OBJECTIVE: Body mass index (BMI) was determined in a population of school students from three provinces of central Italy. Fasting serum leptin concentrations were assayed in a large number of subjects from the same area, to determine their distribution as plotted against the standard deviation score (z-score) of BMI. DESIGN, SUBJECTS AND MEASUREMENTS: Height and weight were recorded from 31 170 subjects (16 175 male and 14 995 female), aged 3 ± 18 y, to construct BMI charts of children and adolescents from central Italy. Percentiles and zscore were calculated using the LMS method of Cole. Serum leptin concentrations were assayed in 1929 subjects (996 male and 933 female) after overnight fasting. RESULTS: BMI percentiles of central Italy were higher than those from standards of other European and USA populations. When plotted against the z-score of BMI, serum leptin values were distributed according to an exponential curve, showing a steep pattern and a wide distribution, as BMI values increased. The hypothesis of the existence of two subgroups, based on a different relation between leptin and BMI, was veri®ed and a separation point between the two subgroups was identi®ed using cluster analysis, discriminant analysis and a novel method developed by our group, hereafter referred to as`regression clustering'. This method allows identi®cation of the value of the independent variable (z-score of BMI) which can be taken as a separation point. This analysis provided the best results and indicated the following separation points: central Italy standard, z-score 0.72 (76.4 th percentile) for males and z-score 0.69 (75.5 th percentile) for females; French standard (the one suggested for a European population by the European Childhood Obesity Group, ECOG), z-score 1.46 (92.8 th percentile) for males and z-score 1.96 (97.5 th percentile) for females. Similar but variable results were obtained when the same analysis was performed on serum leptin concentration, subdivided according to pubertal development (stage I, stage II ± III, stage IV ± V). CONCLUSIONS: Children and adolescents from central Italy had greater BMI percentiles when compared to other European populations. Fasting serum leptin concentrations showed a distribution pattern related to z-score, thus allowing to identi®cation of two different subgroups. The z-scores of BMI, identi®ed as separation points, indicated a trend to leptin production by adipocytes that could be taken as indicators of signi®cant increases of fat mass. This study proposes criteria and a statistical approach that could be useful in the identi®cation of BMI cut-off values when screening children and adolescents for overweight.

Objective: The in¯uence of weight excess reduction on height and height velocity of obese subjects should be evaluated on the basis of appropriate standards, since the pattern of growth of obese subjects is different from that of normal weight subjects. Design, subjects and measurements: Height, weight and triceps skinfold thickness were recorded from 17987 school subjects (9256 males and 8731 females), 3 ± 18 y of age, from three provinces of central Italy, and a growth reference curve of height was constructed. Using BMI (as computed using the tables of Rolland-Cachera et al) and triceps skinfold thickness, normal-weight subjects (NWS) and obese subjects (OS) were identi®ed and speci®c reference curves (mean AE s.d. every sixth month of age) were developed for both groups. Centiles of height were also calculated for OS. Various (2 ± 4) measurements of height in school subjects were performed and a graph of height velocity (HV) was constructed in NWS and in OS using the JPPS method. The yearly mean AE s.d. of HV was also calculated, based on square root transformed data (in order to realise a Gaussian distribution), deriving from successive measurements in total subjects, in NWS and in OS. The z-scores of height and of the square root of HV were calculated in 217 obese subjects (125 males and 92 females) before and during a weight excess reduction programme (WERP). Obese subjects in WERP who showed a reduction of z-score of BMI were considered as`responsive'; those who either maintained or showed an increase of z-score of BMI were considered as`non-responsive'. Obese subjects in WERP were followed for 1 ± 4 y, giving the following results: 0 ± 1 y, 142 responsives and 75 non-responsives; 0 ± 2 y, 76 responsives and 33 non-responsives; 0 ± 3 y, 35 responsives and 30 non-responsives; 0 ± 4 y, 24 responsives and 18 non-responsives. Results: Compared to NWS, OS showed a signi®cantly greater HV in 4 ± 9 y males and in 4 ± 8 y females, but in older children the pubertal spurt was reduced and more precocious. As a result, the height of OS, which was greater in 3 ± 13 year-old males and in 3 ± 11.5 year-old females, subsequently showed a reduction, as compared to that of NWS, in 16 ± 18 year-old males and in 13 ± 18 year-old females. In both responsive and non-responsive groups of obese subjects in WERP, the z-scores of height showed a reduction during WERP when evaluated using the reference curve of the total school population. In contrast, when their growth was evaluated according to the obese-speci®c reference curve, no signi®cant variation was observed comparing both z-scores before and during the WERP. Conclusions: More appropriate information on the growth of obese subjects may be obtained when evaluating the height and HV according to obese-speci®c reference standards from the same population of origin. Adopting this modality, no signi®cant variation of height resulted during WERP in obese children.

Multicollinearity can seriously affect least-squares parameter estimates. Many methods have been suggested to determine those parameters most involved. This paper, beginning with the contributions of Belsley, Kuh, and Welsch (1980) and Belsley (1991), forges a new direction. A decomposition of the variable space allows the near dependencies to be isolated in one sub-space. And this, in turn, allows a corresponding decomposition of the main statistics, as well as a new one proposed here, to provide better information on the structure of the collinear relations.

The paper explores the reasons that have made the Internet even more important for the training of researchers, both for pure statisticians and scientists who use statistics for analysing experimental data. The author discusses various aspects of the role of the Internet and focuses on those contributions that populate the network with valuable tools and services (electronic journals, electronic books, virtual labs, data archives and examples of data analysis, etc.). An extensive bibliography has been provided to guide new users, through the millions of sites of the chaotic World Wide Web, towards the most interesting places for people involved in statistics.

Blood glucose, plasma insulin and plasma glucagon were determined during an arginine test at 0, 30 and 60 min in 1 to 7 and 30 day infants. The results were compared with those obtained in the 2-10-year-old children. Basal levels of blood glucose increased from the 1st to the 30th day; by the 30th day they were comparable with those observed in the 2-10-year-old children. Basal levels of plasma insulin were higher in 1 day infants, lower in 7 day infants and similar to those of the 2-10-year-old children in the 30 day infants. Basal values of plasma glucagon were higher in 1 day infants than in the 2-10-year-old children; the lowest levels were found in 7 day infants. After arginine, blood glucose increased in the 2-10-year-old children, the variations were negligible in the 1 to 7 and 30 day infants. Plasma insulin increased in 2-10-year-old children, there was a slight increase in 1 and 30 day infants and a negligible one in the 7 day infants. Plasma glucagon increased in the 2-10-year old children, and was similar in 1 day infants; the increase was reduced in 30 day infants and virtually absent in 7 day infants. These results suggest that there is a transitory alpha cell insensitivity to arginine during the first week of life.