Dietary patterns and oral and pharyngeal cancer using latent class analysis.

ABSTRACT

The methods traditionally used to identify a posteriori dietary patterns are principal components, factor and cluster analysis. The aim of our study is to assess the relationship between dietary patterns derived with latent class analysis (LCA) and oral/pharyngeal cancer risk (OPC), highlighting the strengths of this method compared to traditional ones. We analyzed data from an Italian multicentric case-control study on OPC including 946 cases and 2,492 hospital controls. Dietary patterns were derived using LCA on 25 food groups. A multiple logistic regression model was used to derive odds ratios (ORs) and corresponding 95% confidence intervals (CIs) for OPC according to the dietary patterns identified. We identified four dietary patterns. The first one was characterized by a high intake of leafy and fruiting vegetable and fruits (Prudent pattern), the second one showed a high intake of red meat and low intake of selected fruits and vegetables (Western pattern). The last two patterns showed a combination-type of diet. We labeled "Lower consumers-combination pattern" the cluster that showed a low intake of the majority of foods, and "Higher consumers-combination pattern" the one characterized by a high intake of various foods. Compared to the "Prudent pattern", the "Western" and the "Lower consumers-combination" ones were positively related to the risk of OPC (OR = 2.56, 95% CI 1.90-3.45 and OR = 2.23, 95% CI 1.64-3.02). No difference in risk emerged for the "Higher consumers-combination pattern" (OR = 1.28, 95% CI 0.92-1.77).