Epidemiology: An Introduction

Citation

Rothman, K. J., Huybrechts, K. F., & Murray, E. J. (2024). Epidemiology: An Introduction (3rd ed.). Oxford University Press.

Chapter Summary

The third edition runs to a little over 300 pages of text in fifteen chapters, each ending with a list of questions for discussion. The preface states the book's aim directly: epidemiology is more than applied statistics, and the emphasis falls on principles and concepts rather than formulas and computation. The first two editions were written by Rothman alone; the preface describes this one as a joint revision in which every chapter was revised, the coverage of infectious disease was expanded after the COVID-19 pandemic, and two chapters (14 and 15) are new.

Chapter 1: Introduction to Epidemiologic Thinking

The book opens, unusually, with confounding. The preface explains the choice: confounding shows why epidemiology is more than "common sense" and why sound principles are needed to avoid fallacious inferences.

Chapter 2: Pioneers in Epidemiology and Public Health

Short profiles of figures who shaped the field, among them Avicenna, Graunt, Snow, Nightingale and Lane-Claypon. In this edition the chapter has been expanded to stress collaborative research, from the studies that linked tobacco smoking to disease to the growth of social epidemiology.

Chapter 3: What Is Causation?

A general model of causation and a discussion of causal inference. The preface argues that these ideas are often skipped in scientific education but belong in any introduction to epidemiology.

Chapter 4: Measures of Disease Occurrence, Association, and Causal Effects

The basic epidemiologic measures: how to quantify how often disease occurs, how to compare occurrence between groups, and how such comparisons relate to causal effects.

Chapter 5: Types of Epidemiologic Studies

The main study designs, presented as ways of making the comparisons defined in Chapter 4.

Chapters 6–7: Dealing with Biases; Random Error and the Role of Statistics

Two chapters on error. Systematic error (bias) comes first, then random error. The authors favour interpreting results by the size of an association and the precision of its estimate, not by significance tests against an arbitrary threshold.

Chapters 8–9: Analyzing Simple Epidemiologic Data; Stratified Analysis

Chapter 8 introduces the basic methods for estimating epidemiologic effects from simple data; Chapter 9 extends them to stratified data, the standard way of controlling confounding in a table.

Chapters 10–11: Using Regression Models in Epidemiologic Analysis; Measuring Interactions

Regression modelling and interaction, presented in elementary terms as groundwork for later courses. The preface notes that the text marks the boundary between an epidemiologic approach to these topics and non-epidemiologic approaches that can mislead the analysis.

Chapter 12: Epidemiology in Clinical Settings

Clinical epidemiology: screening, prognosis, and assessing the safety and effectiveness of treatments.

Chapter 13: Infectious Disease Epidemiology

The key concepts and terminology particular to infectious disease epidemiology, expanded for this edition.

Chapter 14: Field Epidemiology

New in this edition: epidemiology applied to public-health problems, including the principles of disease surveillance, study designs used in the field, and the investigation and control of outbreaks.

Chapter 15: An Introduction to Some Advanced Topics

Also new: a short introduction to a selection of advanced methodological topics, intended as an on-ramp to further reading rather than full coverage.

Key Concepts

Chapters 1–5: Foundations

• Confounding: a mixing of effects in which a third factor distorts the apparent relation between an exposure and a disease; the book's opening example of why epidemiologic reasoning is needed.
• Causal model: a general framework for what it means for a factor to cause disease, the basis for the later discussion of inference.
• Measures of occurrence and effect: quantities that describe how often disease occurs and how much an exposure changes that frequency.
• Study types: the main designs used to compare disease occurrence between exposed and unexposed groups.

Chapters 6–11: Error and Analysis

• Bias (systematic error): error that does not shrink as a study gets larger, treated before random error.
• Estimation over significance testing: judging results by the magnitude of an effect and its precision rather than by a pass/fail threshold.
• Stratification: analysing data within levels of a third variable to remove confounding.
• Interaction: the dependence of one factor's effect on the level of another.

