Incidence vs. Prevalence: Definitions, Differences, Examples

Getting your Trinity Audio player ready...

Contents

• Glossary of Key Terms
• Key Takeaways
• Introduction
• Understanding Incidence
• Understanding Prevalence
• The Relationship Between Incidence and Prevalence
• Side-by-Side Comparison of Incidence and Prevalence
• Worked Examples
• Tips for Undergraduate and First-Year PhD Students
• Incidence and Prevalence in Special Contexts
• Common Data Sources and Study Designs
• Frequently Asked Questions

Glossary of Key Terms

The following terms appear throughout this guide. Reviewing them before reading will improve comprehension of the concepts discussed.

Term	Definition
Attack rate	The proportion of a population exposed to a risk factor that develops a disease during a defined outbreak period; a special form of cumulative incidence.
Cohort	A group of individuals who share a common characteristic and are followed over time in an epidemiological study.
Cross-sectional study	A study design that measures disease and exposure simultaneously at a single point in time; used to calculate prevalence.
Cumulative incidence (CI)	The proportion of a disease-free population that develops a condition over a specified period; also called incidence proportion or risk.
Duration of disease	The average length of time a person lives with a condition; a key driver of prevalence.
Incidence	The rate at which new cases of a disease or condition arise in a population over a specified time period.
Incidence density	See incidence rate; emphasizes that person-time is the denominator.
Incidence rate (IR)	The number of new cases divided by the total person-time at risk in a defined population.
Person-time	A unit combining the number of individuals at risk and the duration they are observed; used as the denominator in incidence rate calculations.
Point prevalence	The proportion of a population that has a condition at a single specific point in time.
Period prevalence	The proportion of a population that has a condition at any time during a defined interval.
Population at risk	The subset of a population that is susceptible to developing a specific condition and therefore eligible to be counted in incidence calculations.
Prevalence	The proportion of a population that has a specific condition at a given time or over a defined period.
Prevalence odds	The ratio of prevalent cases to non-cases; used in certain analytical models.
Steady state	A mathematical condition in which the prevalence of a disease remains stable because incidence and the rate of case removal (recovery or death) are in balance.

Key Takeaways

• Incidence measures new cases over time; prevalence measures all existing cases at a point in time or over a period, making them fundamentally different tools that answer different epidemiological questions.
• The relationship P ≈ IR × D (where P is prevalence, IR is incidence rate, and D is mean disease duration) holds under steady-state conditions and explains why a high-prevalence disease is not necessarily a high-incidence disease.
• Choosing the wrong measure leads to flawed public health conclusions: incidence is preferred for studying etiology and risk factors, while prevalence is preferred for planning healthcare resource allocation.
• Disease duration is the hidden variable connecting incidence and prevalence; an effective treatment that extends survival without curing the disease will increase prevalence even if incidence remains unchanged.
• Students must pay close attention to the denominator and time frame in any epidemiological statistic: small differences in wording, such as “new cases per year” versus “cases at any one time,” signal entirely different measures.

Introduction

Epidemiology rests on two foundational measures: incidence and prevalence. Both describe how common a disease or condition is, but they do so in fundamentally different ways. Confusing the two is one of the most frequent errors in introductory public health and clinical research, and the consequences range from misallocated resources to flawed policy decisions. This guide explains each measure in depth, clarifies the mathematical relationship between them, and provides practical guidance for undergraduate and first-year PhD students who are encountering these concepts in coursework or their own research.

Understanding Incidence

What Is Incidence?

Incidence quantifies the rate at which new cases of a disease or condition develop in a population that was initially free of the condition. It is fundamentally a measure of risk or speed of disease onset. Because incidence focuses exclusively on new cases, it requires that individuals already living with the condition be excluded from the denominator at the start of observation.

Types of Incidence Measures

Two main types of incidence measures are used in epidemiological research, each suited to different study designs and research questions.

Measure	Description
Cumulative incidence (CI)	Proportion of initially disease-free individuals who develop the condition over a defined period. Expressed as a proportion or percentage. Assumes complete follow-up or uses survival analysis to handle losses.
Incidence rate (IR) / Incidence density	Number of new cases divided by the total person-time at risk accumulated by the study population. Expressed as cases per person-year (or per 1,000 or 100,000 person-years). Accommodates varying follow-up times and loss to follow-up.

How Is Incidence Rate Calculated?

