••• CHAPTER 64^ Basic Epidemiology and Clinical Biostatistics^643
❍ How is RR calculated?
RR is the cumulative incidence in the exposed subjects divided by the cumulative incidence in the unexposed
subjects.
❍ How is OR calculated?
The OR is the number of exposed cases times the number of unexposed controls divided by the number of
unexposed cases times the number of exposed controls.
❍ How is AR calculated?
AR is the cumulative incidence in the exposed subjects minus the cumulative incidence in the unexposed subjects.
❍ How is PAR calculated?
PAR is the cumulative incidence in the population (exposed and unexposed) minus the cumulative incidence in
the unexposed.
❍ What study designs are used to study associations?
Observational study designs are used to study association. Observational study designs include cohort studies,
cross-sectional studies, surveillance studies, and case-control studies (see definitions below).
❍ What is causation?
A possible causal relationship is suggested when studies of disease provide evidence for a significant association
between a specific factor and a disease outcome. Experimental studies are designed to test whether a causal
relationship exists between a specific factor and the disease.
❍ What types of experimental study designs are used to study causation?
Experimental study designs used to study causation include randomized controlled trials (RCTs),
group-randomized trials, and multicenter RCTs (see definitions below).
❍ What issues decrease the strength of analytic epidemiology studies?
Selection bias, incidence-prevalence bias, measurement bias, confounders, and effect modifiers can all decrease the
strength of studies.
❍ What is selection (volunteer) bias?
Selection (volunteer) bias refers to the differences between characteristics of people selected for a study and those
who are not selected. For example, volunteers for a study tend to be healthier and better educated.
❍ What is incidence-prevalence bias?
Incidence-prevalence bias occurs where there is a loss of cases (by death or recovery) due to significant periods of
time between exposure and development of the disease or condition.
❍ What is measurement bias?
Measurement bias occurs when individual measurements or classifications of disease or exposure are inaccurate.