Final Epib review

Cohort study

Observational research method that follows a group of people over time to track how specific exposures affect outcome Exposure → Disease

what to use for cohort study (measurement)

use relative risk (risk ratio/incidence ratio)

Cohort study example

Imagine you want to study if eating spicy food causes stomach aches.

Step 1: Find a group of people who don't have stomach aches right now.

Step 2: Sort them into "Spicy Eaters" (Exposed) and "Bland Eaters" (Unexposed).

Step 3: Follow them to see who develops a stomach ache.

Because you know they were healthy at the start, you can prove the exposure came before the disease

Retrospective study

Observational study that uses past record to identify exposures that happened years ago and compare it to now Exposures → Disease

What to use for retrospective study

use relative risk

Retrospective study example

A researcher gets the 1985–1995 personnel files from a paint factory. They group workers into those who worked with benzene (Exposed) and those who worked in shipping (Unexposed). They then check 2025 cancer registry data to see which group had more leukemia

Prospective study

Start now and follow the group into the future (usually in increments) Exposure → Disease

what to use for prospective study

use relative risk

Prospective study example

You find people who already choose to drink 5 energy drinks a day and people who already choose not to. You just follow them to see what happens. The subjects choose their own exposure

Interpret relative risk(risk ratio/incidence ratio)

People exposed had _____ times the risk of _____ compared to the unexposed group

relative risk (risk ratio/incidence ratio)

how much more (or less) likely it is for a disease to occur in an exposed group compared to an unexposed group

Prospective cohort Pros

• Temporality: It is the best for proving the exposure came before the disease because you start with healthy people

• You can study one exposure (like smoking) and look for many different diseases it might cause (lung cancer, heart disease, stroke) all at once

Prospective cohort cons

• Expensive and slow: pay researchers years to follow people

• Loss to follow up: People may move, quit, or just disappear which can ruin the data

Retrospective cohort pros

• Cheaper and faster: you can just look at records to get data

• Ethical: Good for studying disaster of chemical leaks that already happened

Retrospective cohort cons

• Recall bias: people might not remember accurately

• Confounding: harder to control other factors that weren’t recorded

Case Control study

Observational study that is retrospective that compares 2 groups: Patients with a specific condition (case)

Patients without it (Control)

Disease → exposure

what to use for case control study

use odds ratio

Case control study example

Find people who have lung cancer and we look backward to find the exposure

Case control study pros

• Efficient for rare diseases: Find sick people first

• Quick and inexpensive: Since outcome has already happened, you don’t need to follow people for years

Case control study cons

• Recall bias: sick people often think more deeply about their past and “remember” exposures that didn’t actually happen

• Temporality: may be hard to prove that the exposure definitely happened before the disease started

Odds ratio

How much higher the chance of being exposed was for the sick people compared to the healthy people

Odds ratio interpretation

Those with the disease had ___ times the odds of having been exposed to ___ compared to the healthy group

Cross sectional study

“snap shot” study; looks at a single point in time to see how has the exposure and who has the disease right now

Cross sectional study pros

• Good for prevalence: best way to find out how common a disease is in a population at any given time

• Fast and cheap

Cross sectional study cons

• Temporality: can’t 100% establish a cause and if the result was a result of the exposure

• Only a snapshot: doesn’t account for people who died or got cured

What to use for cross-sectional study

use prevalence ratio

Prevalence ratio

Frequency of existing cases of a disease or other condition in a population

Prevalence interpretation

The prevalence of disease was ____ times higher in exposed group compared to unexposed at the time of the study

Randomized control trial (RTC)

Each study participant has the same probability of being assigned to the treatment group

Randomization

Allocation of a participant to the treatment or control group

Random sampling

method od choosing participants to be in the study

Placebo effect

Participants improve because they are told they will or think they will; stems from the power of suggestion

RCT pros

• Eliminates bias- by randomly assigning people to groups, it balances out confounders

• Blinding: you can use placebos where neither the patient nor the doctor knows to prevent expectations from ruining data

RCT cons

• Expensive and slow- recruit large amounts of people

• Ethical constraints: cannot randomize harm

Matching

Selecting pairs of clusters of individuals who are comparable on important variables

Quasi experimental study

Like an experiment but lacks random assignment to treatment and control groups; investigator manipulates the study factor (exposure) but does not assign individuals to treatment groups

Quasi experimental study Pros

• Ethics: allows to study “natural experiments”

• Real- world setting: Groups already existing which reflects real life more

Quasi experimental study cons

• Confounding variables: Because there was no randomization, there could be confounding variables

• selection bias: groups may be different in ways you didn’t measure

Confounding variable

third variable that affects the relationship between the exposure and health outcome

information bias

hawthorne effect/surveillance bias, recall bias, interviewer bias, measurement error, reporting bias

Temporality

Exposure must be before the disease/outcome; only criterium that MUST be met

Strength of association

Measures the magnitude of the relationship between an exposure and an outcome, indicating how likely a casual link exists

