Associations between body mass index and all-cause mortality among individuals with psoriasis: results from the NHANES database retrospective cohort study

Abstract

BackgroundPrevious findings imply a potential positive association between BMI and all-cause mortality in individuals with psoriasis, yet direct evidence remains absent. This study aimed to fill this gap.MethodsWe utilized data from the National Health and Nutrition Examination Survey (NHANES) for the periods 2003–2006 and 2009–2014. Participants’ BMI was categorized as lean (<25), overweight (25 ≤ BMI < 30), and obese (BMI ≥ 30). Psoriasis status was determined through self-reporting. The main outcome measured was all-cause mortality up to December 2019. We accounted for multiple covariates, such as sociodemographic factors and histories of smoking and alcohol consumption. Our statistical analyses mainly included Kaplan–Meier survival analysis, Restricted Cubic Spline (RCS) and Multivariate Cox Regression (MCR). We also applied propensity score matching (PSM) to verify the robustness of our findings.ResultsAmong 22,876 participants, 618 (2.70%) reported a history of psoriasis. An overall effect from the MCR analysis showed that, among individuals with psoriasis, a higher baseline BMI was independently associated with an increased risk of all-cause mortality, noting a 5.5% rise in mortality risk per BMI unit [hazard ratio (HR) = 1.055, 95% CI: 1.004–1.110, p = 0.035]. This significant relationship persisted after PSM. A statistically significant positive correlation was consistent among males, smokers, and individuals younger than 60. However, no such association was found in individuals without a history of psoriasis. Additionally, no significant difference in mortality risk was found between lean and overweight groups with psoriasis, according to the RCS regression and stratified analysis.ConclusionOur findings indicated a trend that, higher BMIs significantly correlated with increased risks of all-cause mortality in people with psoriasis, particularly among obese ones. However, the impact of being overweight on this relationship remains underexplored. Moreover, the necessity to employ alternative metrics beyond BMI for body fat assessment to further investigate these associations is critical.