Cross-Sectional Prevalence Studies
Zuber D. Mulla
Objectives 1. Appreciate that both exposure and outcome status are measured at the same point in time in a cross-sectional study.
2. Calculate and interpret the prevalence ratio and prevalence odds ratio. 3. List the main advantages and disadvantages of a cross-sectional design.
Branches of Epidemiology • Descriptive • Analytic Epidemiology in Medicine by Hennekens & Buring, 1987
Descriptive Epidemiology • Concerned with distribution of disease • Propose hypotheses
Analytic Epidemiology • Focuses on causes of diseases • Tests hypotheses
Overview of Study Designs • Descriptive • Individuals (case series) • Populations (correlational studies)
• Analytic • Observational studies • Intervention studies
Overview of Study • Analytic – Observational studies • Case-control
–Retrospective case-control –Prospective case-control –Nested case-control –Density sampling case-control –Case-cohort
• Cohort (retrospective or prospective) • Cross-sectional prevalence survey
Overview of Study Designs • Analytic –Intervention studies • Nonrandomized controlled clinical trial
• Randomized controlled clinical trial
Independent Variable • • • • • • •
X variable Exposure Risk factor Explanatory Covariate Predictor Regressor
Dependent Variable • Y variable • Outcome • Response • Regressand • Endpoint
• Exposure (independent variable) and the outcome (dependent variable) assessed at the same point in time • No follow-up of subjects • (Panel study)
Cross-Sectional Prevalence Studies • Ideal for chronic diseases
• A disease of short duration, in general, not suitable for a crosssectional study because few individuals would have the disease at any one point in time
Example • You identify a random sample of adults using random digit dialing • Exposure: Obesity (BMI ≥ 30) • Outcome: Depression • Does prevalence of depression vary by obesity status?
Calculating a Prevalence Ratio (PR) Not Depressed depressed
PR = Prevalence of Depression in Obese divided by Prevalence of Depression in Non-obese
A A+B PR = C C+D
Interpretation • PR=1.30 • Obese individuals were 30% more likely than non-obese individuals to have depression.
Advantage • No follow-up • May be quicker to conduct
Disadvantage • Time sequence/Temporality (what came first?) may be not intact: did the exposure really precede the outcome? • For exposures that do not change over time such as race or blood type, no problem.
What came first? Obese
PR vs. Prevalence OR • You can also calculate an odds ratio from a cross-sectional study. • Same formula as before: AD / BC • Logistic regression yields ORs. • Binomial regression will yield RR for a cohort study and PR for a crosssectional study
NHANES • National Health and Nutrition Examination Survey • www.cdc.gov/nhanes
Cited Reference Hennekens CH, Buring JE. Epidemiology in medicine. Boston: Little, Brown and Company, 1987.