Incidence vs Prevalence — Epidemiology's Two Core Measures
Incidence and prevalence are the two fundamental measures of disease frequency in epidemiology. They are often confused but measure distinctly different things: incidence counts new cases, prevalence counts all existing cases. Together they reveal whether a disease is rising or falling, how long patients live with the disease, and what interventions are working.
Disease frequency is the most basic measurement in epidemiology. Two parameters are used: incidence, which counts new cases of a disease occurring in a defined population during a defined time period, and prevalence, which counts all existing cases (new + old) at a given point in time or during a defined period. The distinction matters because incidence measures risk (the chance of getting the disease) while prevalence measures burden (how many people are currently affected).
Two related but distinct measures of incidence:
- Cumulative incidence (CI, also called incidence proportion, attack rate): Number of NEW cases / Total population at risk, over a specified time period. Expressed as a percentage or per 1000, per lakh. Example: 50 new TB cases in a population of 10,000 over 1 year = CI = 50/10000 = 0.5% per year = 500 per lakh per year.
- Incidence rate (IR, also called incidence density, person-time incidence): Number of NEW cases / Sum of person-time at risk. Used when follow-up duration varies between individuals. Expressed as cases per person-year. Example: 50 new cases / 8,500 person-years = 5.9 per 1000 person-years.
Incidence requires: (1) a defined population at risk (excluding those who already have the disease or who are immune); (2) a specified time period; (3) a clear case definition for 'new' cases. Incidence is the measure of risk — the probability that an individual will develop the disease in a given time period.
Two types of prevalence:
- Point prevalence: Proportion of population having the disease at a specific point in time. Example: 200 active TB cases in a population of 50,000 on 31 December 2024 = point prevalence = 4 per 1000.
- Period prevalence: Proportion of population having the disease at any time during a specified period. Includes people who had the disease at the start of the period AND those who developed it during the period. Example: diabetes period prevalence in 2024 = people who had diabetes on 1 Jan 2024 PLUS those diagnosed during 2024.
Prevalence measures burden — how many people need healthcare services for the disease at any given time. It is the measure used for resource allocation, hospital bed planning, and drug procurement.
Under steady-state conditions (stable incidence and duration), prevalence (P) is approximately:
P ≈ I × D
where I = incidence and D = average duration of disease.
This relationship explains two ways to reduce prevalence:
- Reduce incidence (primary prevention — fewer new cases)
- Reduce duration (cure patients faster, or shorten survival through death — perverse but mathematically true)
Conversely, prevalence rises when: incidence rises (epidemic), OR duration increases (better treatment that prolongs life without cure — e.g., HIV/AIDS with ART, diabetes with insulin). This is why HIV prevalence rose sharply in the 2000s even as incidence fell — ART extended patients' lives.
| Disease | India Incidence | India Prevalence | Why the difference? |
|---|---|---|---|
| Tuberculosis | 210 per lakh per year (new cases) | 211 per lakh (active cases at any time) | Treatment lasts 6 months — most patients are cured |
| HIV | Low (declining) | 0.20% adult prevalence | ART extends life — prevalence stays elevated even as incidence falls |
| Diabetes (T2DM) | Rising | 9.6% adult prevalence | Chronic disease — once diagnosed, lifelong |
| Hypertension | Rising | 25% adult prevalence | Chronic, often undiagnosed for years |
| Malaria | Seasonal, varies by year | Low (acute, short-duration) | Acute disease with rapid cure or death — prevalence tracks incidence closely |
Incidence and prevalence are the two most fundamental measures in epidemiology. For UPSC CMS aspirants, the definitions, formulas, the P ≈ I × D relationship, and the Indian disease-specific examples are highly testable PSM topics.