Measures of Association
~2 min read
Lesson 8 of 20
Notes
Analytic epidemiology quantifies the association between exposure and outcome using three principal measures: relative risk (RR), risk difference (RD), and odds ratio (OR).
Relative risk (RR) is the ratio of the incidence in the exposed group to that in the comparison group: RR = I_exposed / I_comparison. The null value is 1 (no difference in incidence between groups). RR > 1 suggests the exposure is a risk factor; RR < 1 suggests a protective effect. RR conveys the strength of the association and provides aetiological insight. It can be calculated from cohort studies and RCTs.
Risk difference (RD), or attributable risk, is RD = I_exposed − I_comparison. The null value is 0. RD > 0 means the exposed group has more cases; RD < 0 means fewer. RD is a measure of the public health impact of an exposure — how many cases in the exposed population are attributable to the exposure, and how many could be prevented by removing it. RD must always be reported with the time period.
Odds ratio (OR) is used in case-control studies, where incidence cannot be calculated. OR = (a/b)/(c/d) from a 2×2 table. The null value is 1. The OR approximates RR when disease prevalence is low (rare disease assumption). When disease is common, OR overestimates RR.
PECOT identifies key study elements: Population, Exposure/comparison, Comparison, Outcome, Time. The GATE frame visualises these elements graphically.
Social determinants of health are the conditions in which people are born, grow, work, live, and age, and the wider forces and systems shaping daily life — including economic policies, social norms, and political systems. The social gradient of health describes the correlation between socioeconomic position and health: those in lower positions have worse health outcomes, measurable by income, occupation, or education.
Selection bias in cross-sectional studies occurs when there is a systematic difference between the sample and source population — that is, when the sample is not representative of the source population due to the selection process. Absolute poverty means income insufficient for minimum nutritionally adequate diet plus essential non-food needs. Relative poverty means income insufficient to purchase necessities relative to the standards of the society.