Stanford Ran 2,000 Simulations on What Happens When Vaccines Disappear: Results Are Devastating

A team of Stanford University epidemiologists ran 2,000 simulations per disease to calculate what happens when childhood vaccination disappears in America, and the findings are almost too catastrophic to process.

The research, published in the Journal of the American Medical Association on 24 April 2025, was led by Stanford epidemiologist Mathew Kiang and infectious diseases physician-scientist Nathan Lo. Their model, a large-scale microsimulation covering all 50 US states and the District of Columbia, assessed what declining or vanishing childhood vaccination would mean for measles, rubella, polio, and diphtheria over a 25-year period.

ProPublica adapted the worst-case scenario, a full 25-year elimination of vaccines, in a detailed interactive published on 27 March 2026. The timing was deliberate. Health and Human Services Secretary Robert F. Kennedy Jr., founder of an antivaccination group, is reportedly considering policy changes that could cause the handful of remaining childhood vaccine manufacturers to exit the US market. The researchers told ProPublica their findings feel more relevant now than when they first published them.

How the Stanford Model Works and What It Measured

Kiang and Lo built what epidemiologists call a microsimulation: a model that assigns every individual in the US population an age, vaccination status, level of immunity, and state of residence, then simulates how diseases spread when an infected traveller arrives from abroad.

Outbreaks typically begin that way, when a US resident catches a disease overseas and returns home. Using vaccination data averaged across 2004 to 2023 to establish baseline immunity by state, the model then ran 2,000 simulations for each disease under multiple scenarios: vaccination staying flat, declining by 5 per cent to 50 per cent, and disappearing entirely.

The team chose four diseases for a specific reason. Measles, rubella, polio, and diphtheria had all been formally eliminated from the United States through vaccination. Measles was declared eliminated in 2000, poliovirus from the Americas in 1994, rubella in 2004, and diphtheria effectively after its vaccine was introduced in the 1940s.

None currently circulates within the country on an ongoing basis. That eliminated status means population immunity is now almost entirely vaccine-dependent, and any sustained drop in vaccination creates a growing pool of susceptible people as new babies are born unprotected.

The authors also included four co-researchers: Kate M. Bubar, Yvonne Maldonado, and Peter J. Hotez, alongside Lo. The full author team spanned Stanford's departments of epidemiology, infectious diseases, and paediatrics, plus Baylor College of Medicine's Texas Children's Hospital Centre for Vaccine Development.

What Happens at Current Vaccination Rates

Even before discussing the worst-case scenario, the model's findings at today's vaccination levels are alarming. The JAMA paper found that at current state-level vaccination rates, measles is already on course to become endemic again within approximately 20.9 years, appearing in 83 per cent of simulations. Under that status-quo projection, the model estimates 851,300 measles cases over 25 years, leading to 170,200 hospitalisations and 2,550 deaths.

The threshold for preventing measles re-establishment sits on a razor's edge. A 5 per cent increase in MMR (measles, mumps, rubella) vaccination would drop projected cases to just 5,800, virtually eliminating the risk.

A 10 per cent decline would produce 11.1 million cases. Kiang described measles as already sitting at a tipping point. 'We're already on the precipice of disaster,' he said in an interview with Stanford Medicine. The other diseases remain unlikely to become endemic under current vaccination rates, but that picture changes rapidly under even moderate declines.

Under a 50 per cent reduction in childhood vaccination, the model projects 51.2 million measles cases, 9.9 million rubella cases, 4.3 million polio cases, 10.3 million hospitalisations, and 159,200 deaths over 25 years. That scenario would also leave an estimated 51,200 children with post-measles neurological complications, 10,700 babies born with congenital rubella syndrome, and 5,400 people paralysed by polio.

The Worst-Case Scenario: Zero Vaccines for 25 Years

The ProPublica adaptation took the model's most extreme input, total vaccine unavailability for a full quarter century, and visualised each disease in turn. The average outcomes from those 2,000 simulations per disease are striking in scale.

For measles, the model predicts an average of 290,129 deaths over 25 years, within a 95 per cent uncertainty interval of 285,271 to 294,286 deaths. Before the vaccine, measles killed between 400 and 500 Americans annually; the model suggests a vaccine-free future could produce death tolls at roughly 100 times that annual rate. Measles spreads with extraordinary efficiency, with one infected person passing the virus to as many as 20 unvaccinated contacts. The virus can remain airborne in a room for up to two hours after the infected person leaves.

For polio, the model predicts an average of 23,066 cases of paralytic poliomyelitis, ranging from zero to 74,934. One in every 200 unvaccinated people who contracts polio is assumed to become paralysed, a rate drawn from established clinical data. Rubella, usually mild in adults, is most devastating to unborn children: the model projects 41,441 cases of congenital rubella syndrome, meaning babies born with heart defects, deafness, blindness, or intellectual disabilities. Those figures would not emerge immediately, as unvaccinated children would first need to reach reproductive age, but the model shows cases climbing steeply after about 15 years.

Diphtheria, the so-called strangling angel, is far rarer and harder to catch than measles, but kills roughly one in ten unvaccinated people who contract it. The model's average is 138,284 deaths, but the range runs from zero to more than 1.4 million, reflecting the disease's unpredictable transmission dynamics. In the worst-case simulation, over a million Americans could die from a disease most people today have never encountered.

The mathematics of herd immunity offer no sentiment: once vaccination falls below the threshold required to break transmission, the diseases do not pause to wait for a political consensus to shift.