目的: 探讨体重校正腰围指数(weight-adjusted waist index，WWI)与美国成人急性、亚急性、慢性疼痛之间的相关性。方法: 采用横断面研究，提取1999—2004年美国国家卫生和营养检查调查(National Health and Nutrition Examination Survey，NHANES)数据库中关于成人腰围、体质量和疼痛等变量，及性别、年龄、种族、婚姻状况、教育水平、家庭收入、体力活动情况、饮酒、吸烟和糖尿病患病状况等协变量数据。采用多分类Logistic回归分析构建3种模型，评估WWI与急性、亚急性和慢性疼痛之间的相关性。模型1未对协变量进行校正，模型2对年龄、性别、种族、婚姻状况、教育水平和家庭收入情况进行校正，模型3进一步校正了体力活动、饮酒、吸烟和糖尿病患病情况等所有协变量。结果: 共纳入12 694例参与者，平均年龄为(50.8±18.7)岁, 其中9 614例(75.74%)未出现超过24 h的疼痛，870例(6.85%)出现急性疼痛，354例(2.79%)出现亚急性疼痛，1 856例(14.62%)出现慢性疼痛。所有参与者的WWI为(10.95±0.85) cm/$\sqrt{\mathrm{kg}}$，根据WWI的四分位数分为4组：Q1组为(7.90~10.36) cm/$\sqrt{\mathrm{kg}}$，Q2组为(10.37~10.94) cm/$\sqrt{\mathrm{kg}}$，Q3组为(10.95~11.53) cm/$\sqrt{\mathrm{kg}}$，Q4组为(11.54~15.20) cm/$\sqrt{\mathrm{kg}}$。随着WWI的增加，参与者的急性、慢性疼痛状态的差异有统计学意义(P < 0.001)。模型1中，与Q1组相比，Q2组和Q4组的急性疼痛风险降低(Q2组：OR=0.765，95%CI：0.615~0.953，P=0.017；Q4组：OR=0.648，95%CI：0.503~0.835，P < 0.001)；与Q1组相比，Q2组、Q3组和Q4组的慢性疼痛风险均增加(Q2组：OR=1.365，95%CI：1.149~1.622，P < 0.001；Q3组：OR=1.291，95%CI：1.082~1.541，P=0.005；Q4组：OR=1.874，95%CI：1.579~ 2.224，P < 0.001)。模型2中，与Q1组相比，其他3组慢性疼痛风险增加(Q2组：OR=1.359，95%CI：1.137~1.624，P=0.001；Q3组：OR=1.260，95%CI：1.039~1.528，P=0.019；Q4组：OR=1.735，95%CI：1.413~2.132，P < 0.001)。模型3中与Q1组相比，Q4组的慢性疼痛风险增加49.2%(OR=1.492，95%CI：1.208~1.842，P < 0.001)。在模型2和模型3中，急性疼痛与WWI未见相关性(均P>0.05)；3个模型均未发现亚急性疼痛与WWI存在相关性(均P>0.05)。结论: WWI与美国成人急性疼痛、亚急性疼痛之间未见明显相关性，但随着WWI的增加，慢性疼痛风险增加, 所以有必要通过大规模前瞻性研究进一步验证这一结论。

Objective: To investigate the relationship between the weight-adjusted waist index (WWI) and acute, subacute pain or chronic pain among American adults. Methods: There was a cross-sectional study. Data from the 1999-2004 National Health and Nutrition Examination Survey (NHANES) concerning waist circumference, weight, pain status and covariates (age, gender, race, marital status, education level and income, physical activity, alcohol consumption, smoking status, and diabetes) were extracted for analysis. Multinomial Logistic regression was conducted across the three models to investigate the associations between WWI and acute, subacute and chronic pain. Model 1 did not involve any adjustments. Model 2 involved adjustments for age, gender, race, marital status, education level, and income. Model 3 was further adjusted for physical activity, alcohol consumption, smoking, and diabetes status. Results: This study involved 12 694 participants with an average age of (50.6±18.7) years. Among all the participants, 9 614 people (75.74%) had no pain, 870 people (6.85%) experienced acute pain, 354 people (2.79%) suffered from subacute pain, and 1 856 people (14.62%) experienced chronic pain. The WWI of all the participants was (10.95±0.85) cm/$\sqrt{\mathrm{kg}}$, divided into four groups based on quartiles: Group Q1 (7.90-10.36) cm/$\sqrt{\mathrm{kg}}$, group Q2 (10.37-10.94) cm/$\sqrt{\mathrm{kg}}$, group Q3 (10.95-11.53) cm/$\sqrt{\mathrm{kg}}$ and group Q4 (11.54-15.20) cm/$\sqrt{\mathrm{kg}}$. With the increase of WWI, the analysis revealed a significant statistical difference in the participants' acute and chronic pain status (all P < 0.001). In Model 1, the prevalence of acute pain was lower in group Q2 and group Q4 compared with group Q1 (group Q2: OR=0.765, 95%CI: 0.615-0.953, P=0.017; group Q4: OR= 0.648, 95%CI: 0.503-0.835, P < 0.001). Compared with group Q1, the prevalence of chronic pain increased in group Q2, group Q3, and group Q4 (group Q2: OR =1.365, 95%CI: 1.149-1.622, P < 0.001; group Q3: OR=1.291, 95%CI: 1.082-1.541, P=0.005; group Q4: OR=1.874, 95%CI: 1.579-2.224, P < 0.001). In Model 2, compared with group Q1, an increase in chronic pain prevalence was still associated with an increase in WWI in other three groups (group Q2: OR=1.359, 95%CI: 1.137-1.624, P=0.001; group Q3: OR=1.260, 95%CI: 1.039-1.528, P=0.019; group Q4: OR=1.735, 95%CI: 1.413-2.132, P < 0.001). In Model 3, group Q4 had a 49.2% increased prevalence of chronic pain compared to group Q1 (OR = 1.492, 95%CI: 1.208-1.842, P < 0.001). However, in Models 2 and 3, no significant relationship was observed between acute pain and WWI (all P>0.05). And none of the three models identified a significant association between subacute pain and WWI (all P>0.05). Conclusion: For American adults, there was no significant correlation between WWI and acute pain or subacute pain. However, as WWI increases, so does the prevalence of chronic pain. Further validation of this conclusion through large-scale prospective studies is warranted.