Chapters 12–15: Applications and Extensions

• Clinical epidemiology: epidemiologic methods applied to screening, prognosis and treatment decisions.
• Surveillance and outbreak investigation: the core tasks of field epidemiology.

Critical Analysis

Strengths

• Concepts before calculation. The preface sets out a book about epidemiologic thinking, not a statistics manual, and the order of the contents bears this out: history, causation and measures come before any analysis chapter. For a first course this gives students a reason for each method before they meet it.
• A clear position on statistical inference. The authors state that they interpret results through effect size and precision rather than significance testing, and they keep regression and interaction for after bias, random error and stratification. Students learn a consistent approach from the start.
• Little background required. The preface says that little or no background is needed and that basic notions such as the definition of a cause are developed with simple examples, which makes the book usable outside medicine and public health.
• Revised for current practice. The third edition adds a field epidemiology chapter covering surveillance and outbreaks, expands the infectious disease material after COVID-19, and closes with a chapter pointing to advanced methods. Each chapter ends with discussion questions, and the preface points to Episheet, a public-domain spreadsheet workbook that carries out many of the analyses in the book.

Limitations

• Questions without printed answers. The end-of-chapter questions are framed as a basis for discussion; the preface refers readers to a publisher website for discussion and reader-contributed answers rather than providing worked solutions. Students who want graded drill will need another source.
• Elementary treatment of modelling. Regression and interaction are deliberately introduced only in elementary terms, and Chapter 15 surveys advanced topics without developing them. Students who must fit and interpret models will need a methods text.
• Brief applied chapters. Clinical, infectious disease and field epidemiology each receive a single chapter of about twelve to twenty pages, enough for orientation but not for detailed training in, for example, outbreak investigation.

Where It Fits in the Course

This is the most concept-centred of the introductions in this course. Gordis Epidemiology covers similar ground in a longer, heavily illustrated textbook, with more attention to screening, randomized trials and the use of epidemiology in health services and policy, and with end-of-chapter review questions and answers; the two work well together, one for the reasoning and the other for practice. Essential Epidemiology by Webb, Bain and Page is a full course textbook aimed at public health and the health professions, with more space for applications in prevention, screening, outbreaks and policy.

Rothman, Huybrechts and Murray also prepare the ground for the graduate-level books. Modern Epidemiology, of which Rothman is a co-author, is a comprehensive reference on causal inference, validity and analysis, and Chapters 3, 6, 7 and 15 here form a natural lead-in to it. Mark Woodward's Epidemiology: Study Design and Data Analysis is a statistically oriented text that works through real data sets, and it suits students who move on from Chapters 8–11 to analyse data themselves.

How to Use This Book

The book suits first-year students of public health, medicine, nursing and other health sciences, and anyone who reads epidemiologic research and wants to understand how its conclusions are reached. Comfort with simple arithmetic and proportions is enough; no prior statistics is assumed.

A practical reading plan:

• Core (in order): Chapters 1 and 3–7. They establish causal thinking, the basic measures, the study designs, and the two kinds of error that shape every interpretation. Chapter 2 is short and worth reading early for context.
• Analysis: Chapters 8 and 9 are central for anyone who will compute rates, risks and stratified estimates. Download Episheet and reproduce the book's examples as you go; this turns the concepts into working skills without specialized software.
• Bridging chapters: Chapters 10 and 11 can wait until a statistics or methods course is under way; read them as an orientation to models and interaction.
• Choose by interest: Chapter 12 for clinical students, Chapters 13 and 14 for public health and infectious disease, and Chapter 15 for those planning further methods training.

Work through the discussion questions at the end of each chapter with classmates, since no answer key is printed. Pair the book with a source of numerical exercises, such as the review questions in Gordis Epidemiology, and when you meet a published study, try to name its design, its main sources of bias and how precise its estimates are.