Epidemiology

Part of Public Health

The study of disease distribution and determinants in populations — the scientific foundation for understanding why diseases spread, who is at risk, and what interventions work.

Why This Matters

Epidemiology is how medicine becomes science rather than superstition. Without it, a healer who sees three recoveries after giving willow bark tea might conclude the tea works — but they cannot distinguish this from spontaneous recovery, placebo effect, or the three patients happening to be unusually healthy. With even basic epidemiological thinking, the healer compares outcomes in patients who received the tea versus those who did not, counts the difference, and makes a judgment based on rates rather than impressions.

In public health terms, epidemiology is how you know whether your sanitation improvements are actually reducing diarrhea rates, whether your malaria prevention campaign is working, and whether one community’s higher death rate compared to another is due to disease, malnutrition, or some other factor. Without this analytical framework, public health operates on opinion and tradition.

The mathematical tools of formal epidemiology — confidence intervals, multivariate analysis — require resources you may not have. But the conceptual framework is accessible to any literate person and profoundly improves decision-making. This article provides that framework.

Key Epidemiological Measures

Incidence: The number of NEW cases of a disease occurring in a population in a defined time period. Expressed as rate: cases per 1,000 (or 100,000) people per year.

• Example: 12 new cholera cases in a community of 800 in one month = 12/800 per month = 15 per 1,000 per month = 180 per 1,000 per year

Prevalence: The total number of EXISTING cases (new + old) in a population at a specific moment.

• More relevant for chronic diseases (tuberculosis, malnutrition)
• Less relevant for acute diseases (cholera — patients either recover quickly or die)

Case fatality rate (CFR): The proportion of people with a specific disease who die from it.

• CFR = deaths from disease / cases of disease × 100%
• Example: 3 deaths among 12 cholera cases = 25% CFR
• High CFR means the disease is dangerous; low CFR means most recover
• CFR changes with treatment quality — comparing CFRs before and after an intervention measures the intervention’s impact

Attack rate: In an outbreak, the proportion of exposed people who develop the disease.

• Attack rate = cases / exposed persons × 100%
• Used in outbreak investigation to identify the source
• Example: 20 people attended a feast; 12 developed vomiting within 6 hours; attack rate = 60%

Basic reproduction number (R₀, “R naught”): The average number of new cases one case generates in a fully susceptible population. R₀ > 1 means the disease spreads. R₀ < 1 means it dies out.

• You cannot calculate R₀ precisely without laboratory tools, but you can estimate whether a disease is spreading (case counts growing) or contracting (case counts shrinking).

The Epidemiological Triad

Disease occurs at the intersection of three factors:

Agent: The pathogen or cause — a bacterium, virus, parasite, nutritional deficiency, toxin. Understanding the agent tells you about transmission routes, treatment options, and incubation period.

Host: The individual at risk — their age, sex, immune status, nutritional status, genetics. Understanding host factors tells you who is most vulnerable and why. Children with malnutrition have higher attack rates and CFRs for most infections than well-nourished adults.

Environment: The physical and social conditions that facilitate or inhibit exposure — water quality, housing density, seasonal factors, trade routes. Understanding environmental factors tells you where to focus prevention.

Most epidemiological problems can be analyzed through this triad: What is the agent? Who are the hosts at greatest risk and why? What environmental conditions are enabling transmission?

Descriptive Epidemiology: Time, Place, Person

When investigating an outbreak, systematically describe cases in three dimensions:

Time — When did cases start? Plot a histogram of case onset by day. The shape of this “epidemic curve” suggests the transmission type:

Curve Shape	Interpretation
Abrupt peak, rapid decline	Point source (all exposed to same source at one time) — contaminated water or food
Gradual rise, plateau, gradual decline	Propagated spread (person-to-person transmission) — infectious disease spreading through community
Multiple peaks	Multiple sources or waves of transmission

Place — Where do cases cluster? Map cases on a sketch of the community. Cases clustered around a water source suggest waterborne transmission. Cases in one household cluster suggest person-to-person within household. Cases scattered randomly suggest airborne or widespread environmental exposure.

This mapping approach was used by John Snow in 1854 to identify the Broad Street water pump as the source of a London cholera outbreak — one of the founding events of epidemiology — using only a dot map and interviews.

Person — Who is affected? Compare age distribution, sex, occupation, behaviors, and locations of cases to the general community. If cases are disproportionately among children who attend the same school, the school is implicated. If cases are among adults who all drink from the same well, the well is implicated.

Analytical Epidemiology: Measuring Association

Once you have a hypothesis about what is causing an outbreak, test it by measuring whether the suspected exposure is actually associated with disease.

Relative risk (RR) (for cohort studies — where you follow a group forward): RR = (Risk in exposed) / (Risk in unexposed)

• RR = 1: no association
• RR > 1: exposure increases risk
• RR < 1: exposure decreases risk (protective)

Example: 100 people drank from Well A; 80 developed diarrhea → risk = 80% 100 people drank from Well B; 10 developed diarrhea → risk = 10% RR = 80% / 10% = 8 — drinking from Well A is 8 times more likely to cause diarrhea

Odds ratio (OR) (for case-control studies — when looking backward at what cases were exposed to): OR = (Odds of exposure among cases) / (Odds of exposure among controls)

• Interpretation is similar to RR for rare diseases

Attributable fraction: What proportion of cases would be eliminated if the exposure were removed? AF = (Exposed cases - Unexposed cases) / Total cases This tells you how much disease burden you can prevent by eliminating a specific risk factor.

Confounding and Bias

Not every association indicates causation. Two factors can be associated statistically without one causing the other.

Confounding: A third variable is associated with both the exposure and the outcome. Example: areas with higher rates of shoe-wearing have lower rates of hookworm infection. This does not mean shoes are the key factor — both shoe-wearing and hookworm protection correlate with wealth, and wealth correlates with many health-protective behaviors.

Selection bias: Cases and controls are not comparable because of how they were selected.

Information bias: Cases and controls are asked about exposures differently, or recall their exposures differently.

For practical public health purposes, the key question is: does the association make biological sense? Is the magnitude large (an RR of 8 is hard to explain away by confounding; an RR of 1.2 might be)? Is there a dose-response relationship (more exposure → more disease)? Does removing the exposure reduce disease rates?

Applying Epidemiology to Community Decision-Making

The most practical application of epidemiological thinking is before-and-after comparison:

• Measure disease rates before a public health intervention
• Implement the intervention
• Measure disease rates after
• Compare the change

This simple pre-post comparison, while not a rigorous randomized trial, provides meaningful evidence for practical decision-making. If diarrhea rates fall by 60% after latrine construction, the most parsimonious explanation is that the latrines work.

Use this framework to justify resource allocation decisions, evaluate programs honestly, and build a community evidence base that improves public health over years and generations.