Department of Orthopedics, The Second Hospital of Shanxi Medical University, Taiyuan, 030001, Shanxi, P.R. China.
Academy of Medical Sciences, Shanxi Medical University, Taiyuan, P.R. China.
Lipids Health Dis. 2024 Oct 14;23(1):334. doi: 10.1186/s12944-024-02324-5.
The link between body roundness index (BRI) and osteoarthritis (OA) has yet to be validated. Our aim was to explore this connection between BRI and OA risk.
This cross-sectional study utilized the 1999-2018 National Health and Nutrition Examination Survey retrieved data. To assess the association between BRI and OA risk, we performed weighted multivariable regression analysis (MVRA), with smooth curve fitting for potential nonlinear association and subgroup analysis and interaction tests for relationships in specific subgroups. A 7:3 ratio was adopted for the random division of the acquired data into training and validation sets. Subsequently, least absolute shrinkage and selection operator regression, along with MVRA, were conducted for the training set to isolate variables for a prediction model. This model was visualized using the nomogram and was followed by evaluation. Finally, the validation set was utilized to validate the model.
This study enrolled 12,946 individuals. Following the adjustment for all covariables, OA risk increased by 18% with every unit rise in BRI (odd ratio [OR] = 1.18; 95% confidence interval [CI]: 1.13-1.23; P < 0.0001). Upon regarding BRI as a categorical variable, it was divided into quartiles for subsequent analysis. In comparison to quartile 1, the risk of OA was increased in quartile 2 (OR = 1.58; 95% CI: 1.22-2.03; P = 0.0006), quartile 3 (OR = 1.83; 95% CI: 1.40-2.40; P < 0.0001) and quartile 4 (OR = 2.70; 95% CI: 1.99-3.66; P < 0.0001). Smooth curve fitting revealed no non-linear relationships. None of the subgroups showed a statistically significant interaction (all P > 0.05). After selecting the variables, a prediction model was developed. The prediction model exhibited favorable discriminatory power, high accuracy, and potential clinical benefits in training and validation sets.
The BRI was positively associated with OA risk. Our predictive model demonstrated that combining BRI with other easily accessible factors was helpful in assessing and managing high-risk OA groups.
体圆指数（BRI）与骨关节炎（OA）之间的联系尚未得到证实。我们的目的是探讨 BRI 与 OA 风险之间的这种联系。
本横断面研究利用了 1999 年至 2018 年全国健康和营养检查调查的数据。为了评估 BRI 与 OA 风险之间的关联，我们进行了加权多变量回归分析（MVRA），并对潜在的非线性关联进行了平滑曲线拟合，以及在特定亚组中进行了亚组分析和交互测试。所获得的数据采用 7:3 的比例随机分为训练集和验证集。随后，对训练集进行最小绝对收缩和选择算子回归以及 MVRA，以分离预测模型的变量。使用列线图可视化该模型，并对其进行评估。最后，使用验证集验证模型。
这项研究共纳入了 12946 人。在调整所有协变量后，BRI 每升高一个单位，OA 风险增加 18%（优势比[OR] = 1.18；95%置信区间[CI]：1.13-1.23；P < 0.0001）。当将 BRI 视为分类变量时，将其分为四分位数进行后续分析。与四分位 1 相比，四分位 2（OR = 1.58；95% CI：1.22-2.03；P = 0.0006）、四分位 3（OR = 1.83；95% CI：1.40-2.40；P < 0.0001）和四分位 4（OR = 2.70；95% CI：1.99-3.66；P < 0.0001）的 OA 风险增加。平滑曲线拟合未显示出非线性关系。所有亚组的交互作用均无统计学意义（均 P > 0.05）。在选择变量后，建立了预测模型。该预测模型在训练集和验证集均表现出良好的区分能力、高准确性和潜在的临床获益。
BRI 与 OA 风险呈正相关。我们的预测模型表明，将 BRI 与其他易于获得的因素相结合有助于评估和管理高危 OA 人群。
