Objective: To assess, particularly in longitudinal studies, how close or far the food intakes of population groups are from a reference dietary pattern.

Design: Computation of an index, called the Mediterranean Adequacy Index (MAI), by dividing the sum of the percentage of total energy from typical Mediterranean food groups by the sum of the percentage of total energy from non-typical Mediterranean food groups. The reference Italian-Mediterranean diet utilised was that of subjects from Nicotera in 1960.

Setting: Different areas in Italy.

Subjects: Men aged 45-65 years at the start of the study from rural areas of Italy in the Seven Countries Study followed for 26 years (Crevalcore and Montegiorgio), elderly men and women from Perugia followed for 11 years, men and women from Pollica (Salerno) followed for 32 years, and families from Rofrano (Salerno) followed for 41 years.

Results: The median value of MAI among 40-59-year-old men from Nicotera in 1960 was 7.2; in Crevalcore it was 2.9 in 1965 and 2.2 in 1991; and in Montegiorgio, 5.6 in 1965 and 3.9 in 1991. In Pollica, the values in men ranged from 5.6 to 6.3 in 1967 and from 2.4 to 4.5 in 1999; for women, the corresponding ranges were 4.2-7.2 and 2.7-4.1. In elderly men of Perugia, median MAI value was 4.9 in 1976 and 3.2 in 1987; for women, the corresponding values were 3.1 and 2.6.

Conclusions: The diet of these Italian population groups has changed over the last four decades, progressively abandoning the nutritional characteristics of the reference Italian-Mediterranean diet. The MAI proposed is simple to compute; it has satisfactory discriminating power particularly for longitudinal dietary data with only a few limitations.