Dose-Response Association of Waist-to-Height Ratio Plus BMI and Risk of Depression: Evidence from the NHANES 05–16

BACKGROUND: Obesity or underweight has been found to be associated with depression, but the relationship remains to be determined so that more precise prevention strategies can be implemented. Body mass index (BMI) and waist–height ratio (WHR) were used as indicators to study the dose–response relationship between depression and obesity or underweight. METHODS: We obtained basic information and disease-related data for 13,975 adults from the National Health and Nutrition Examination Survey (NHANES) 2005–2016 dataset. The depressive status was determined based on the PHQ-9 scale (>4). Logistic regression was used to analyze the association and risk of BMI, WHTR and depressive status. Based on the results of logistic regression, the dose–response relationship between BMI, WHTR and depressive state was analyzed using restricted cubic splines (RCS). RESULTS: The adjusted model showed that compared with the fourth quartile (Q4) of BMI, the odds ratios (ORs) of depression for Q1, Q2 and Q3 were 0.63 (0.56–0.71), 0.61 (0.54–0.68) and 0.74 (0.66–0.82), and compared with the fourth quartile (Q4) of WHtR, the odds ratios (ORs) of depression for Q1, Q2 and Q3 were 0.55 (0.49–0.62), 0.57 (0.51–0.64) and 0.64 (0.57–0.71), respectively. The restricted cubic spline regression depicted a U-shaped dose–response relationship between continuous changes of obesity indicators and the risk of depression (P1, P2 < 0.001). When the participants’ BMI reached approximately 25kg/m(2) with the reference value of BMI was 18.5kg/m(2), the risk of depression was minimized (OR=0.68, 95% Cl=0.56–0.83). When the WHtR reached approximately 0.52 with the reference value of WHtR was 0.40, the risk of depression was minimized (OR=0.69, 95% Cl=0.54–0.88). CONCLUSION: We found a significant U-shape correlation between BMI, WHtR and depression. People with slight overweight have the lowest risk of depression. However, according to the International Obesity standards, the population at these levels of weight may have an obesity-chronic disease risk, and this is not recommended.