What 100 Years and Over 200 Human Experiments Revealed About Cold and Flu Transmission

Over a century of controlled human experiments attempted to spread cold and flu from sick people to healthy ones. The results were far weaker than most people assume, and the methods far more questionable. This section of the knowledge base examines what that evidence actually shows, why it matters, and what alternative explanations remain under-investigated.

• 203 documented human transmission experiments found a modal outcome of zero contagion across more than a third of all studies.
• The scientific foundations of germ theory, including Koch's and Rivers' postulates, could not be satisfied by their own authors.
• Non-infectious factors, including humidity, temperature, atmospheric pollutants, airway pH, and psychological expectation, can produce cold and flu symptoms through documented physiological mechanisms.
• The most comprehensive influenza transmission experiments ever conducted, run by the US military during the 1918 Spanish flu epidemic, failed to infect a single volunteer across 25 separate protocols.
• Cell culture methods used in virology produce illness in controls at rates that undermine conclusions drawn from positive results.

What the transmission experiments actually showed

Between 1906 and 2020, researchers conducted more than 200 controlled experiments attempting to transmit cold or influenza from sick people to healthy volunteers. The participants were exposed to everything: contaminated objects, direct nasal inoculation, breathing infectious air, kissing, drinking mucus mixed into milk, and sitting in sealed rooms with people coughing in their faces. Across all 203 experiments, the most common single result was that nobody became ill.

The mean positive rate across all studies was 32%. After adjusting for inert control groups, which produced a 10% illness baseline on their own, the adjusted rate drops to around 28%. Four outlier experiments accounted for nearly half of all positive cases. Remove those, and the overall rate falls to 22%. Seventy-three experiments, more than a third of the total, produced zero cases.

These were not small or poorly resourced studies. The US Navy series conducted during the 1918 Spanish influenza pandemic used 62 healthy volunteers, all in excellent physical condition, many of whom had never had influenza. Across eight experiments at Deer Island, Boston, none became ill, including one experiment where sick men breathed and coughed directly into the faces of healthy sailors for up to 50 minutes. The same result was replicated at Angel Island and Gallups Island.

The problem with interpreting the positive results

Even the positive results are difficult to interpret. Two fundamental validity problems affect the entire experimental record. The first is external validity: direct inoculation of mucus into nasal cavities does not reflect how transmission supposedly occurs in daily life, so even a high positive rate cannot be extrapolated to the real world. The second is internal validity: almost none of the 203 experiments used positive controls, random sampling, or adequate blinding.

Without these controls, a participant who develops symptoms after inoculation could have done so from the inoculation procedure itself, from a chemical or allergic reaction to the non-viral constituents of the fluid, from the nocebo effect (illness triggered by expectation), from natural coincidental acquisition, or from an uncontrolled environmental factor. The number of high-quality experiments that unambiguously demonstrate contagion across the full record is zero.

Several specific findings undermine confidence further. In one experiment, Schmidt inoculated 43 healthy volunteers with plain saline and 19% developed a common cold, a higher rate than his active inoculation groups. In Experiment 37.2, uninfected chicken embryo fluid, containing no mucus from sick people, caused cold-like symptoms in 50% of controls. In the Hebrides island experiments, healthy volunteers housed with people carrying experimentally induced colds developed no illness across multiple exposure protocols, while a single visitor with a naturally acquired cold transmitted to three of four people he sat with.

The foundations of germ theory under scrutiny

The scientific criteria designed to prove that specific germs cause specific diseases were created by the same researchers who then failed to satisfy them. Koch's postulates, the four-step evidentiary standard for bacterial causation, could not be fulfilled by Koch himself in relation to cholera. Rivers' postulates, the equivalent framework for viruses, could not be fulfilled by Rivers. Rather than treating this as a disqualifying finding, subsequent researchers invented explanatory workarounds, most notably the concept of asymptomatic infection, to preserve the theory when evidence contradicted it.

