Incidence (epidemiology) Study Guide

📖 Core Concepts Incidence – counts new cases occurring in a defined population during a set time period. Incidence Proportion (Cumulative Incidence) – probability that an individual free of disease at baseline becomes a case by the end of follow‑up. Incidence Rate – new cases divided by total person‑time at risk; accounts for varying entry/exit times. Prevalence – proportion of all cases (new + existing) present at a specific point or period; a snapshot of disease burden. Person‑time – sum of each participant’s time under observation while at risk (e.g., person‑years). Attack Rate – incidence proportion applied to an outbreak setting (short‑term, defined population). Attributable Risk – difference in incidence between exposed and unexposed groups; isolates the effect of the exposure. Rate Ratio – ratio of two incidence rates; tells how many times higher (or lower) the rate is in one group versus another. --- 📌 Must Remember Incidence Proportion = # new cases ÷ # persons at risk (baseline). Incidence Rate = # new cases ÷ total person‑time at risk. Prevalence = # total cases ÷ total population. Use incidence proportion when the risk set stays essentially constant over a fixed follow‑up. Use incidence rate when follow‑up times differ or loss‑to‑follow‑up is common. Approximation: Prevalence ≈ Incidence × Average disease duration (only when incidence ≈ constant & duration stable). Attack rate = # cases during outbreak ÷ # people at risk in the defined population. Attributable risk = Incidenceexposed − Incidenceunexposed. Rate ratio (RR) = Incidence rate in group A ÷ Incidence rate in group B. --- 🔄 Key Processes Calculate Incidence Proportion Identify cohort free of disease at start. Count new cases during the observation window. Divide by the number at risk at baseline. Calculate Incidence Rate For each participant, record time contributed while at risk. Sum all times → total person‑time. Divide # new cases by total person‑time. Calculate Prevalence Count all existing cases at the chosen time point (new + old). Divide by the total population size (or subgroup size). Derive Attack Rate (Outbreak) Define the exposed population during the outbreak. Count cases that arise within the short exposure period. Divide cases by the at‑risk population. Compute Attributable Risk & Rate Ratio Obtain incidence (or rate) for exposed and unexposed groups. Subtract to get attributable risk. Divide to get rate ratio. --- 🔍 Key Comparisons Incidence Proportion vs. Incidence Rate Incidence Proportion: uses number of people; assumes uniform follow‑up. Incidence Rate: uses person‑time; handles variable follow‑up. Incidence vs. Prevalence Incidence: measures risk of new disease. Prevalence: measures burden (all cases) at a moment. Attack Rate vs. Incidence Proportion Attack Rate: same formula but limited to a brief outbreak window. Incidence Proportion: used for routine cohort follow‑up over longer periods. Attributable Risk vs. Rate Ratio Attributable Risk: absolute difference in risk (or rate). Rate Ratio: relative comparison (how many times higher). --- ⚠️ Common Misunderstandings “Incidence = Prevalence” – false; incidence counts new cases, prevalence counts all cases. Using incidence proportion when follow‑up varies – leads to biased risk estimates; switch to incidence rate. Assuming prevalence always reflects current incidence – ignores disease duration; a chronic disease can have high prevalence despite low incidence. Treating attack rate as a general incidence measure – only valid for short, well‑defined outbreaks. --- 🧠 Mental Models / Intuition “Water tank” analogy: Incidence is the faucet flow (new water entering), Prevalence is the water level in the tank (total water). The level rises when inflow exceeds outflow (recovery or death). Person‑time as “seat‑hours”: each participant occupies a seat for the time they’re at risk; total seat‑hours = person‑time. Duration bridge: If people stay sick long, the tank fills (high prevalence) even if the faucet drips slowly (low incidence). --- 🚩 Exceptions & Edge Cases Approximation Prevalence ≈ Incidence × Duration fails for: Age‑specific measures where incidence or duration changes with age. Situations with rapidly changing incidence (e.g., epidemics). When the risk set changes dramatically (mass migration, massive loss‑to‑follow‑up), neither simple incidence proportion nor rate may capture true risk without advanced modeling. --- 📍 When to Use Which Fixed, complete follow‑up, stable risk set → use Incidence Proportion. Variable entry/exit, censoring, or loss‑to‑follow‑up → use Incidence Rate (person‑time). Quick snapshot of disease burden → compute Prevalence (point or period). Outbreak with short exposure window → calculate Attack Rate. Assessing impact of an exposure → compute Attributable Risk (absolute) and Rate Ratio (relative). --- 👀 Patterns to Recognize High prevalence + low incidence → disease of long duration (e.g., diabetes). Low prevalence + high incidence → acute, short‑lasting disease (e.g., influenza). Incidence rate expressed per 1,000 or 100,000 person‑years → watch for denominator units. Questions that give person‑time → they are prompting an incidence rate calculation, not proportion. --- 🗂️ Exam Traps Mistaking person‑years for persons – a common distractor when the answer choices list “cases per 1,000 persons” instead of “per 1,000 person‑years”. Using baseline cases in denominator – baseline prevalent cases must be excluded from the risk set for incidence calculations. Choosing prevalence when the question asks for risk – prevalence does not convey the probability of becoming a case. Assuming attack rate works for chronic diseases – attack rate is reserved for short‑term outbreaks; applying it to chronic conditions is a red flag. Ignoring disease duration in prevalence‑incidence relationship – if the answer relies on the approximation, verify that incidence is stable and duration is constant.