Design of Observational Studies

An observational study is an empiric investigation of effects caused by treatments when randomized experimentation is unethical or infeasible. Observational studies are common in most fields that study the effects of treatments on people, including medicine, economics, epidemiology, education, psychology, political science and sociology. The quality and strength of evidence provided by an observational study is determined largely by its design. Design of Observational Studies is both an introduction to statistical inference in observational studies and a detailed discussion of the principles that guide the design of observational studies.Design of Observational Studies is divided into four parts. Chapters 2, 3, and 5 of Part I cover concisely, in about one hundred pages, many of the ideas discussed in Rosenbaum’s Observational Studies (also published by Springer) but in a less technical fashion. Part II discusses the practical aspects of using propensity scores and other tools to create a matched comparison that balances many covariates. Part II includes a chapter on matching in R. In Part III, the concept of design sensitivity is used to appraise the relative ability of competing designs to distinguish treatment effects from biases due to unmeasured covariates. Part IV discusses planning the analysis of an observational study, with particular reference to Sir Ronald Fisher’s striking advice for observational studies, "make your theories elaborate."The second edition of his book, Observational Studies, was published by Springer in 2002.

From the book reviews:“The book is both a pragmatic how-to manual and a delightfully written manifesto. … The book contains four parts. … People very new to statistics could read this book quite readily, with the caveat that mathematical notation, language, and prose is used far more than in most classic introductory statistics books. This book seems most useful for statisticians and applied statisticians who are looking to learn about a new area that they might not have been exposed to before.” (Luke W. Miratrix, Psychometrika, Vol. 79, 2014)“This is for those who want to improve whatever their basic design is—for those people, this is a very good book. … have several quick, but useful, guides: key elements in a design, solutions to common problems (very useful), a symbol glossary, a listing of acronyms, a glossary of statistical terms, suggested readings for a course on the design of observational studies, and an index. … anyone who is not an expert on the varieties of matching will profit from reading this.” (Richard Goldstein, Technometrics, Vol. 53 (2), May, 2011)Finally, the book covers all the relevant issues in designing and analyzing treatment effects in observational studies, with the exception of observer bias, i.e., the bias present when the assessment of the outcomes are not valid; see Haro et al. (2006) for further details. Nowadays, it is yet unusual to address hidden biases in observational studies and, therefore, this book is an essential reading for statisticians who want to go a step beyond from the likely naïve sentence: 'We assume no unmeasured confounding in the study’.” (Journal of Biopharmaceutical Statistics , 2011, Issue 1)“Graduate students and researchers in statistics, biostatistics, econometrics, or academic researchers in statistically oriented fields of psychology and social sciences. … ‘Design of Observational Studies’ talks about statistics. … The book will be suitable for a seminar course for graduate students with previous knowledge of the subject area, or practicing statisticians seeking guidance in design of observational research and a language to discuss the issues. ‘Design of Observational Studies’ is an important book.” (Erkki P. Liski, International Statistical Review, Vol. 78 (1), 2010)