Vaccines go through a multi-stage testing process that typically spans about eight years from initial development to approval. Each stage is designed to catch different types of safety problems, starting with laboratory and animal studies, moving through three phases of human clinical trials, and continuing with monitoring long after a vaccine reaches the public.
Preclinical Testing in the Lab
Before a vaccine candidate is ever given to a person, it undergoes extensive testing in laboratory settings and animal models. Researchers use a combination of cell-based (in vitro) and animal-based (in vivo) studies to assess whether the vaccine triggers an immune response and whether it causes toxic effects in living tissue. These studies also establish the identity, purity, and potency of the vaccine.
The choice of animal model depends on what researchers need to learn. Mice are commonly used to study one branch of the immune response (cellular immunity), while rabbits help researchers evaluate antibody production. Mini-pigs are a good model for vaccines delivered through the skin, and primates like baboons are sometimes used when their biology closely mirrors human physiology, such as for nasal vaccines. Hamsters have been used for specific diseases like RSV. If a vaccine shows signs of toxicity or fails to produce an adequate immune response in these models, it doesn’t move forward.
Phase 1: First Tests in People
Phase 1 trials involve a small group of generally healthy volunteers. The primary goal is safety, not effectiveness. Researchers give the vaccine at increasing doses and carefully track what happens. They record local reactions at the injection site (pain, redness, swelling) and systemic reactions throughout the body (fever, fatigue, headache, muscle aches). Blood samples are drawn to get early signals about whether the vaccine is prompting an immune response.
These trials also help determine the right dose. If a lower dose produces a strong immune response with fewer side effects, that dose moves forward. Phase 1 trials typically enroll dozens of participants and can last several months.
Phase 2 and Phase 3: Expanding the Evidence
Phase 2 trials broaden the participant pool to hundreds of people, often including groups that reflect the intended population for the vaccine, such as children or older adults. Researchers continue tracking safety while gathering more detailed data on immune response and refining the dosing schedule.
Phase 3 is where the scale increases dramatically, enrolling thousands to tens of thousands of participants. This larger group makes it possible to detect less common side effects that wouldn’t show up in smaller studies. Participants are typically randomized to receive either the vaccine or a placebo, and neither participants nor researchers know who received which (a double-blind design). Safety data is collected throughout, often for months after each dose, and the trial also measures how well the vaccine actually prevents disease in real-world conditions.
Independent Safety Boards Watch the Data in Real Time
During clinical trials, an independent group called a Data Safety Monitoring Board (DSMB) reviews the accumulating safety data at regular intervals. These board members cannot have any financial, professional, or proprietary interest in the trial’s outcome. They can’t be involved in conducting the study, and representatives of the vaccine manufacturer are not eligible to serve on the board, though they may attend open portions of meetings.
The DSMB has real authority. After each review, the board recommends whether the trial should continue as designed, be modified, or be stopped entirely. If the board identifies a serious and immediate safety concern, the chair is required to notify the sponsoring agency that same day, in writing. This structure means that even while a trial is still running, an independent group has the power to shut it down if the safety data warrants it.
FDA Review and Expert Advisory Panels
Once a manufacturer believes it has enough evidence, it submits a Biologics License Application (BLA) to the FDA. This package includes all preclinical data, manufacturing details, and the full results from every phase of clinical trials. The FDA then conducts its own independent review of the raw data. Based on an analysis of vaccines approved between 1996 and 2020, the median FDA review period from BLA submission to approval is 12 months, though some reviews have taken close to two years.
The FDA also convenes a public advisory committee called the Vaccines and Related Biological Products Advisory Committee (VRBPAC). This panel of outside experts reviews the safety, effectiveness, and appropriate use of the vaccine in public meetings where the data is presented and debated openly. The committee’s recommendations inform the FDA’s final decision, though the agency is not bound by them. The entire process from the start of clinical development to FDA approval has a median length of about 8 years.
Every Batch Is Tested Before Release
Safety testing doesn’t end with approval. Federal regulations require that every single manufactured lot of a licensed vaccine must pass a series of tests before it can be shipped. No lot can be released until these tests are complete. The specific requirements include potency testing (confirming the vaccine will produce the intended immune response), sterility testing (confirming no bacterial or fungal contamination in the final containers), and purity testing (confirming the product is free of extraneous material beyond what’s unavoidable in manufacturing). Injectable products are also tested for fever-causing substances using rabbit models, though certain product categories like viral vaccines are exempt from this particular test.
Monitoring Continues After Public Use
Once a vaccine is in widespread use, several surveillance systems track its safety on an ongoing basis. The Vaccine Adverse Event Reporting System (VAERS) accepts and analyzes reports of health problems that occur after vaccination. Anyone, including patients, parents, and healthcare providers, can submit a report. VAERS is designed to detect early warning signals rather than prove cause and effect; a report to VAERS means a health event happened after vaccination, not necessarily because of it.
The Vaccine Safety Datalink (VSD) takes a different approach. It uses health records from a large network of healthcare organizations to actively monitor vaccine safety and conduct rigorous studies on adverse events following immunization. Because VSD works with medical records rather than voluntary reports, it can compare rates of health problems in vaccinated and unvaccinated populations, making it better suited for determining whether a vaccine is actually causing a particular issue.
These post-market systems are especially important for catching very rare side effects, the kind that occur in one out of every 100,000 or one million people. No clinical trial, even a large Phase 3 study, enrolls enough participants to reliably detect events that uncommon. Post-market surveillance effectively turns the entire vaccinated population into an ongoing safety study.
How mRNA Vaccines Fit In
mRNA vaccines go through the same clinical trial phases as traditional vaccines, but there has been regulatory debate about whether they require additional preclinical testing. Because the technology involves delivering genetic instructions into cells, some regulators have considered whether mRNA vaccines should be classified alongside gene therapies, which can require biodistribution studies, genotoxicity studies, and up to 15 years of follow-up safety monitoring. In practice, mRNA vaccines have been regulated under the standard vaccine framework, since they do not alter DNA and the mRNA itself degrades quickly.
Clinical trial data has shown that mRNA vaccines tend to produce more noticeable short-term reactions (sore arms, fatigue, fever) compared to other vaccine types. A systematic review of 28 clinical trials comparing different COVID-19 vaccine platforms as boosters found that mRNA vaccines were the most reactogenic, while inactivated vaccines were the least. These reactions reflect the immune system responding vigorously and are tracked carefully during trials to distinguish them from genuine safety concerns.