Diseño de cuestionarios de frecuencia de consumo para estudiar la relación dieta-cáncer en Colombia

Abstract

In Colombia, cancer ranks third as a cause of death in the adult population. Thirty-five percent of cancer deaths are attributable to dietary factors and that 90% of colorectal cancers can be prevented by appropriate diets. A dietary questionnaire was developed to study the relationship between diet and cancer for five cities in Colombia (Barranquilla, Cartagena, Santa Marta, Bogotá and Bucaramanga and its metropolitan area), based on 24-hour dietary survey data accumulated by the National Survey of Nutritional Status-2005. Each questionnaire had 3 sections and involved six stages in its development. A checklist representing thirteen nutrients was formulated by means of the method of the variance maximization (Max_r); it was supplemented with nine categories of consumption frequency. The two complementary sections inquired about the consumption of food, number of meals per day, and methods of preparation, related with cancer development. Between 22 and 32 foods were discriminated for each questionnaire. Only nine foods were common in the checklists of each of the five cities. Nine of the thirteen nutrients are represented with correlations higher than 0.80, with a minimum 0.47 and a maximum of 1.0. For the first time in Colombia specific questionnaires were developed to study the diet-cancer relationship. The checklist was comprehensive and it incorporated between-person variability, as well as the habits and food culture of each city. We propose nine analysis alternatives for the questionnaire. The questionnaires will prove useful in epidemiological research, although they require calibration with studies of reproducibility and validity.