Concise Biostatistical Principles & Concepts

About the Book

Biostatistics deals with making sense of data. While statistical inference is essential in our application of the research findings to clinical decision-making regarding the care of our patients, statistical inference without clinical relevance or importance can be very misleading and even meaningless. This textbook has attempted to deemphasize p value in the interpretation of clinical and biomedical data by stressing the importance of confidence intervals, which allow for the quantification of evidence. For example, a large study due to a large sample size that minimizes variability may show a statistically significant difference while in reality the difference is too insignificant to warrant any clinical relevance. Covers these relevant topics in biostatistics: Design Process, Sampling & Reality in Statistical Modeling Basics of Biostatistical Reasoning & Inference Central Tendency Theorem & Measures of Dispersion Most commonly used & abused parametric test – t test Most commonly used & abused non-parametric test – chi squared statistic Sample size and power estimations Logistic/Binomial Regression Models – Binary Outcomes Time-to-Event Data - Survival Analysis & Count Data – Poisson Regression ANOVA, ANCOVA – Mixed Effects Model (Fixed and Random), RANOVA,GEE Simple & Multiple Linear Regression Models Correlation Analysis (Pearson & Spearman Rank) Clinical & Statistical Significance – p value as a function of sample size Clinical and biomedical researchers often ignore an important aspect of evidence discovery from their funded or unfunded projects. Since the attempt is to illustrate some sets of relationships from the data set, researchers often do not exercise substantial amount of time in assessing the reliability and validity of the data to be utilized in the analysis. However, the expected inference or the conclusion to be drawn is based on the analysis of the un-assessed data. Reality in statistical modeling of biomedical and clinical research data remains the focus of scientific evidence discovery, and this book. This text is written to highlight the importance of appropriate design prior to analysis by placing emphasis on subject selection and probability sample and the randomization process when applicable prior to the selection of the analytic tool. In addition, this book stresses the importance of biologic and clinical significance in the interpretation of study findings. The basis for statistical inference, implying the quantification of random error is random sample, which had been perpetually addressed in this book. When studies are conducted without a random sample, except when disease registries/databases or consecutive subjects are utilized, as often encountered in clinical and biomedical research, it is meaningless to report the findings with p value.

About the Author

Laurens Holmes, Jr. Educated at the Catholic University of Rome, Italy , University of the Health Sciences, Antigua, School of Medicine, University of Amsterdam, Faculty of Medicine, and the University of Texas, Texas Medical Center, School of Public Health, Laurens (Larry) Holmes, Jr., is currently a clinical epidemiologist (Orthopedic Department), Head of the Epidemiology Laboratory at the Nemours Center for Childhood Cancer Research, and Chief Methodologist at the Nemours/A.I.duPont Children Hospital , Office of Health Equity & Inclusion. He is also an adjunct professor of clinical trials and molecular epidemiology at the Department of Biological Sciences, University of Delaware, Newark, DE. He is recognized for his work on epidemiology and control of prostate cancer, but has also published papers on other aspects of hormonally-related malignancies, cardiovascular and chronic disease epidemiology utilizing various statistical methods. Dr. Holmes is a strong proponent of reality in the statistical modeling of cancer and non-experimental research data, where he presents on the rationale for tabular analysis in most non-experimental research data which are often not randomly sampled, and hence meaningless to apply statistical inference to such data. Franklin Opara Dr. Opara completed his undergraduate from Texas Southern University, earned his master degree from George Washington University, then received his medical degree from UTESA- School of Medicine, Dominican Republic, and also obtained his doctorate degree from Walden University, Minnesota. Dr. Opara currently heads the Health Policy Research Division at the American Health Research Institute, and together with Dr. Holmes, he examines the role of race/ethnicity in geo-epidemiologic mapping of diseases. Within other positions, he has served as a Chief Consultant at Priority Women’s Health Alliance specialized in clinical issues in women’s healthcare. Dr. Opara is best recognized for his contributions in health disparities in cardiovascular and chronic diseases namely hypertension, and also clinical research designs and management in the areas of teen pregnancy, healthcare outcomes and injury prevention. He is a co-author and has published many papers in scientific journal on disparities in health outcomes.