The incidence rate is calculated by dividing new cases by person-time. Person-time is the sum of the time each individual spends under observation while remaining disease-free and at risk.

Formula: Incidence Rate = Number of new cases / Total person-time at risk

Example: In a cohort of 500 initially disease-free adults followed for varying periods, suppose 25 develop hypertension and the total follow-up accumulates to 2,000 person-years. The incidence rate is 25 / 2,000 = 0.0125 cases per person-year, or 12.5 cases per 1,000 person-years.

Component	Value	Notes
New cases	25	Only individuals who were disease-free at baseline and developed hypertension during follow-up are counted.
Person-years at risk	2,000	Each participant contributes time only while they are at risk; time after diagnosis, death, or loss to follow-up is excluded.
Incidence rate	12.5 per 1,000 person-years	Interpreted as: for every 1,000 disease-free adults observed for one year, 12.5 new cases are expected to arise.

When Should You Use Incidence?

Incidence is the appropriate measure when the goal is to understand causation, disease onset, or the effect of a risk factor on the development of a new condition. It is the preferred measure for etiological studies because it directly quantifies the probability of an event occurring.

• Identifying risk factors for disease onset in prospective cohort studies
• Evaluating the effectiveness of vaccines or preventive interventions in randomized controlled trials
• Monitoring outbreak dynamics, such as the attack rate of an infectious disease
• Calculating absolute risk and attributable risk in public health surveillance
• Comparing disease rates across populations or time periods with different follow-up structures

Common Pitfalls in Calculating Incidence

Several errors repeatedly appear in student work and published literature when incidence is calculated or interpreted.

Pitfall	Explanation and Correction
Including prevalent cases in the numerator	Only individuals who are disease-free at the start of the observation period should be counted as new cases. Including existing cases inflates incidence.
Using the entire population as the denominator	Individuals who already have the disease are not at risk of developing it (they already have it) and must be excluded from the denominator.
Ignoring person-time in the denominator	Using a simple head count instead of person-time conflates studies with very different follow-up periods and introduces bias.
Misspecifying the at-risk period	The observation window must be clearly defined. Vague time frames make the resulting rate uninterpretable and incomparable across studies.
Reporting incidence without a time unit	An incidence rate without a time unit (e.g., per year, per month) is meaningless. Always specify the time dimension.

Understanding Prevalence

What Is Prevalence?

Prevalence measures the proportion of a population that has a specific condition at a given moment or during a defined period. Unlike incidence, prevalence does not distinguish between old and new cases: anyone with the condition counts, regardless of when they developed it. Prevalence is a snapshot rather than a rate of change.

Point Prevalence vs. Period Prevalence

The two main forms of prevalence differ in their time dimension and are suited to different analytical purposes.

Type	Definition	Example
Point prevalence	Proportion of the population with the condition at a single, precisely defined moment.	Proportion of adults with diabetes on January 1 of a given year in a defined city.
Period prevalence	Proportion of the population with the condition at any point during a specified interval. Includes both individuals who had the condition at the start and those who developed it during the interval.	Proportion of workers who experienced at least one episode of low-back pain during a 12-month calendar year.

How Is Prevalence Calculated?

Prevalence is expressed as a proportion. The formula is straightforward, but care must be taken to define the numerator and denominator correctly.

Formula: Prevalence = Number of existing cases / Total population at a given time (or during a given period)

Example: In a survey of 10,000 adults, 1,200 report currently having asthma. Point prevalence = 1,200 / 10,000 = 0.12, or 12%.

What Factors Raise or Lower Prevalence?

Prevalence rises and falls depending on factors that affect how many people enter or leave the pool of existing cases. Understanding these drivers is critical for interpreting prevalence data.

Factor	Effect on Prevalence	Mechanism
Rising incidence	Increases prevalence	More new cases enter the prevalent pool.
Longer disease duration (e.g., better treatment, chronic disease)	Increases prevalence	Cases remain in the pool longer before dying or recovering.
Higher cure rate	Decreases prevalence	Cases leave the pool more rapidly.
Higher case fatality rate	Decreases prevalence	Cases leave the pool through death.
In-migration of cases	Increases prevalence	New cases enter the pool without arising locally.
Out-migration of cases	Decreases prevalence	Cases leave the geographic pool.

When Should You Use Prevalence?

