Food consumption patterns in Italy: the INN-CA Study 1994–1996

European Journal of Clinical Nutrition (2001) 55, 571±588 ß 2001 Nature Publishing Group All rights reserved 0954±3007/01 $15.00 www.nature.com/ejcn Original Communication Food consumption patterns in Italy: the INN-CA Study 1994 ± 1996 A Turrini1*, A Saba1, D Perrone2, E Cialfa1 and A D'Amicis1 1 Istituto Nazionale di Ricerca per gli Alimenti e la Nutrizione (INRAN), Rome, Italy; and 2Istituto Nazionale di Statistica (ISTAT), Rome, Italy Objectives: The present study was aimed at surveying the Italian food consumption patterns in the 90s. It represented the second nationwide food intake survey that was carried out by the Istituto Nazionale di Ricerca per gli Alimenti e la Nutrizione (INRAN). Design: Cross-sectional study. Setting: Free-living households. Subjects: Sampled subjects: 1147 households randomly selected to be representative of the four main geographical areas (North-West, North-East, Centre, South). Analysed subjects: 1978 individuals out of 2734 initially collected in 15 Collaborative Centres strategically scattered through the national territory. Methods: A mixed 7-day based survey technique was applied in order to survey both individual and household consumption. At individual level, food intake was recorded by a self-compiled diary. At household level food data collection was by compilation of a food inventory by the dietician, a purchased=wasted foods diary and a recipes form both compiled by the person responsible for food related activities (decision of purchase, purchase, preparation of meals) in the household. All recorded data were monitored by the dieticians who visited the households participating in the study, at least three times. Household members were interviewed in order to collect socio-demographic, lifestyle and motivational information. Furthermore, ®eld-workers performed the food coding and the input of data by an ad hoc developed software. The survey design allowed an internal quantitative check of food data. Several check steps were centrally performed Results: The complex methodology caused the 46.8% response rate and afterwards a reduction of the analysed units (72% of the surveyed individuals). However, the analysed sample provided suf®ciently reliable data for outlining the most relevant aspects of dietary patterns in Italy. In order to better interpret the results, controversial aspects are also illustrated and discussed in the text. The disparity analysis found that traditional diversities among Italian regions are still alive. In general, males eat more than females. Analysis according to the age class (children: 1 ± 9 years old; adolescents: 10 ± 17; adults: 18 ± 64; elderly: > 64) showed differences especially between the group of children vs all the others, but also in the two groups of young individuals vs adults and the elderly, that could represent clues of an incorrect way of eating. Conclusions: Taking into account the different methodology, the comparison with previous results showed changes in the Italian average diet complying with an increasing attention to healthy aspects by large sectors of the population, but also opposite tendencies that should be carefully monitored. Sponsorship: Minister delle Politiche Agricole e Forestali (Mi.P.A.F.). Descriptors: Italian diet; food consumption; data collection; nationwide survey; individual survey; 7 day food record European Journal of Clinical Nutrition (2001) 55, 571±588 *Correspondence: A Turrini, UnitaÁ di Statistica ed Economia Alimentare, Istituto Nazionale di Ricerca per gli Alimenti e la Nutrizione, Via Ardeatina 546, I-00178 Rome, Italy. E-mail: Guarantor: A Turrini. Contributors: AT designed the study in collaboration with the other researchers, projected and supervised data management at the co-ordinating centre, did the analyses and drafted the paper. AS participated in the study design, supervision of data collection and interpretation of the results. DP developed the data-entry and data-check software, in addition to participating in the database design. EC promoted the initiative of the project in addition to participating in the study design, recruitment of the collaborative group and interpretation of the results with very appreciated contributes. AD'A recruited and co-ordinated the collaborative group, in addition to participating in the study design, interpretation of the results, and all versions of the paper. Received 14 August 2000; revised; 2 January 2001; accepted 10 January 2001 Food consumption in Italy A Turrini et al 572 Introduction Outlining food consumption patterns is key in assessing dietary adequacy and safety, representing important aspects in evaluating the relationship between diet and health, including the role of consumers' attitudes, preferences and lifestyle related to socio-economic situation and cultural models. In Italy, different sources provide data on food supply and consumption. In particular, in the public sector, the National Institute of Statistics (ISTAT) delivers Food Balance Sheets (ISTAT, 1981 ± 1990) yearly, also published by the Food and Agriculture Organisation (FAO) (FAO, 1991 ± 1994, 1994 ± 1997), Household Budget Survey results (ISTAT, 1978 ± 1995) and more recently information on Italian eating habits (ISTAT, 1998). Other consumption data are supplied by private marketing companies. The types of data provided (food supply, consumption and frequencies) are not immediately suitable for the evaluation of the nutritional aspects of dietary pro®les, because the data structure is meant for economic analysis. However, these data are often used for international comparisons or trend analysis, mainly because they are carried out at regular time intervals by adopting comparable methodologies. On the other hand, dietary information for speci®c purposes (single food components and=or population groups) are copiously provided by several research teams within diverse contexts, such as for example epidemiological studies, clinical trials, etc (Turrini, 1993). However, the informative power of this large amount of data is reduced by the extreme fragmentation. In synthesis, either currently collected data provide a comprehensive picture but they are not designed for nutritional analyses, or ad hoc surveyed data are much too speci®c for comprehensive studies on food behaviour with nutritional purposes. For this reason, the Istituto Nazionale di Ricerca per gli Alimenti e la Nutrizione (INRAN) carried out the ®rst nationwide food intake study in the 1980s (Saba et al, 1990). In the ®rst study, data were collected at household level only and per-capita consumption was assessed by utilising ancillary information, such as the number of Table 1 family members, the number and type of meals consumed at home and the importance of the latter for each person (Saba et al, 1990; Cialfa et al, 1991). The survey comprised approximately 12 000 households ( 36 000 individuals) scattered throughout the national territory and represe (...truncated)