Mathwords logoMathwords

Cumulative Incidence — Definition, Formula & Example

Cumulative incidence is the share of people who were disease-free at the start of a study and then developed the disease during a stated time period. It estimates an individual's risk of getting the disease over that period.

For a closed cohort followed over a fixed interval [0,t][0, t], the cumulative incidence (also called incidence proportion or risk) is the number of new cases arising in [0,t][0, t] divided by the number of individuals at risk at time 00. It is a dimensionless proportion between 00 and 11 and is meaningful only when paired with the length of the interval. It assumes everyone is followed for the full interval, with no losses to follow-up and no competing deaths.

Key Formula

CI=number of new cases during the periodnumber of people at risk at the start of the periodCI = \frac{\text{number of new cases during the period}}{\text{number of people at risk at the start of the period}}
Where:
  • CICI = Cumulative incidence, a proportion from 0 to 1, often written as a percentage or per 1,000 people
  • new cases\text{new cases} = People who first develop the condition during the follow-up period
  • people at risk\text{people at risk} = People who do not have the condition at the start and could develop it

Worked Example

Problem: A study enrolls 2,600 adults and follows them for 5 years. At enrollment, 100 of them already have type 2 diabetes. During the 5 years, 150 of the remaining adults are newly diagnosed. Find the 5-year cumulative incidence of type 2 diabetes.
Step 1: Remove the people who already have the disease. They cannot become new cases, so they are not at risk.
2600−100=25002600 - 100 = 2500
Step 2: Divide the new cases by the number of people at risk.
CI=1502500=0.06CI = \frac{150}{2500} = 0.06
Step 3: State the result with its time period. You can also scale it to cases per 1,000 people.
0.06×1000=60 per 10000.06 \times 1000 = 60 \text{ per } 1000
Answer: The 5-year cumulative incidence is 0.06. That is 6%, or 60 new cases per 1,000 at-risk adults over 5 years.

Why It Matters

In clinical trials and cohort studies, you compare cumulative incidence between an exposed or treated group and a control group. The ratio of the two gives the relative risk. Vaccine efficacy, for example, is computed as 1−CIvaccine/CIplacebo1 - CI_{\text{vaccine}} / CI_{\text{placebo}}.

Common Mistakes

Mistake: Reporting cumulative incidence without a time period, or calling it a rate.
Correction: A 6% cumulative incidence over 1 year means something very different from 6% over 10 years. Always state the interval. Cumulative incidence is a proportion. The incidence rate is a different measure that divides by person-time, such as cases per 1,000 person-years.

Related Terms

  • Control Group — Its cumulative incidence is the baseline for comparing risk
  • Response Variable — New disease occurrence is the yes-or-no response measured
  • Hypothesis Testing — Tests whether two groups' cumulative incidences truly differ
  • Relative Risk — Compares cumulative incidence between two groups as a ratio
  • Cohort — Group of people followed over time to measure it