An incidence rate measures how frequently new events (new cases of disease, accidents, financial defaults, etc.) occur in a defined population over a specified period. It is a core metric in epidemiology and is also used in market research and finance to estimate the likelihood of new, time-bound events (for example, new foreclosures or loan defaults). Unlike prevalence, which counts existing cases at a point in time, incidence counts only new occurrences during the observation period.
Key takeaways
– Incidence measures new cases; prevalence measures all existing cases at a point in time.
– Two common formats:
• Cumulative incidence (risk, proportion): new cases / population at risk during a fixed period.
• Incidence rate (person-time, incidence density): new cases / total person-time at risk.
– Incidence rate ratio (IRR) = incidence rate of group A ÷ incidence rate of group B; used to compare rates between groups.
– Correct denominator (population at risk) and time framing are essential for valid interpretation.
– Incidence is used in public health, clinical trials (adverse-event rates), market research, and financial risk analysis.
How incidence rates work (conceptual)
– Define the event (what counts as a “case”).
– Define the population at risk (exclude those who already have the event or are not susceptible).
– Choose the observation period (e.g., one year, 6 months).
– Count only new events that occur during that period.
– Express the result as a proportion, a rate per N people (e.g., per 100,000), or per unit person-time (e.g., cases per 1,000 person-years).
Formulas and how to calculate
1) Cumulative incidence (risk or incidence proportion)
– Formula: cumulative incidence = (number of new cases during period) / (population at risk at start of period)
– Example: If 200 new foreclosures occur among 10,000 homeowners in one year → 200 / 10,000 = 0.02 = 2%.
2) Incidence rate (person-time or incidence density)
– Formula: incidence rate = (number of new cases) / (total person-time at risk)
– Person-time is the sum of the time each person was observed and at risk (e.g., person-years).
– Example: If 10 people develop a disease over 2,500 person-years of observation → incidence rate = 10 / 2,500 = 0.004 per person-year = 4 per 1,000 person-years.
3) Incidence rate ratio (IRR) — comparing two groups
– Formula: IRR = incidence rate in group A / incidence rate in group B
– Interpretation: IRR > 1 means higher incidence in group A; IRR 1 → group A has a higher incidence rate than group B by factor IRR. Example: IRR = 2 means twice the rate.
– IRR use person-time.
– Small numbers create unstable rates; present confidence intervals.
– Comparisons across populations need standardization (age, sex) to be fair.
Fast facts (selected)
– Incidence is used to assess risk and inform prevention strategies.
– In U.S. surveillance data, TB is often reported as cases per 100,000; HIV incidence in 2022 was reported as 11.3 per 100,000 people (U.S. statistics).
– In clinical trials, regulatory reviewers use incidence rates of adverse events to evaluate drug safety.
The bottom line
Incidence rates quantify how quickly new events occur in a population during a defined period. Correct definition of the event, clear specification of the at-risk population, and appropriate choice between cumulative incidence and person-time incidence are central to producing valid, interpretable results. Incidence measures are widely used across public health, clinical research, market research, and financial risk analysis for planning, evaluation, and decision-making.
Sources and further reading
– Investopedia. “Incidence Rate.” by user)
– Centers for Disease Control and Prevention. “Tuberculosis — United States, 2023.”
– Food and Drug Administration. “Conducting a Clinical Safety Review of a New Product Application and Preparing a Report on the Review.”
– HIV.gov. “U.S. Statistics.”
– Walk through a custom incidence calculation with your own data (e.g., cohort counts and follow-up times).
– Provide formulas and a short Excel template for calculating cumulative incidence, incidence rates, and IRRs with confidence intervals.