SDoH) have not been comprehensively studied in relation to Parkinson's disease (PD), although they are well known to be associated with other chronic diseases. Here we assessed the relationship of a multidimensional, comprehensive measure of SDoH with the prevalence of PD in US adults.

METHODS

Among 48,637 adults in the National Health and Nutrition Examination Survey (NHANES) between 2001 and 2020, we identified PD cases based on self-reported antiparkinsonian medication use. We formed a weighted multidomain SDoH score from eight indicators that assessed economic security, education, healthcare access, and social environment. Both multivariable logistic regression and restricted cubic spline (RCS) were used to estimate odds ratios and dose-response relationships and were adjusted for demographic and lifestyle factors.

RESULTS

More intensive adverse SDoH burden was significantly associated with higher PD prevalence (fully adjusted OR 1.31, 95% CI 1.16-1.47 per unit increase), and RCS analysis showed no evidence of nonlinearity (p for nonlinearity = 0.370). Unemployment (OR 4.15, 95% CI 2.70-6.40) and food insecurity (OR 1.96, 95% CI 1.39-2.76) were strongly associated with higher PD prevalence, whereas reported lack of health care access was associated with lower identified PD (OR 0.27, 95% CI 0.15-0. 49), pointing to likely ascertainment bias. When stratifying, we observed meaningful effect modification by age (p < 0.001), with stronger effects in middle-aged individuals (≤ 60 years: OR 1.35, 95% CI 1.13-1.61), compared with an attenuation in older individuals.

CONCLUSION

Poor SDoH were independently associated with higher PD prevalence, particularly among adults aged 60 years and younger. Unemployment and food insecurity showed strong positive associations, whereas the inverse association with healthcare access likely reflects ascertainment bias. These findings are hypothesis-generating and suggest that SDoH may be relevant to PD screening and epidemiologic research. Longitudinal studies are needed to clarify temporality and causality.