Cross-Sectional and Ecological Studies
~1 min read
Lesson 7 of 20
Notes
Epidemiological studies are divided into descriptive and analytic categories. Descriptive epidemiology studies the distribution of health-related states in specified populations, asking who, what, where, and when. Analytic epidemiology studies determinants — examining associations between exposures and outcomes and inferring causation.
Cross-sectional studies are conducted via surveys. They measure exposures and outcomes simultaneously at a single point in time, providing prevalence data. Uses include: describing and comparing distribution of exposures and outcomes; generating hypotheses (e.g., via prevalence ratio); and planning health services. Strengths: cheap, quick, can assess multiple exposures and outcomes simultaneously.
Limitations of cross-sectional studies: because exposure and outcome are measured at the same time, temporal sequence cannot be established — we cannot determine whether exposure preceded outcome. They do not measure incidence. They are unsuitable for rare, transient (may have resolved by the time of measurement), or variable (results differ depending on timing) exposures or outcomes.
The GATE frame for a cross-sectional study shows a vertical time arrow (simultaneous measurement).
Ecological studies (correlational studies) compare exposures and outcomes across populations rather than individuals. Each data point represents a population. Uses: compare population-level factors (e.g., air pollution, water fluoridation); consider hypotheses (assess consistency with ecological data). Relatively inexpensive when routinely collected data exist.
The ecological fallacy is the key limitation: applying group-level associations to individuals is invalid. A higher average exposure rate in a population does not mean that the exposed individuals within it are the ones who develop disease. Confounding cannot be controlled (no individual data, no random allocation), and causation cannot be established.
Distinction: cross-sectional studies generate hypotheses by finding associations in individual-level data; ecological studies consider hypotheses by checking whether aggregate population-level data is consistent with the hypothesis.