Virus isolation presents an additional problem. The method used to detect viruses in cell culture relies on observing cell death, which virologists call a cytopathic effect, and inferring from it that a virus must be present. This inference is circular: the method is specifically designed to kill cells, and the same cell death occurs in control conditions containing no material from sick people. Cells in the control condition of the Enders and Peebles measles study died at comparable rates to cells in the experimental condition, yet the study was accepted as landmark evidence for viral isolation.

Alternative causes with documented mechanisms

Several non-infectious mechanisms can produce the same physiological changes attributed to viral infection, and each has a body of experimental evidence behind it.

Airway surface liquid (ASL), the thin fluid lining the respiratory tract, normally sits at a pH between 6.9 and 7.1. When it drops below 6.9, three compounding problems occur: epithelial cells are damaged, mucus becomes more viscous, and mucociliary clearance, the primary mechanism for removing inhaled debris, is impaired. If ASL pH falls below 6.7 for more than 24 hours, respiratory cells begin to die and slough off the airway wall. Epithelial sloughing and impaired clearance are both described in the medical literature as hallmarks of viral respiratory infection, yet they occur in response to ASL acidification without any pathogen. Three independent causes of ASL acidification have been documented: inhaled particulate matter, sulfur dioxide exposure (which is 4 to 6 times higher in winter due to increased fossil fuel combustion), and dietary acid-base imbalance.

Absolute humidity reliably predicts influenza outbreaks in temperate climates. For every 0.5 g/m3 drop below 8 to 12 g/m3, influenza risk increases by around 58%. Yet in tropical climates, influenza peaks at high humidity, directly contradicting the temperate-climate mechanism. No single viral explanation can account for both patterns simultaneously.

Cold temperature exposure has been associated with increased illness rates in both epidemiological data and controlled experiments. In one experiment, 13 of 90 participants (14.4%) who immersed their feet in icy water developed colds within five days, compared with 5 of 90 controls (5.5%). During World War I, two experiments involving a combined 8,000 soldiers found that those sent outside into cold, wet conditions fell ill at four times the rate of those who remained in warm barracks.

The nocebo effect, illness produced by negative expectation, is documented in controlled experiments. In one study, a participant was told he had received infected mucus; he developed a severe cold by that evening. The following day, he was told he had actually received saline. His symptoms resolved within an hour. The role of social contagion, through which symptoms and illness beliefs spread through shared observation and expectation, is documented in cases from the Dancing Plague of 1518 to mass psychogenic illness events in schools and workplaces in the 20th and 21st centuries.

Why this investigation matters

The question of whether colds and flu are transmitted person to person, and by what mechanism, has direct implications for how we understand illness, respond to outbreaks, and assess public health interventions. If the experimental record does not support contagion as reliably as is commonly assumed, then the non-infectious explanations deserve serious investigation rather than dismissal.

The investigation does not require accepting that viruses do not exist or that infectious disease is impossible. It requires only applying the same evidentiary standard to contagion claims that would be applied to any other causal claim: that the evidence is sufficient before the conclusion is treated as settled. Across 114 years of transmission experiments, that standard has not been met.

Where these ideas come from

The ideas in this section of the knowledge base originate from the work of Daniel Roytas, specifically Can You Catch a Cold? Untold History and Human Experiments, self-published on 23 March 2024. Roytas is a naturopath and health researcher who spent several years systematically reviewing the primary scientific literature on cold and flu transmission, assembling one of the most comprehensive summaries of human contagion experiments available in a single work. His approach is evidence-led throughout, citing primary sources directly rather than relying on secondary interpretation. If you want to engage with the full argument, the complete body of cited experiments, and the author's own conclusions, the original work is worth reading in full.

The knowledge base itself is an independent work. Every concept has been studied, rewritten from scratch, and restructured for use in a multi-source advisory system. Nothing from the original has been reproduced. The knowledge has been transformed, not copied. The source is named clearly because the ideas deserve proper credit, and because the original work stands on its own merits.