The aim of this study was to evaluate the relative validity of indicators of food and beverage intake obtained from the telephone interview surveillance system (VIGITEL). A random sample (n = 100) was evaluated from the total sample of approximately two thousand adults studied by the system in 2009 in Belém, Pará State, Brazil. The indicators were protective factors (adequate consumption of fruit, vegetables, and leafy vegetables) and risk factors (consumption of saturated fat, soft drinks, and alcoholic beverages) for chronic non-communicable diseases. The telephone interview results were compared with those of three 24-hour recalls (reference standard). The reference standard showed underestimation in the indicators' frequency, except for soft drinks and alcoholic beverages. The mean consumption frequencies were generally higher in the exposed group interviewed by the VIGITEL system. We cannot conclude that the VIGITEL system is not a good indicator of consumption, since the reference standard also shows limitations. Nevertheless, its use as a surveillance tool in Brazil is justifiable.