Prevalence is best suited for questions about burden, planning, and the current state of health in a population. It is not suitable for studying causes of disease because it captures existing cases, which may have survived a selection process unrelated to the exposure of interest.

• Estimating the number of people who need treatment or services at a given time
• Planning hospital capacity, staffing levels, and medical supply procurement
• Tracking the burden of chronic diseases such as diabetes, hypertension, or chronic obstructive pulmonary disease in population surveys
• Identifying subgroups with disproportionately high disease burden for targeted public health programs
• Comparing disease burden across countries or regions using standardized cross-sectional surveys

The Relationship Between Incidence and Prevalence

The Prevalence Pot Model

A useful mental model is the prevalence pot: think of existing cases as water in a pot. Incidence fills the pot by pouring in new cases. Recovery and death drain the pot by removing cases. When the rate of filling equals the rate of draining, the water level (prevalence) remains stable, a condition called steady state.

What Is the Mathematical Relationship Between Incidence and Prevalence?

Under steady-state conditions, prevalence is approximated by the product of incidence rate and mean disease duration. This formula reveals how a disease with a low incidence rate can still have high prevalence if it lasts a long time.

Formula: P ≈ IR × D

Where P = prevalence, IR = incidence rate, D = mean disease duration (in the same time units as the incidence rate).

Disease	Incidence Rate	Mean Duration	Approximate Prevalence
Acute influenza	5 per 100 per year	0.02 years (~1 week)	≈0.1%
HIV/AIDS (pre-treatment era)	0.5 per 100 per year	10 years	≈5%
Type 2 diabetes	0.8 per 100 per year	20+ years (lifelong)	~16% or more
Ebola (acute outbreak)	High locally; rare globally	~0.04 years (~2 weeks)	Very low globally despite high local IR

Why Does This Relationship Matter Practically?

Misunderstanding this relationship leads to incorrect conclusions about disease dynamics. A declining prevalence may not mean that fewer people are getting sick; it could mean that an effective new therapy is curing people faster, reducing duration (D) even if IR is unchanged.

• A new cure that shortens disease duration will reduce prevalence even if incidence remains constant.
• An aging population may show rising prevalence of chronic diseases simply because improved survival extends duration, not because more people are contracting the disease.
• A rising incidence rate for a quickly resolving condition (such as a seasonal virus) may produce very little change in prevalence at any snapshot point in time.
• Public health campaigns targeting prevention reduce incidence; campaigns targeting treatment adherence affect duration and thus prevalence.

Limitations of the P ≈ IR × D Formula

The formula is an approximation that requires specific conditions. Students should be aware of when it does and does not apply.

Assumption	Consequence When Violated
Steady state (stable prevalence)	The formula overestimates or underestimates prevalence when disease trends are changing rapidly, such as during a new epidemic.
Low prevalence (less than ~10%)	At high prevalence, the formula underestimates the true value. More precise formulas are needed.
Homogeneous population	If disease duration or incidence differs substantially across subgroups, aggregate estimates can be misleading.
No migration effects	In-migration of cases or disease-free individuals distorts both prevalence and incidence estimates.

Side-by-Side Comparison of Incidence and Prevalence

Feature	Incidence	Prevalence	Practical Implication
What it counts	New cases only	All existing cases	A prevalent-case study cannot establish that exposure preceded disease.
Time dimension	Rate or risk over a period	Proportion at a point or period	Incidence requires follow-up; prevalence can use a single survey.
Denominator	Population at risk (disease-free)	Total population	Always check whether the denominator excludes existing cases.
Best study design	Cohort study; RCT	Cross-sectional survey	Choosing the right design depends on which measure you need.
Primary use	Etiology; causal inference	Burden; resource planning	Both are necessary; neither alone tells the full story.
Affected by disease duration?	No (directly)	Yes, strongly	Long-lasting diseases accumulate high prevalence even with low IR.
Expressed as	Rate (per person-time) or risk (%)	Proportion (%)	Always include units for incidence; prevalence is unitless.

Worked Examples

Example 1: Chronic Disease (Type 2 Diabetes)

A regional health authority surveys 50,000 adults and finds 6,000 with type 2 diabetes. Separately, a cohort study follows 10,000 diabetes-free adults for five years and records 200 new diagnoses, with total accumulated follow-up of 48,000 person-years.

