Objective: To describe the diversity in dietary patterns existing across centres/regions participating in the European Prospective Investigation into Cancer and Nutrition (EPIC). Design and setting: Single 24-hour dietary recall measurements were obtained by means of standardised face-to-face interviews using the EPIC-SOFT software. These have been used to present a graphic multi-dimensional comparison of the adjusted mean consumption of 22 food groups. Subjects: In total, 35 955 men and women, aged 35-74 years, participating in the EPIC nested calibration study. Results: Although wide differences were observed across centres, the countries participating in EPIC are characterised by specific dietary patterns. Overall, Italy and Greece have a dietary pattern characterised by plant foods (except potatoes) and a lower consumption of animal and processed foods, compared with the other EPIC countries. France and particularly Spain have more heterogeneous dietary patterns, with a relatively high consumption of both plant foods and animal products. Apart from characteristics specific to vegetarian groups, the UK 'health-conscious' group shares with the UK general population a relatively high consumption of tea, sauces, cakes, soft drinks (women), margarine and butter. In contrast, the diet in the Nordic countries, The Netherlands, Germany and the UK general population is relatively high in potatoes and animal, processed and sweetened/refined foods, with proportions varying across countries/centres. In these countries, consumption of vegetables and fruit is similar to, or below, the overall EPIC means, and is low for legumes and vegetable oils. Overall, dietary patterns were similar for men and women, although there were large gender differences for certain food groups. Conclusions: There are considerable differences in food group consumption and dietary patterns among the EPIC study populations. This large heterogeneity should be an advantage when investigating the relationship between diet and cancer and formulating new aetiological hypotheses related to dietary patterns and disease. Of all the environmental exposures diet is a universal exposure, comprising a complex mixture of different compounds that varies over time, space and according to a number of historical, ethnic, religious, agricultural, socioeconomic and psychological factors, at the individual and population levels 1 . The lack of unbiased dietary methods and the difficulty of measuring long-term individual cumulative dietary exposure, when measurements are obtained at fixed point(s) in time of the subject's life, are important methodological limitations for estimating diet accurately as an exposure 2,3 . These drawbacks may partly explain the lack of consistency in estimates of the association between diet and disease, particularly with regard to cancer 4 , and the attenuation in relative risk estimates of the actual relationship between diet and the outcome diseases due to random errors in dietary measurements 5 -7 . Another emerging possible ...

BackgroundWhile the timing of reproductive events varies across populations, a downward trend in age at menarche has nevertheless been reported in most of the developed world over the past century. Given the impact of change in age at menarche on health conditions, this study sought to examine secular trends in age at menarche among women living in Navarre (Northern Spain) who participated in a population-based breast cancer screening programme.MethodsThe study was based on 110545 women born from 1925 to 1962. Trends were tested using a linear regression model, in which year of birth was entered continuously as the predictor and age at menarche (years) as the response variable, using size of town and region of birth as covariates.ResultsAmong women born in Navarre between 1925 and 1962, age at menarche declined steadily from an average of 13.72 years in the 1925-1929 birth-cohorts to 12.83 years in the 1958-1962 birth-cohorts. Controlling for size of town or city of birth, age at menarche declined by an average of 0.132 years every 5 years over the period 1925-1962. This decline was greater in women born in rural versus urban settings. Trends were also different among regions of birth.ConclusionWe report a population-based study showing a downward trend in age of onset of menarche among Spanish women born in the period 1925-1962, something that is more pronounced among women born in rural settings and varies geographically.

Objective: To assess the consumption of vegetables and fruits (V&F) in adults from ®ve regions in Spain according to sex, age and educational level. Design: Cross-sectional study within the members of the EPIC cohort in Spain. Setting: Three regions of the north of Spain (Asturias, Guipu Âzcoa and Navarra) and two regions of the south of Spain (Granada and Murcia). Subjects: 41 448 healthy volunteers (15 365 men, 25813 women), aged 29±69 y. Interventions: Information on habitual diet during the previous year was collected by means of a computerised version of a diet history questionnaire. Results: Among men, the mean daily consumption of vegetables and of fruits was 273.7 g (3.4 servings) and 348.3 g (4.4 servings) respectively. Among women, the corresponding vegetables and fruit intakes per day were 244.4 g (3.1 servings) and 349.4 g (4.4 servings). The total V&F intake tended to increase with age and educational level. Overall, 74% of subjects consumed 400 gad (5 servings) of vegetables and fruit. Conclusions: Consumption of vegetables and fruits in healthy adults in Spain is considerably higher than in most European countries and the United States; this complies with what is considered to be the Mediterranean diet. Despite some regional differences, there were no clearly differentiated patterns of V&F intake between southern and northern regions within Spain.

RESUMENFundamento. Existe un consenso general en la recomendación de aplicar con carácter poblacional programas de cribado de cáncer de mama. El Departamento de Salud puso en marcha en el año 1990 un Programa de Detección Precoz de Cáncer de Mama, con el objetivo de disminuir la mortalidad por este tumor, que en este momento está finalizando su séptima vuelta. El presente trabajo tiene como objetivo analizar los resultados obtenidos, describiendo los principales indicadores de proceso y predictores de impacto según los factores más determinantes. Material y métodos.Se han recopilado los datos de las exploraciones de cribado realizadas desde el comienzo del programa hasta el 31 de diciembre de 2002 clasificando por edad en el momento de la exploración, tipo de exploración y vuelta.Resultados. Durante el período se han realizado 354.575 exploraciones. Indicadores de proceso: en el 11,22% se recomendó realizar una exploración adicional. El 0,67% de las mujeres han sido enviadas a unidades hospitalarias para valoración diagnóstica. Se ha confirmado un tumor maligno en el 0,34% de las mujeres exploradas. A 8,37% de las mujeres se les ha recomendado realizar una exploración intermedia antes de la nueva de cribado.Indicadores predictores de impacto: la captación global es de 88,68%, la participación global de 87,91% y la adherencia al programa de 96,32%. La tasa de detección alcanza la cifra de 3,95 por 1.000 mujeres exploradas. De los tumores detectados, 19,17% son intraductales, y 37,23% de los invasivos tienen un tamaño igual o inferior a 10 mm. El 74,89% del total de tumores no presenta invasión ganglionar. Conclusiones.Considerando las referencias que para programas de cribado establece la Unión Europea, todos los indicadores predictores calculados superan los estándares exigidos.Palabras Clave. Cáncer de mama. Cribado poblacional. Indicadores. Evaluación. ABSTRACTBackground. There is a general consensus on the recommendation of applying breast cancer screening programs to the population. In March 1990, the Health Department set in motion a Program for the Early Detection of Breast Cancer, with the aim of reducing mortality due to this tumour. At present, the program is reaching the end of its seventh round. The present study aims to analyze the results obtained, describing the principal indicators of the process and the impact predictions according to the most determinant factors.

No abstract