Why Now Daily.

Published

Vaccine Efficacy vs Effectiveness: Trial and Real-World Measures

Vaccine efficacy estimates relative risk reduction under controlled trial conditions, while vaccine effectiveness estimates performance in routine real-world use, usually through observational data.

Timeline

  1. 2024-08-14: CDC updated its public methods explainer for influenza vaccine efficacy and effectiveness studies.
  2. 2025-03-10: WHO published an updated explainer distinguishing vaccine efficacy, effectiveness and protection.
  3. 2026-10-12: Current WHO, CDC and peer-reviewed methodological sources were reviewed for this guide.

Vaccine efficacy and vaccine effectiveness both compare the frequency of a defined outcome in vaccinated and comparison groups, but they describe different study settings. Efficacy usually comes from a randomized controlled trial in which assignment, storage, dosing and follow-up are tightly managed. Effectiveness describes performance in routine practice, commonly using observational studies after rollout. The distinction is about how evidence was generated. It does not mean efficacy is “real” and effectiveness is unreliable, or that every real-world estimate must be lower than the trial result. [1][2][3]

The familiar percentage is a relative risk reduction for the study outcome. If 10 of every 1,000 people in the comparison group develop that outcome and 2 of every 1,000 vaccinated people do, the risk ratio is 0.2 and vaccine efficacy or effectiveness is 1 minus 0.2, or 80%. That means an 80% lower risk during that study, not that 80% of recipients are completely protected or that each vaccinated person has a fixed 20% chance of illness. The absolute risk reduction in this example is 8 cases per 1,000. [1][4]

Every percentage needs an endpoint. Protection against laboratory-confirmed infection, symptomatic illness, hospitalization, severe disease and death are different questions and can produce different values for the same vaccine. A trial may require a particular symptom definition and testing protocol; a real-world study may rely on health records or a test-negative design. Comparing “95% against symptomatic disease” with “80% against hospitalization” as though the larger number names a better vaccine ignores what was counted. [1][2]

Randomization helps balance measured and unmeasured differences between trial groups, reducing selection bias and confounding. Blinding and standardized case detection can reduce other systematic errors. Trials still apply only to the enrolled population, products, schedules, circulating strains, follow-up period and endpoints studied; rare outcomes or underrepresented groups may remain uncertain. A confidence interval shows statistical precision around the estimate. A point estimate without its interval, case counts and protocol is an incomplete summary of the evidence. [1][2][3]

Effectiveness studies show what happens when eligibility, access, storage, dose timing, prior infection, health status and exposure vary in practice. Cohort, case-control and test-negative designs can study very large populations and outcomes that trials were not powered to measure. Because vaccination is not randomly assigned, researchers must address confounding, selection, outcome classification and differences in health-seeking behavior. Multiple well-designed studies across systems can strengthen confidence, while disagreement may reflect methods or populations rather than a biological contradiction. [2][3][5]

Time is part of the estimate. Immune protection takes time to build, may change after additional doses and can wane. The pathogen may evolve, and the dominant strain or season can change between a trial and later effectiveness study. Age, immune status and comorbidities may modify protection, while the interval since vaccination changes who belongs in each comparison. A useful statement therefore reads like “effectiveness against outcome X in population Y during period Z after schedule S,” rather than attaching one permanent number to a brand. [1][2]

Program impact is broader than individual direct effectiveness. High coverage can reduce transmission and create indirect protection, while missed doses, unequal access or weak delivery can limit cases prevented even if the product performs well among recipients. Researchers distinguish direct, indirect, total and overall effects depending on the comparison. Immunogenicity is another separate concept: an antibody or cellular response may support vaccine evaluation, but it is not automatically identical to demonstrated protection against a clinical outcome unless a validated correlate links them. [3][5][6]

To read a vaccine claim, identify the study design, exact outcome, comparison group, population, dose schedule, follow-up window, circulating strain, case counts, confidence interval and whether the number is relative or absolute. Then check who funded the study, whether the protocol was prespecified and whether independent evidence agrees. An efficacy or effectiveness estimate informs policy and personal decisions but does not guarantee an individual result. Use current recommendations from the relevant health authority and a clinician who knows the person’s age and medical circumstances. [1][2][3][4]

Sources

  1. World Health Organization — Vaccine Efficacy, Effectiveness and Protection
  2. CDC — How Flu Vaccine Effectiveness and Efficacy Are Measured
  3. CDC — Manual for the Surveillance of Vaccine-Preventable Diseases: Definitions
  4. CDC — Principles of Epidemiology: Measures of Association
  5. PubMed Central — Distinguishing Vaccine Efficacy and Effectiveness
  6. World Health Organization — Measuring Vaccine Effectiveness

Related stories