Calculation	Result
Point prevalence	6,000 / 50,000 = 12.0%
Incidence rate	200 / 48,000 = 0.0042 per person-year = 4.2 per 1,000 person-years
Estimated mean duration (using P ≈ IR × D, rearranged)	D = P / IR = 0.12 / 0.0042 ≈ 28.6 years
Interpretation	Diabetes has a very long duration, which is why prevalence is high even though the annual incidence rate is relatively modest.

Example 2: Acute Infectious Disease (Influenza)

During a winter season, 5,000 new influenza cases are recorded in a city of 200,000 susceptible residents over 12 weeks. At any single week, approximately 800 residents are currently ill.

Calculation	Result
Cumulative incidence (attack rate)	5,000 / 200,000 = 2.5% over 12 weeks
Weekly point prevalence (example week)	800 / 200,000 = 0.4%
Interpretation	Despite a meaningful seasonal incidence, point prevalence at any moment is low because influenza resolves quickly (short duration).

Example 3: The Effect of a New Treatment on Prevalence

A disease has an incidence rate of 10 per 1,000 person-years and a mean duration of 8 years before treatment is introduced, yielding a prevalence of approximately 8%. A new treatment reduces mean duration to 3 years (by curing more patients) without affecting incidence.

Scenario	Calculation	Prevalence
Before new treatment	IR × D = 0.010 × 8	≈8%
After new treatment (lower duration)	IR × D = 0.010 × 3	≈3%
Interpretation	Prevalence fell by 62.5% not because fewer people got sick, but because the disease resolved faster. Incidence was unchanged.

Tips for Undergraduate and First-Year PhD Students

How Should You Approach Reading Papers for the First Time?

When reading a paper, locate the measure name in the abstract or methods, then immediately check the denominator. The denominator tells you almost everything about whether the reported figure is an incidence measure or a prevalence measure.

• Ask: does the denominator exclude people who already have the condition? If yes, it is an incidence measure; if no, it is likely a prevalence measure.
• Check whether the study followed individuals over time (cohort design = suitable for incidence) or measured everyone at one moment (cross-sectional design = suitable for prevalence).
• Look at the units: a rate expressed per person-year signals incidence; a proportion expressed as a percentage with no time unit attached usually signals prevalence.
• Be suspicious of papers that use the words “rate” and “proportion” interchangeably; these are not synonymous, and such usage may indicate methodological imprecision.

Common Exam and Assignment Mistakes to Avoid

The following errors appear frequently in undergraduate examinations and first-year PhD qualifying papers. Review each one carefully before submitting any written work.

Mistake	How to Avoid It
Describing prevalence as a “rate”	Prevalence is a proportion, not a rate. A rate requires a time dimension in the denominator. Use “prevalence proportion” if precision is needed.
Using total population as denominator for incidence	The incidence denominator must be the population at risk: people who do not yet have the disease. Remove prevalent cases before calculating.
Confusing cumulative incidence with incidence rate	Cumulative incidence is a proportion (dimensionless, between 0 and 1); incidence rate has dimensions of 1/time and can exceed 1. They require different denominators.
Applying the P ≈ IR × D formula to non-steady-state situations	Always verify whether the disease is in approximate steady state (stable prevalence over time) before applying this formula. Emerging pandemics violate this assumption.
Ignoring the time frame	Every incidence measure must be accompanied by the observation period. “Twenty new cases” is meaningless without knowing whether that occurred over one month or ten years.
Treating high prevalence as evidence of high risk	High prevalence may simply reflect long disease duration. Always look at incidence to assess actual risk of developing the disease.

How to Choose Between Incidence and Prevalence in Your Own Research

The choice of measure should follow directly from your research question. Defining the question precisely before selecting a measure prevents the most common design errors in student projects.

Research Question Type	Appropriate Measure	Rationale
What causes this disease?	Incidence	Causal inference requires establishing that exposure preceded disease onset, which only incidence-based designs (cohort, RCT) can demonstrate.
How many people in our hospital system need treatment for X?	Prevalence	Resource allocation depends on the current burden, not the rate of new cases.
Did our prevention program reduce the number of new cases?	Incidence	Program evaluation requires measuring change in the rate of new disease onset.
How has the burden of a chronic disease changed over a decade?	Both	Interpreting trends requires understanding whether changes reflect shifts in incidence, duration, or both.
What proportion of a population carries a biomarker?	Prevalence	Biomarker surveys are cross-sectional and measure the existing distribution, not onset.

