Comprehensive Exam Study Guide (Epidemiology)

[Last update: 4/14/04]

This brief document is intended to help you prepare for the MPH comprehensive exam in epidemiology. Reference is made to relevant chapters in Epidemiology Kept Simple (EKS), both editions. 

Format of the exam. The exam is mostly open ended but also includes several multiple questions. It is a closed-book test. Formulas are provided. Please bring a calculator to the exam. There will questions about basic definitions. You will also interpret several study abstracts. The following topics are addressed:

BASIC DEFINITION OF TERMS � epi and public health definitions as covered in Chapter 1 (both editions).

DISEASE FREQUENCY / INCIDENCE & PREVALENCE The three ways to quantify disease in a population are: (1) incidence proportion (risk, cumulative incidence), (2) incidence rate (incidence density,  person-time rate, and (3) prevalence. (Although these statistics differ, lay people will often call them "risks" or "rates.") You should know what goes into the numerator and denominator of each of these measures. These statistics are covered in Chapter 6 of both editions. 

MEASURES OF ASSOCIATION AND POTENTIAL IMPACT The effect of an exposure on risk can be quantified in absolute terms (risk difference) or relative terms (risk ratio). The impact of an exposure can be quantified as an attributable fraction. These measures are covered in Chap 8 (both editions).

EPIDEMIOLOGIC STUDIES Epidemiologists use five main study types to investigate relations in populations: (1) trials (including community trials, field trials, and clinical trials), (2) ecological studies, (3) cross-sectional field surveys, (4) cohort studies, and (5) case-control studies. These are covered in Chap 9 in the first edition and in Chapters 9 - 11 in the second edition.

ERROR IN EPIDEMIOLOGIC RESEARCH - There are two types of errors in any type of study: random error (imprecision) and systematic error (bias). We use p values and confidence intervals to address random errors. Systematic error comes in three varieties: selection bias (study subjects are not representative of the larger population), information bias (non-random measurement error), and confounding (systematic distortion due to extraneous factors). Error in research is discussed in Chap 10 and 11 in the first edition and in Chap 12 in the second edition.