HTH 450 Chapter 12: Estimating Risk - Is There an Association?

Vocabulary practice flashcards covering measures of risk, association, study designs, relative risk interpretations, odds ratios, and matched-pairs case-control analysis from Chapter 12.

Absolute Risk

The incidence of a disease in a population. It does not take into account the risk of disease in unexposed individuals and does not indicate whether an exposure is associated with an increased risk of disease.

Randomized Clinical Trial (Design)

A study design where a defined study population is randomly assigned to receive either a new treatment or a current treatment, and participants are followed to determine whether they improve or do not improve.

Cohort Study (Design)

An observational study design where a defined population is not randomly assigned, but classified into exposed and not exposed groups, and followed forward in time to observe whether disease develops.

Case-Control Study (Design)

An observational study design that begins by selecting individuals who have the disease ('cases') and individuals who do not have the disease ('controls'), and subsequently looks back to determine past exposure status.

Relative Risk (RR)

The probability of an event (developing a disease) occurring in exposed people compared with the probability of the event occurring in unexposed people, calculated as $\text{Relative risk} = \frac{\text{Risk in exposed}}{\text{Risk in unexposed}}$.

Relative Risk Interpretation ($\text{RR} = 1$)

Indicates that the risk of disease in the exposed group is equal to the risk in the unexposed group, signifying no association.

Relative Risk Interpretation ($\text{RR} > 1$)

Indicates that the risk of disease in the exposed group is greater than the risk in the unexposed group, signifying a positive association that is possibly causal.

Relative Risk Interpretation ($\text{RR} < 1$)

Indicates that the risk of disease in the exposed group is less than the risk in the unexposed group, signifying a negative association that is possibly protective.

High Cholesterol Level (Doyle et al., 1964)

Serum cholesterol level defined as $220\,\text{mg/dL}$ or greater.

Odds

The ratio of the probability that an event will occur ($P$) to the probability that the event will not occur ($1 - P$), defined as $\text{Odds} = \frac{P}{1 - P}$.

Odds Ratio in a Cohort Study

The ratio of the odds that an exposed individual develops disease to the odds that an unexposed individual develops disease, calculated from a $2 \times 2$ table as $\frac{a / b}{c / d} = \frac{ad}{bc}$.

Odds Ratio in a Case-Control Study

The ratio of the odds that a case was exposed to the odds that a control was exposed, calculated from a $2 \times 2$ table as $\frac{a / c}{b / d} = \frac{ad}{bc}$.

Cross-Products Ratio

The ratio obtained by multiplying diagonal cells in a $2 \times 2$ table ($a \times d$ divided by $b \times c$), which yields the odds ratio in both cohort and case-control studies: $\text{OR} = \frac{ad}{bc}$.

Rare Disease Assumption

The principle stating that the odds ratio is a good estimate of the relative risk when the disease is infrequent (rare), but is not a good estimate when the disease is common.

Individual Matching

A technique in case-control studies where controls are selected by pairing each one to a case based on variables related to disease risk, such as sex, age, or race, and results are analyzed as pairs rather than individuals.

Concordant Pairs

Case-control pairs in a matched study in which both the case and the matched control had the same exposure experience (both exposed or both unexposed).

Discordant Pairs

Case-control pairs in a matched study in which the case and the matched control had different exposure experiences (one exposed and the other not exposed).

Matched-Pairs Odds Ratio

The measure of association in a matched case-control study calculated exclusively from discordant pairs using the formula $\text{Matched pairs odds ratio} = \frac{b}{c}$.