Department of Epidemiology, University of North Carolina-Chapel Hill Gillings School of Global Public Health, Chapel Hill, USA.
Department of Epidemiology, University of California-Berkeley, School of Public Health, Berkeley, USA.
Cancer Causes Control. 2020 Feb;31(2):193-202. doi: 10.1007/s10552-019-01261-4. Epub 2020 Jan 1.
The association between diet quality and head and neck cancer (HNC) was explored using a population-based case-control study of 1170 HNC cases and 1303 age-, race-, and sex-matched controls from the United States. Diet quality was assessed with three diet quality scores (DQS): (a) Healthy Eating Index 2005 (HEI-2005), (b) Mediterranean Diet Score (MDS), and (c) HNC-specific Mediterranean Diet Score (MDS-HNC), a modified MDS that we developed to be more applicable to HNC. Logistic regression models estimated adjusted odds ratios (ORs) and 95% confidence intervals (CIs) representing diet quality-incident HNC associations. We examined effect measure modification (EMM) by body mass index (BMI), race, cigarette smoking, and alcohol consumption and associational heterogeneity by HPV-positivity and tumor site. A one standard deviation summary DQS decrement suggested a consistent inverse association (ORs (CIs)) for the HEI-2005, MDS, and MDS-HNC: 1.35 (1.21, 1.50), 1.13 (1.02, 1.25), and 1.17 (1.06, 1.31), respectively. This association did not vary by tumor site or tumor HPV status, though additive EMM by alcohol use and by BMI was observed. Our findings suggest the Mediterranean diet can be used to study HNC in American populations, and that poor diet quality elevates HNC incidence, particularly among alcohol users.
采用美国 1170 例头颈部癌症（HNC）病例和 1303 例年龄、种族和性别匹配对照的基于人群的病例对照研究，探讨了饮食质量与 HNC 之间的关联。使用三种饮食质量评分（DQS）评估饮食质量：（a）2005 年健康饮食指数（HEI-2005），（b）地中海饮食评分（MDS），和（c）我们开发的更适用于 HNC 的改良 MDS，即 HNC 特定的地中海饮食评分（MDS-HNC）。逻辑回归模型估计了代表饮食质量与 HNC 关联的调整后比值比（OR）和 95％置信区间（CI）。我们通过体重指数（BMI）、种族、吸烟和饮酒检查了效应量修饰（EMM），并通过 HPV 阳性和肿瘤部位检查了关联异质性。DQS 降低一个标准差的综合指标提示，HEI-2005、MDS 和 MDS-HNC 与 HNC 呈一致的负相关（OR（CI））：1.35（1.21，1.50）、1.13（1.02，1.25）和 1.17（1.06，1.31）。尽管观察到酒精使用和 BMI 的附加 EMM，但这种关联不受肿瘤部位或肿瘤 HPV 状态的影响。我们的研究结果表明，地中海饮食可用于研究美国人群中的 HNC，而不良的饮食质量会增加 HNC 的发病率，尤其是在饮酒者中。