Building Intuition: Quick Mental Checks

Developing fast intuitive checks will help you catch errors before they reach a supervisor or journal reviewer. Practice asking these questions reflexively whenever you encounter an epidemiological statistic.

• Can someone already have the disease and still be in the denominator? If yes, it is probably prevalence.
• Is there a time unit attached to the number (per year, per month)? If yes, it is probably an incidence rate.
• Is the study design prospective with follow-up? If yes, incidence is typically the appropriate measure.
• Is the disease acute with rapid resolution, or chronic and long-lasting? Acute diseases tend to show large differences between incidence and prevalence; chronic diseases can show very similar trends.
• Does the wording say “new cases” or “cases at a point in time”? Those phrases are your clearest signal.

Tips Specifically for First-Year PhD Students

Graduate research introduces additional complexity: you may be generating your own incidence or prevalence estimates, writing literature reviews that synthesize heterogeneous studies, or designing original studies. The following guidance targets common challenges at that stage.

• When writing a systematic review or meta-analysis protocol, specify in your inclusion criteria whether you will accept incidence studies, prevalence studies, or both, and justify that choice in relation to your research question.
• Be precise in your methods section: state explicitly whether your study estimates cumulative incidence, incidence density, point prevalence, or period prevalence, and provide the exact formula and denominator definition you used.
• When comparing your estimates to previously published figures, verify that earlier studies used the same measure and definition. A comparison between period prevalence and point prevalence of the same condition is not valid.
• In grant applications and thesis proposals, define the at-risk population with care: reviewers and ethics boards will scrutinize your denominator definition, and a vague denominator is a common reason for protocol revision requests.
• If you are using administrative health data or electronic health records, distinguish between a first recorded diagnosis (incident case) and any recorded diagnosis (prevalent case), as these databases often contain existing patients at baseline, creating a prevalent-case bias if not handled correctly.
• Consult a biostatistician before finalizing your study design: the decision between a person-time denominator and a simple population denominator has direct consequences for the statistical models you can use and the assumptions those models require.

Incidence and Prevalence in Special Contexts

Infectious Disease Epidemiology

In infectious disease settings, incidence is often reported as an attack rate (cumulative incidence during an outbreak) or as a secondary attack rate (proportion of susceptible household or close contacts who develop the disease after exposure to a primary case). Prevalence matters less in acute outbreaks but becomes critical for chronic infections such as HIV or hepatitis C, where the treated and untreated case pools both require monitoring.

Measure	Use in Infectious Disease	Example
Attack rate (cumulative incidence)	Quantifies overall risk during a defined outbreak	30% of wedding guests developed norovirus within 48 hours of the event.
Secondary attack rate	Measures transmission efficiency within a contact group	25% of susceptible household members developed COVID-19 after exposure to an index case.
Point prevalence of infection	Surveys active infections at a moment in time; used in HIV surveillance	0.3% of adults in a national survey tested positive for active HIV infection in a given year.

Mental Health Epidemiology

Mental health research frequently uses period prevalence (for example, 12-month prevalence) rather than point prevalence, because many mental health conditions are episodic: a person may be well on the day of a survey but may have had a diagnosable episode in the preceding year. Incidence of first onset is also tracked but requires careful distinction from recurrence, relapse, and recrudescence.

• Lifetime prevalence captures anyone who has ever met diagnostic criteria and is not a measure of the current burden on health systems.
• 12-month prevalence is the most commonly reported figure in global mental health surveys such as the World Mental Health Surveys.
• Incidence of first onset is a more demanding measure to calculate, requiring confirmation that no prior episode existed; it is therefore less frequently reported.

Cancer Epidemiology

Cancer registries report both incidence (new cancer diagnoses per year per 100,000 population) and prevalence (total number of people living with a diagnosis). The difference is striking for cancers with improving survival: five-year prevalence of certain cancers has risen sharply not because more people are getting cancer, but because more are surviving longer after diagnosis.

Cancer Registry Measure	Definition
Age-standardized incidence rate	Incidence rate adjusted for the age distribution of a standard reference population, enabling comparisons across countries or time periods with different age structures.
Five-year prevalence	Number of people who have been diagnosed with a specific cancer in the past five years and are still alive, regardless of disease status.
Net survival	Probability of surviving a specific cancer for a defined period in the hypothetical absence of other causes of death; related to but distinct from prevalence.

