The roles of epidemiology in public health, and the steps to solving a health problem
The official definition
Epidemiology is the study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to control health problems.
DistributionWho gets sick, where and when (person, place, time)
DeterminantsWhy and how: the causes and risk factors
ApplicationUsing what you learn to stop the problem
The word itself: epi (upon) + demos (people) + logos (study). Epidemiology helps find what about the agent, host and environment affects health, which causes matter most, which groups are at highest risk, and whether health programs work.
Public health recognizes six core jobs that make up the work of epidemiology (five were named in the mid-1980s; policy development was added later). A single epidemiologist might do all six over the course of a career, sometimes all six in the same investigation.
01
Surveillance
Continuously watching disease patterns so problems are caught early. (Section 6!)
02
Field Investigation
Going out (literally, "shoe leather") to find cases and clues during an outbreak.
03
Analytic Studies
Running rigorous studies (like case-control or cohort studies) to test a hypothesis.
04
Evaluation
Checking whether a program or intervention is actually working as intended.
05
Linkages
Teaming up with labs, doctors, and other agencies: field epi is a team sport.
06
Policy Development
Turning evidence into recommendations, regulations, and guidelines.
No matter which of these six roles they're playing, every epidemiologist follows the same basic logic when solving a health problem, sometimes called the epidemiologic approach:
#
Count
Find and count the cases or health events, describing them by time, place, and person.
÷
Divide
Divide the number of cases by the size of the population at risk to calculate a rate.
⇄
Compare
Compare that rate over time, or against another group, to see if it's really unusual.
Step zero: define a "case." Before anyone can count anything, epidemiologists agree on a case definition, a set of standard criteria (symptoms, sometimes a lab test, and limits on person, place and time) that decide whether a person counts as a case. A shared case definition means a health department in 2024 can compare its numbers fairly against 2019, or against another state entirely.
Two Kinds of Epidemiology
🔍 Descriptive
Asks
Who? What? Where? When?
Does
Describes the problem by person, place and time
Result
Generates a hypothesis (a smart guess)
🧪 Analytic
Asks
How? Why?
Does
Compares groups, like people who ate the salad and people who didn't
Result
Tests the hypothesis, and checks whether an intervention works
Watch or assign? In an observational study, investigators just watch what happens; it can be descriptive or analytic. In an experimental study, investigators decide who gets what, like a vaccine trial. More in Chasing the Curve, chapter 1.
The Public Health Approach: From Problem to Response
1
Surveillance
What's the problem? Count and track cases.
2
Risk factor identification
What's the cause? Find who's at risk, and why.
3
Intervention evaluation
What works? Test ways to prevent it.
4
Implementation
How do you do it? Put what works into action.
Another version tests use: collect data, assess it (draw conclusions), test the hypothesis, then take action. Same idea, different words.
✓ Check Yourself
Q1Put the three basic actions an epidemiologist takes with data in order.
Drag to reorder (or use the arrows), then press Check.
Count the cases (by time, place and person)
Divide by the population at risk to get a rate
Compare rates across times, places or groups
Count, divide, compare. A rate lets you compare groups of different sizes fairly.
Q2Epidemiology is the study of the distribution and ________ of health events in populations.
Determinants: the causes and risk factors. Distribution is who, where and when.
Q3Descriptive or analytic? Sort each one.
Mapping where every case lives
Counting cases by age group
Comparing people who ate the salad with people who didn't
Testing whether a new vaccine prevents flu
Descriptive work describes who, where and when, and gives you a hypothesis. Analytic work compares groups to test it.
Q4Put the public health approach in order.
Drag to reorder (or use the arrows), then press Check.
Surveillance
Risk factor identification
Intervention evaluation
Implementation
What's the problem? What's the cause? What works? How do you do it?