Non-linear association between weight-adjusted-waist index and obstructive sleep apnea: a cross-sectional study from the NHANES (2005–2008 to 2015–2020)

Background The principal objective of the present investigation is to undertake an in-depth exploration of the relationship that exists between the newly introduced weight-adjusted waist index (WWI), employed as a surrogate way for corpulence, and obstructive sleep apnea (OSA). Methods Analysis using cross-sectional data from 11,545 NHANES participants across 2005–2008 and 2015–2020. Obesity via WWI (waist circumference over sqrt of body weight). OSA via 3 NHANES QnA items: monthly excessive sleepiness, weekly wheezing/snoring/breathing stoppage, weekly snoring. Relationships between WWI and OSA probed with weighted multivariate logistic regression and smoothed curve fitting. Also did subgroup, interaction tests and threshold effect analysis. Excluded those with incomplete WWI, OSA or hypertension data as they might have different health profiles. We excluded participants with incomplete data on WWI, OSA, or hypertation-related items, as those with missing data might have different health profiles. Results The study, encompassing a cohort of 11,545 participants, revealed that 5,727 individuals were diagnosed with OSA. Upon conducting fully adjusted models, A positive relevance between WWI and OSA was established, with an odds ratio of 1.57 (95% CI : 1.44, 1.71), indicating a significant relationship. Notably, participants falling within the highest quartile of WWI exhibited a markedly heightened propensity for OSA, being 2.58 times more likely to suffer from it than those in the bottom quartile [ OR : 2.58 (95% CI : 2.10, 3.17)]. Rigorous subgroup analyses and interaction tests further confirmed the robustness of this positive association across various subgroups, thereby affirming the consistency of the observed relationship. Additionally, a noteworthy non-linear association and saturation phenomenon were discerned between the WWI and OSA, demarcated by an inflection point at 11.70 cm/√kg. Conclusion Our research has clearly shown a significant positive correlation, along with a saturation effect, between WWI and OSA in the American population. However, the cross-sectional design limits causal inference, and the exclusion of certain participants may affect the generalizability of the findings. Future longitudinal studies are needed to explore causality and address potential biases associated with participant exclusion, ultimately improving the broader applicability of the results.