EPIB301: Study Guide Definitions

Population

Epidemiology focuses on disease occurrence among groups of people rather than individuals

Prevalence

The number of existing cases of a disease or health condition in a population at a designated time.

Incidence

The number of new health-related events (instances of people falling ill) in a defined population during a specified period.

Intervention study

An investigation involving an intentional change in a subject's status

Therapeutic trial

A study involving curative drugs or new surgical procedures intended to improve a patient's health.

Prophylactic trial

A trial designed to evaluate the effectiveness of a substance

Crossover

A clinical trial design involving a switch of study treatments for a patient

Randomization

A process using chance to assign subjects to either an intervention or control group

Manipulation

The control of the study factor (exposure) by the investigator rather than the subjects.

Blinding/Masking

A procedure where the subject (single-blind) or both the subject and investigator (double-blind) are unaware of which treatment is being received to reduce bias.

Ecologic comparison study

Assesses the correlation between exposure and disease rates among different groups over the same time period.

Ecologic trend study

Assesses the correlation between changes in exposure and changes in disease within the same group over different times.

Ecological fallacy

The error of making inferences about individuals based only on group-level data.

Correlation

A recognized pattern in data that shows the direction and strength of the relationship between variables.

Case definition

A conceptual and operational definition used to accurately classify and identify individuals who have a disease.

Odds

The ratio of the probability of an event occurring to the probability of it not occurring.

Recall bias

A type of systematic error where there is differential accuracy in remembering past events or exposures between study groups.

Temporality

The requirement that the cause must precede the effect for a causal association to exist.

Cohort

A population group distinguished by a common characteristic (such as birth year or exposure status) followed over time.

Attrition

The loss of subjects during a study due to factors like dropouts

Placebo

A nonactive substance that resembles the actual drug or treatment being tested.

Placebo effect

A beneficial health outcome resulting from a person's belief or anticipation that an intervention will help.

Clinical endpoints

The final outcomes measured in a trial

Efficacy

The extent to which an intervention produces a beneficial result under ideal conditions.

Number needed to treat

The expected number of people who must receive a treatment to prevent one unfavorable outcome.

Equipoise

The principle that a trial should only be conducted when there is genuine uncertainty about which treatment is better.

Distributive justice

The ethical requirement that study populations must be appropriate and non-exploitative.

Beneficence

The ethical obligation to maximize benefits while balancing them against risks to do good for the participant.

Informed Consent

A process where researchers explain risks and benefits to subjects so they can voluntarily decide to participate.

Respect for persons

The principle that participants must volunteer without undue influence and fully understand their role in the research.

Prevalence formula (disease in sample)

(Total with disease) / (Total study sample) = (A + C) / (A + B + C + D)

Prevalence formula (exposure in sample)

(Total exposed) / (Total study sample) = (A + B) / (A + B + C + D)

Incidence in exposed group

(New cases among exposed) / (Total exposed at risk) = A / (A + B)

Incidence in unexposed group

(New cases among unexposed) / (Total unexposed at risk) = C / (C + D)

General incidence rate

(Number of new cases) / (Total population at risk during the same time period)

Odds of exposure among cases

(Exposed cases) / (Unexposed cases) = A / C

Odds of exposure among controls

(Exposed controls) / (Unexposed controls) = B / D

Odds Ratio (OR) formula

(Odds of exposure in cases) / (Odds of exposure in controls) = (A/C) / (B/D) or cross-product (AD)/(BC)

Relative Risk (RR) formula

[A / (A + B)] / [C / (C + D)]

What makes a study design different than anecdotal evidence

Study design is formal

Observational vs. experimental research

Observational: no control over exposure; Experimental: researcher manipulates exposure and uses randomization.

More than one level of exposure

Exposure variable is continuous (e.g.

Descriptive vs. analytical studies

Descriptive: who

Typical uses of cross-sectional studies

Intervention planning

How case-control groups are formed

Based on disease status: cases have disease

Possible sources of cases and controls

Cases: disease registries

Why OR is normally not a measure of risk

Case-control studies are retrospective and select based on outcome

Population vs. hospital controls

Population: random from geographic area (representative but expensive); Hospital: from same facility (easier but may not represent general population).

Population-based vs. exposure-based cohort

Population-based: entire population or sample

Prospective

retrospective

Interpretation of correlation value

r from -1 to +1: direction (+ = both increase

What an odds ratio compares

Compares odds of exposure among cases to odds of exposure among controls.

Interpretation of OR

OR=1: no association; OR>1: exposure is risk factor; OR<1: exposure is protective factor.

Why cohort studies establish temporality

Start with disease-free participants

What relative risk compares

Compares risk (incidence) of disease in exposed group to risk in unexposed group.

Interpretation of RR

RR=1: no association; RR>1: exposure is risk factor; RR<1: exposure is protective factor.

Why RR or OR = 1 means no association

Numerator and denominator are equal

Why experimental designs are gold standard

Investigator controls exposure (dose/timing) and uses randomization to minimize confounding

What randomization accomplishes

Makes groups comparable by balancing known and unknown confounders

Clinical endpoints in experimental study

Final outcomes measured: rates of disease

Goal of therapeutic intervention

Evaluate drugs or procedures to cure or improve health of patients already suffering from a condition.

Goal of prophylactic intervention

Evaluate effectiveness of substance or program (e.g.

Ecologic study definition

Examines group/population as unit of analysis

Types of ecologic studies

Comparison: correlation among different groups same time; Trend: changes in exposure and disease within same group over time.

Ecologic study measure

Correlation (Pearson's r).

Ecologic study advantages

Quick

Ecologic study disadvantages

Ecological fallacy (incorrect individual inferences from group data); imprecise exposure measurement.

Cross-sectional study definition

Observational study measuring outcome and exposure simultaneously at a single point in time (snapshot).

Cross-sectional study unit of analysis

Individual.

Cross-sectional study measure

Prevalence (often called prevalence studies).

Cross-sectional study advantages

Relatively quick and inexpensive; useful for health planning and examining risk factor trends.

Cross-sectional study disadvantages

Cannot determine temporality; inefficient for rare or short-duration diseases.

Case-control study definition

Analytical study comparing diseased (cases) vs. non-diseased (controls) to identify differing exposure levels.

Case-control study directionality

Retrospective (looks backward from effect to cause).

Case-control study measure

Odds Ratio (OR).

Case-control study advantages

Best for rare diseases; efficient for long latent periods; allows evaluation of multiple risk factors.

Case-control study disadvantages

Vulnerable to recall bias; inefficient for rare exposures; difficult to establish clear temporal relationship.

Cohort study definition

Follows group with common characteristic over time to compare incidence of outcomes.

Cohort study measure

Relative Risk (RR).

Cohort study advantages

Establishes temporality; permits direct determination of risk; can examine multiple outcomes.

Cohort study disadvantages

Lengthy and expensive; susceptible to attrition; inefficient for rare diseases.

Experimental (RCT) definition

Investigator manipulates exposure and uses randomization to assign subjects to groups.

Experimental (RCT) purpose

Evaluate efficacy of therapeutic (curative) or prophylactic (preventive) measures.

Experimental (RCT) measure

Relative Risk (RR) or other clinical endpoints.

Experimental (RCT) advantages

Gold standard for causal inference; greatest control over exposure; minimizes confounding via randomization.

Experimental (RCT) disadvantages

Highly expensive and resource-intensive; limited external validity; ethical constraints prevent testing harmful exposures.