Summary
Highlights
Analytic epidemiology shifts focus from 'who, where, and when' (descriptive epidemiology) to 'why and how' health outcomes occur, identifying risk factors and causal relationships. Unlike descriptive studies, analytic studies include comparison groups and form the basis for evidence-based recommendations.
Analytic studies are categorized into observational studies, where researchers observe relationships without intervention, and experimental studies, where interventions are introduced. This chapter primarily focuses on observational studies, including case-control, cohort, case-crossover, and nested designs.
In case-control studies, participants are selected based on disease status (cases have it, controls don't), and researchers look back in time for exposure history. This design is useful for rare diseases, long latency periods, and when studies need to be fast and cost-effective. The primary measure of association is the odds ratio.
The odds ratio indicates the likelihood of cases being exposed compared to controls. An odds ratio greater than one suggests a positive association, one indicates no association, and less than one suggests a protective effect. A 2x2 table aids in its calculation.
Cohort studies select participants based on exposure status (exposed vs. unexposed) and follow them over time to observe disease development. They can be prospective (present to future) or retrospective (using historical data). These studies are ideal for common exposures, multiple outcomes, and estimating incidence and relative risk.
Cohort studies allow for the calculation of the risk ratio (relative risk), which compares disease probability in exposed versus unexposed groups, and the rate ratio, which accounts for varying follow-up periods using person-time. A risk ratio greater than one indicates increased risk, and less than one indicates decreased risk.
Case-crossover studies use each participant as their own control, comparing exposure before disease onset to an earlier time, useful for acute exposures. Nested case-control studies are embedded within a cohort, selecting controls from those at risk when a case arises, offering cost-effectiveness and maintaining temporal structure.
All studies are susceptible to bias and confounding. Common biases in case-control studies include selection bias, recall bias, and interviewer bias. Cohort studies may face loss to follow-up and the healthy worker effect. Confounding occurs when a third factor is associated with both exposure and outcome. Strategies to address confounding include restriction, matching, stratification, and multivariable analysis.
Analytic epidemiology aims to answer 'why' and 'how' diseases occur, primarily utilizing case-control and cohort study designs. Key measures of association include the odds ratio, risk ratio, and rate ratio. Specialized designs offer flexibility. Mitigating bias and controlling for confounding are crucial for credible findings.