Jacques Allard

Is Serum Ferritin an Additional Cardiovascular Risk Factor for All Postmenopausal Women

Background: Most of the studies on cardiovascular disease (CVD) risk factors in menopause have focused on serum lipid(lipoprotein) abnormalities and were conducted in populations which were not well controlled for several important influential factors. Methods: Two homogenous groups of 30 apparently healthy Caucasian premenopausal women and 3–5 years postmenopausal women who were nonobese, nonsmoking and not using estrogen were compared in a well-controlled cross-sectional design. Fasting serum ferritin and plasma total homocysteine (tHcy) were evaluated concomitantly to classical serum lipid(lipoprotein) risk factors. Relationships between risk factors and the influence of other contributing variables such as diet and body weight were also examined. Results: Serum total cholesterol (p < 0.01), low-density lipoproteins (LDL; p < 0.05) and triglycerides (p < 0.05) of postmenopausal women were greater than that of their menstruating counterparts, even though they ate a CVD-preventive diet, had similar body weight and body fat distribution. Their serum ferritin was almost 3-fold greater (p < 0.0001) but was still within normal limits, except for the 38.5% of postmenopausal women who exhibited values above the 80 µg/l limit that has been associated with sharp increases in the rate of heart disease in either gender. Serum ferritin was low in one third of the postmenopausal group (as low as in the premenopausal control group, whose dietary iron intake was slightly below the nutritional recommendation). The mean plasma tHcy of the postmenopausal group was almost twice as elevated (p < 0.0001). Both ferritin and tHcy were found to be linked to serum cholesterol. The correlation between tHcy and triglycerides was also significant. Conclusion: Early menopause is not associated with blood iron overload and CVD risk factor in an important proportion of women.

Robust centroid method

This paper describes a robust centroid method, which is a variant of principal component analysis. A genetic local search algorithm is presented to perform the calculations. Simulations are carried out to appraise the performance of the genetic local search algorithm. A real data set with missing data and multiple outliers is analyzed.

The Z-Plot: A Graphical Procedure for Contingency Tables with an Ordered Response Variable

Publisher Summary This chapter provides an overview of the Z-Plot, which serves as graphic procedure for contingency tables with an ordered response variable. The Z-plot is used as an exploratory data analytic tool to screen two-way and multiway contingency tables, but no formal inferential conclusion can be deduced from it. However it may help to choose among the formal statistical approaches such as Goodmans R or R + RC association model, McCullaghs proportional odds model, or more complicated models, for ordered categorical data that are available to analyze such datasets. The chapter illustrates the use of Z-plots using three examples: (1) Z-plot of evolution of the opinion of a sample of youths on military service, (2) Z-plot representing the results of a taste-testing experiment for five treatments, and (3) parallel Z-plots for the severity of persistent wheeze according to age, city, and maternal smoking status.

Acknowledgement to the 2004 Reviewer

Martinez, A., Pamplona, Spain Mokuda, O ., Ichihara-City, Japan Moser, U., Basel, Switzerland Neuhäuser-Berthold, M., Giessen, Germany Pallauf, J., Giessen, Germany Palou, A., Palma de Mallorca, Spain Pedersen, J., Oslo, Norway Pool-Zobel, B., Jena, Germany Porrini, M., Milan, Italy Rethy, L., Budapest, Hungary Rimbach, G.H., Reading, UK Rodriguez-Amaya, D.B., Campinos, Brazil Rust, P., Vienna, Austria Salem, N., Rockville, Md., USA Schümann, K., Munich, Germany Simopoulos, A.P., Washington, D.C., USA Singer, P., Bensheim-Auerbach, Germany Sirtori, C.R., Milan, Italy Sjöström, M., Huddinge, Sweden Somoza, V., Garching, Germany Stahl, W., Düsseldorf, Germany Trichopoulo, A., Athens, Greece Trichopoulos, D., Boston, Mass., USA Trippo, U., Bergholz-Rehbrücke, Germany Uauy, R., Santiago, Chile Urbanek, R., Vienna, Austria Villalpando, S., Cuernavaca, Mexico Von Rücker, A., Bonn, Germany Von Schacky, C., Munich, Germany Wahlqvist, M.L., Melbourne , Australia Wahrendorf, J., Heidelberg, Germany Watzl, B., Karlsruhe, Germany Wenk, C., Zürich, Switzerland Winisch, W., Vienna, Austria Wisker, E., Kiel, Germany Zunft, H.-J., Bergholz-Rehbrücke, Germany Adam, O., Munich, Germany Antal, M., Budapest, Hungary Arthington, J., Ona , Fla., USA Berg, A., Freiburg, Germany Bitsch, I., Giessen, Germany Boeing, H., Bergholz-Rehbrücke, Germany Böhm, V., Jena, Germany Classen, H., Hohenheim, Germany Clinton, S., Columbus, Ohio, USA Cunningham, C., Winston-Salem , N.C., USA De Blas, J.C., Madrid , Spain Dobos, D., Essen, Germany Fürst, P., Stuttgart, Germany Gadek-Wesierski, J., Vienna, Austria Genser, D., Vienna, Austria Götz, M., Vienna, Austria Grassmann, J., Munich, Germany Hagfors, L., Umea, Sweden Hauner, H., Munich, Germany Hautvast, J., Wageningen, The Netherlands Herrera, E., Madrid, Spain Joost, H.J., Bergholz-Rehbrücke, Germany Kafatos, A., Heraklion, Greece Kalapos, M., Budapest, Hungary Kar, A ., Deoghar, India Kato, Saiama, Japan Keller, U., Basel, Switzerland Knasmüller, S., Vienna, Austria Koletzko, B., Munich, Germany König, J., Vienna, Austria Krawinkel, M., Giessen, Germany Kudlácková, M., Bratislava, Slovak Republik Laplace, J.P., Paris, France Lemmens, R., Vienna, Austria Linseisen, J., Munich, Germany Magee, P., Belfast, UK Marangoni, F., Milan, Italy

Chapter 8 – The Z-Plot: A Graphical Procedure for Contingency Tables with an Ordered Response Variable

Publisher Summary This chapter provides an overview of the Z-Plot, which serves as graphic procedure for contingency tables with an ordered response variable. The Z-plot is used as an exploratory data analytic tool to screen two-way and multiway contingency tables, but no formal inferential conclusion can be deduced from it. However it may help to choose among the formal statistical approaches such as Goodmans R or R + RC association model, McCullaghs proportional odds model, or more complicated models, for ordered categorical data that are available to analyze such datasets. The chapter illustrates the use of Z-plots using three examples: (1) Z-plot of evolution of the opinion of a sample of youths on military service, (2) Z-plot representing the results of a taste-testing experiment for five treatments, and (3) parallel Z-plots for the severity of persistent wheeze according to age, city, and maternal smoking status.

The Z-Plot

Publisher Summary This chapter provides an overview of the Z-Plot, which serves as graphic procedure for contingency tables with an ordered response variable. The Z-plot is used as an exploratory data analytic tool to screen two-way and multiway contingency tables, but no formal inferential conclusion can be deduced from it. However it may help to choose among the formal statistical approaches such as Goodmans R or R + RC association model, McCullaghs proportional odds model, or more complicated models, for ordered categorical data that are available to analyze such datasets. The chapter illustrates the use of Z-plots using three examples: (1) Z-plot of evolution of the opinion of a sample of youths on military service, (2) Z-plot representing the results of a taste-testing experiment for five treatments, and (3) parallel Z-plots for the severity of persistent wheeze according to age, city, and maternal smoking status.