Specificity of Association

Exposure consistently leads to a single outcome (little to no alternative explanations for the relationship)

Biological gradient/ dose response relationship

As exposure to the risk factor increases, risk if disease/outcome either increases or decreases

Consistency

Relationships demonstrated in multiple studies across different populations, study designs, and circumstances are more likely to be casual

Plausibility

A mechanism exists that connects the exposure to the outcome

Coherence

Proposed association must not contradict current scientific knowledge

Analogy

Proposed association mirrors another association already established in the literature

Necessary cause

A factor/event that must occur for the specific disease to occur

Sufficient cause

A set of factors that, when come together, inevitably produces a specific health outcome

Causal inference

process of determining whether an association between an exposure and an outcome is truly caused by that exposure rather than bias or confounding

Internal validity

results represent the true relationship between exposure and outcome for the participants studied, free from systematic errors

External validity

generalizability: the extent to which study results apply to populations, settings, or times beyond the study sample

Under Observation umbrella

Cohort, Case-control, Cross sectional

Under experiment umbrella

Randomized control trial (RCT), Quasi experiment

Agent

Entity (pathogen) that causes the disease; necessary but not sufficient for infectious diseases to develop

Host

Organism infected with the agent (usually humans or animals); susceptibility is determined by several factors like age and immune system

Environment

can serve as a reservoir for infectious agents (can be physical, social, cultural, etc)

Attack rate

Frequenctly used to describe incidence of food borne illness, infectious disease, and other epidemics

Direct transmission

Immediate transfer of infectious agents between the previous host and the new host

Direct transmission examples

• Skin-skin contact

• exposure to bodily fluids

• HIV

Indirect transmission

Transfer with something bridging the gap between the previous host and the new host

Indirect transmission examples

• Vector borne infections (agent is carried by living things)

• vehicle borne infections (agent is carried by inanimate things)

• Fomite borne infections (agent is carried by contaminated objects or surfaces)

Chain of infection

Reservoir → portal of exit → Mode of transmission → Agent → portal of entry → Host

Infectivity

Proportion of exposed individuals who become infected (regardless of whether they become ill)

Pathogenicity

Proportion of exposed individuals who become ill (clinically apparent cases)

Virulence

Measure of disease severity (proportion of cases that resulted in death)

Primary prevention

Action taken to prevent onset of disease/injury

Primary prevention example

• Immunizations

• Wearing masks

secondary prevention

Actions taken to identify and treat a disease/condition in its early stages before it becomes more serious

secondary prevention example

• Breast cancer screening programs

• testing

Tertiary prevention

Managing an established and long-term disease to prevent it from getting worse

Tertiary prevention examples

• Physical therapy for stroke patients

• Meds for chronic illnesses

Generation time

Average time between the point when an individual becomes infected and the point when that person infects another individual; infection to infection

Incubation period

The time between exposure to an infectious agent and the first signs and symptoms of disease

virulence

Severity of harmfulness of a disease; capacity to cause death

Risk assessment

To estimate the health risks of specific exposures, critical for policy decisions

Hazard identification

Examines the evidence that associates exposure to an agent with its toxicity

Dose-response assessment

Measures the relationship between the amount of exposure and the occurrence of unwanted health effects

exposure assessment

identifies exposed populations and estimates the exposure

Risk characterization

Estimates the number of excess unwarranted health events expected

Risk management

Implements policies to reduce exposure to risks

Social and behavior epidemiology

the study of how social, behavioral, and psychosocial factors influence the distribution of health outcomes across populations

Social epidemiology

Focuses on the social determinants and distribution of health

Behavioral epidemiology

Investigates how individual behaviors influence health outcomes

Psychosocial epidemiology

Examines how social factors and personal behavior interact to affect health

Social determinants of health

Non-medical factors that influence health outcomes

Social determinants of health examples

• income, education, employment

• Healthcare access

Stress

Chronic stress can lead to health issues like mental health, diabetes, and cardiovascular diseases

Personality factors

How behavior patterns like hostility can influence heart disease

Psychosocial epidemiology

the impact of psychosocial factors on health outcomes

Ecosocial theory

Focuses on how the interaction of social environments and biological insults over the life course affects health outcomes

Moderators

Factors that influence the relationship between exposures and outcomes

Equipoise

state of genuine uncertainty regarding which of two or more treatments is more effective

Respect of person’s autonomy

Participants have the right to decide what happens to them. It also includes protecting those with diminished autonomy

Beneficence

The obligation to "do no harm" and to maximize possible benefits while minimizing possible harms

Justice

Fair distribution of burdens and benefit; participants should not be in research studies if they’re unlikely to benefit

informed consent

A participant is given all necessary information about a study and voluntarily agrees to participate

Institutional review board (IRB)

Independent committee that reviews research plans to ensure they are ethical before the study can begin