Common Data Sources and Study Designs

Data Source / Design	Incidence or Prevalence?	Key Strengths and Limitations
Prospective cohort study	Incidence (best)	Establishes temporal sequence; expensive and time-consuming; loss to follow-up can introduce bias.
Cross-sectional survey	Prevalence (best)	Fast and inexpensive; cannot establish causality; subject to survivor bias for conditions with high case fatality.
Disease registry (e.g., cancer, diabetes)	Both	Covers defined populations over time; completeness of registration varies; definitions may change over time.
Electronic health records	Both (with caveats)	Large sample sizes; incident cases require careful identification of first diagnosis; may underrepresent undiagnosed or uninsured individuals.
Vital statistics and mortality data	Incidence of death; not disease incidence	Complete in many settings; conflates disease incidence with case fatality; not suitable alone for estimating disease incidence.
Serological surveys	Prevalence of infection or immunity	Captures both symptomatic and asymptomatic cases; cross-sectional; cannot determine timing of infection without serial sampling.

Frequently Asked Questions

Can incidence ever be greater than 1 (or 100%)?

Cumulative incidence, expressed as a proportion, cannot exceed 1 (or 100%) because it is bounded by the number of people at risk. However, incidence rate (expressed per person-time) can technically exceed 1 per person-year if the event is highly common and the follow-up window is long. For example, in a study of common colds in a school-age population, children may average more than one cold per year, producing an incidence rate greater than 1 per person-year.

Is prevalence always lower than cumulative incidence over the same period?

Not necessarily. Period prevalence over a given interval will generally be higher than point prevalence at the start of that interval because it captures new cases arising during the period as well as those present at the start. Cumulative incidence counts only the new cases, so period prevalence will be higher when existing (baseline) cases are numerous. The comparison depends on the time frame and the disease characteristics.

Why does prevalence sometimes fall even when incidence is rising?

This counterintuitive pattern can occur when a new treatment dramatically shortens disease duration (D) at a rate that outpaces the rise in incidence rate (IR). Because P ≈ IR × D, if D falls faster than IR rises, the product (prevalence) will decrease. This was observed historically for certain infections when highly effective short-course therapies became available.

How do I handle competing risks when calculating incidence?

Competing risks arise when individuals can exit the at-risk pool for reasons other than developing the disease of interest, for example, dying from an unrelated cause. Standard incidence rate calculations remove such individuals from the denominator at the time they exit. More advanced methods, such as cause-specific hazard models and cumulative incidence functions, are used in research to partition competing events appropriately. First-year students should be aware that ignoring competing risks can overestimate the risk of the event of interest.

What is the difference between incidence and incidence rate?

In everyday usage, “incidence” sometimes refers loosely to cumulative incidence (a proportion, dimensionless) and sometimes to incidence rate (cases per person-time). Formally, “incidence rate” or “incidence density” refers specifically to the person-time-based measure. Being explicit about which measure you are reporting prevents ambiguity in your writing and communication.

Can prevalence be used to draw causal inferences?

Prevalence alone is insufficient for causal inference for two reasons. First, prevalent case studies cannot confirm that exposure preceded the disease: the disease or its treatment may have altered the exposure. This is known as Neyman bias or prevalence-incidence bias. Second, prevalent cases are survivors: individuals who died quickly after disease onset are excluded, which may systematically differ from those who survive long enough to be surveyed. Incidence-based designs are required for etiological conclusions.

Are there situations where prevalence is more useful than incidence for policy?

Yes. For chronic, noncommunicable diseases where the goal is to estimate current need for treatment, care, and support services, prevalence is directly applicable. Budget planning for dialysis units, insulin supplies, antihypertensive medications, or mental health services all depend on knowing how many people currently have the condition, not how many will newly develop it in the coming year. For such planning purposes, prevalence is the primary metric, and incidence is used as a secondary indicator to project future burden.

How does screening affect the measured incidence and prevalence of a disease?

Introducing or expanding a screening program creates a temporary surge in apparent incidence (detected new cases) without necessarily reflecting a true increase in disease occurrence. This is called lead-time bias or length-biased sampling in more complex scenarios. Simultaneously, prevalence measured by the screening program will be higher than that measured without screening, because previously undetected cases are now identified. Students should interpret changes in reported incidence or prevalence around the introduction of new screening programs with caution.