2.1 Prevalence and incidence

Prevalence and Incidence Fundamentals

Prevalence and incidence are the two core ways epidemiologists measure how much disease exists in a population. Prevalence tells you the total burden of disease at a given moment, while incidence tells you how quickly new cases are appearing. Together, they shape nearly every public health decision, from allocating hospital beds to evaluating whether a vaccination campaign is working.

Prevalence vs. Incidence Definitions

Prevalence is the proportion of a population that has a specific condition at a given point in time. It counts all existing cases, whether someone got sick yesterday or ten years ago. You'll usually see it expressed as a percentage or a ratio (e.g., "5 per 100").

Incidence is the rate at which new cases develop in a population over a defined time period. It only counts people who were disease-free at the start of that period and then developed the condition. It's typically expressed as a rate per person-time (e.g., "2 per 1,000 person-years").

Three key differences to keep straight:

• Timeframe: Prevalence is a snapshot (one point in time). Incidence covers a span of time.
• Which cases count: Prevalence includes all current cases (new + old). Incidence includes only new cases.
• What it tells you: Prevalence reflects disease burden. Incidence reflects disease risk.

Calculation of Prevalence and Incidence

Prevalence formula:

$Prevalence = \frac{\text{Number of existing cases}}{\text{Total population}} \times 100$

Example: If 50 people in a community of 1,000 currently have hypertension, the prevalence is:

$\frac{50}{1{,}000} \times 100 = 5\%$

Incidence rate formula:

$\text{Incidence Rate} = \frac{\text{Number of new cases}}{\text{Person-time at risk}}$

The denominator here is person-time at risk, not just the total population. Person-time accounts for the fact that different people may be observed for different lengths of time. If 10,000 people are each followed for 1 year, that's 10,000 person-years. If 20 new cases appear, the incidence rate is:

$\frac{20}{10{,}000 \text{ person-years}} = 0.002 \text{ per person-year} = 2 \text{ per 1,000 person-years}$

Incidence rates are often scaled to per 1,000 or per 100,000 person-years to make the numbers easier to read and compare.

Interpretation of Health Rates

What prevalence tells you: Prevalence reflects how much of a burden a disease places on a healthcare system right now. Chronic diseases like diabetes tend to have high prevalence because people live with them for years. A high prevalence signals that a population needs ongoing treatment resources, clinic capacity, and disease management programs.

What incidence tells you: Incidence reflects the current risk of developing a disease. A rising incidence rate is an early warning sign, especially for infectious diseases. For example, a spike in influenza incidence during winter signals that transmission is accelerating and public health response may be needed.

Public health applications:

• Comparing disease frequency across populations (e.g., diabetes prevalence in urban vs. rural areas)
• Evaluating prevention strategies (e.g., did a vaccination program reduce the incidence of measles?)
• Identifying high-risk groups for targeted interventions (e.g., lung cancer screening for heavy smokers)

Factors Affecting Prevalence and Incidence

Prevalence and incidence don't move independently. Understanding what drives each one helps you avoid misinterpreting the data.

Factors that raise or lower prevalence:

• Disease duration: Longer-lasting conditions accumulate more existing cases. Diabetes has high prevalence partly because people live with it for decades.
• Survival rates: Better treatments for cancer keep patients alive longer, which increases prevalence even if incidence stays flat.
• Migration: People moving into or out of a region change the local prevalence.
• Treatment effectiveness: Effective HIV antiretroviral therapy extends life, raising prevalence. A cure, by contrast, would lower it.

Factors that raise or lower incidence:

• Exposure to risk factors: Occupational hazards (e.g., asbestos exposure) directly increase incidence of related diseases.
• Changes in diagnostic criteria: When the definition of autism spectrum disorder broadened, reported incidence rose even though the underlying biology may not have changed.
• Screening programs: Mammography detects breast cancer cases earlier, which can temporarily inflate incidence figures.
• Population susceptibility: Genetic predisposition or lack of prior immunity (e.g., a novel virus) affects how many new cases arise.

Factors that affect both:

• Age and sex distribution of the population (cardiovascular disease rates differ sharply by age and sex)
• Socioeconomic conditions and access to healthcare
• Environmental exposures like air pollution
• Genetic predisposition within a population

The Prevalence-Incidence Relationship

There's a useful approximation that connects these two measures:

$\text{Prevalence} \approx \text{Incidence} \times \text{Average duration of disease}$

This only holds when prevalence is relatively low and the population is roughly stable, but it captures an important idea: prevalence depends on both how fast new cases appear and how long each case lasts. A disease can have low incidence but high prevalence if it lasts a long time (e.g., well-managed type 1 diabetes). Conversely, a disease can have high incidence but low prevalence if cases resolve quickly (e.g., the common cold).

Applications in Epidemiological Studies

Prevalence is most useful for:

• Cross-sectional studies that assess disease burden at a single time point (e.g., surveying obesity prevalence in a state)
• Health services planning, where you need to know how many people currently require care
• Evaluating chronic disease management to see whether programs are keeping patients healthier over time

Incidence is most useful for:

• Cohort studies that follow disease-free people over time to identify causes of disease (e.g., the Framingham Heart Study tracked thousands of participants to uncover cardiovascular risk factors)
• Clinical trials that measure whether an intervention prevents new cases
• Outbreak surveillance, where tracking new cases of an infectious disease like COVID-19 informs containment decisions

Using both together:

• Estimating disease prognosis and expected duration of illness
• Building predictive models for risk assessment (e.g., cardiovascular risk calculators)
• Measuring the long-term impact of public health interventions, such as whether smoking bans reduced both new lung cancer cases (incidence) and total lung cancer burden (prevalence) over a decade