ATSDR has used a basic test battery of kidney biomarkers to examine early indicators of kidney dysfunction in several health investigations of persons who live near hazardous waste sites. Six biomarkers have been used recently in a longitudinal study conducted by ATSDR. They are alanine aminopeptidase (AAP), N-acetyl-B-D-glucosaminidase (NAG), albumin (ALB), retinol-binding protein (RBP), intestinal alkaline phosphatase (IAP), and epidermal growth factor (EGF). Often it is desirable to explore the data through graphical methods. This may lead to the realization of trends, the recognition of important factors, and the detection of outliers. However, the correlation of the six biomarkers and the high dimension of the data can create problems when exploring the data through graphical techniques. To reduce the dimensionality of the data, a principal components analysis was performed. The goal of principal components analysis is to linearly transform possibly correlated variables into a small number of uncorrelated variables called principal components. The principal components are then plotted to graphically reveal patterns in a complex data set that are not readily seen by visual inspection. A demonstration of how this may be used to explore the kidney biomarker data for possible trends and the detection of “unusual” data points (outliers) will be presented. This will assist researchers in recognizing people who may be at greatest risk of kidney dysfunction. In the longitudinal study, this technique leads to the recognition of clusters that share similar characteristics; e.g. gender and site.
Learning Objectives: Participants will learn some exploratory techniques for analyzing kidney biomarker data. These techniques will be useful for identifying unusual values, which is beneficial for public health researchers. The methods will also help idnetify important factors for modelling the biomarkers. A data set from a recent ATSDR investigation will be used for illustration.
Keywords: Indicators, Biostatistics
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.