Competition between evidence-based and alternative medical paradigms in non-infectious diseases

In the study of how diseases spread and how people respond to them, scientists have long focused on infectious illnesses. In those scenarios, the behavior of a single person can directly change the risk for everyone else; if one person avoids a crowd or gets vaccinated, they lower the chance that a virus will jump to a neighbor. This creates a complex social puzzle where individual choices and biological transmission are tightly linked. However, a vast number of health challenges are not caused by germs that pass from person to person. These are non-infectious conditions, such as chronic pain or recurring metabolic issues, where the risk of falling ill does not depend on how many other people are currently sick. For decades, models of human behavior and disease have struggled to explain what happens in these non-contagious cases. If a disease cannot spread socially, why do beliefs about how to treat it spread so rapidly? Why do some populations stick with treatments that offer slower recovery times, even when better medical options are available?

Researchers Hermes Benito-Andrés, Alfonso de Miguel-Arribas, and Carlos Gracia-Lázaro have built a new mathematical framework to answer these questions. They created a model that treats the competition between two different ways of thinking about health—evidence-based medicine and alternative medicine—as a cultural battle, separate from the biology of the disease itself. In their simulation, the disease does not spread between people. Instead, the "infection" is the idea of which treatment to use. The researchers found that when people decide what to do based on how quickly they see others getting better or worse, the system behaves in a surprising way. It does not naturally settle on the single best treatment, nor does it oscillate wildly. Instead, the population gets locked into a specific outcome that depends entirely on where it started.

The core of this discovery lies in how people react to health news. The model assumes that individuals do not simply count how many people are currently healthy or sick. Instead, they react to the change in health. If a group of people using a specific treatment suddenly starts recovering faster, that positive shift creates a loud social signal that draws more people to that treatment. Conversely, if a group's health suddenly worsens, that negative shift pushes people away. The researchers found that this focus on change, rather than the current state, creates a system with a unique property: a continuous line of possible stable outcomes. In a typical system, you might expect the population to eventually converge on one perfect answer, like everyone choosing the most effective medicine. Here, that does not happen. The system can settle anywhere along a spectrum of possibilities, from a population that is almost entirely evidence-based to one that is almost entirely alternative.

Which point on that line the society actually reaches is determined by the initial conditions and the "volume" of the social signals. The researchers showed that if the social amplification of one treatment is stronger than the other—if stories of success or failure for that treatment travel faster or louder—it can tilt the balance. However, the most striking finding is that the system has a kind of memory. Once the population settles into a specific balance of treatments, it cannot easily move to a different balance unless an outside force actively pushes it. If a temporary campaign convinces people to switch treatments, the system does not just bounce back to where it was. It lands on a completely new, permanent equilibrium. The society retains a "deterministic memory" of that temporary push. The final mix of treatments the population lives with is not just a reflection of which medicine is scientifically superior, but a record of the history of how people reacted to health changes and how those reactions were amplified by social networks.

This model helps explain why some alternative medical practices remain popular even when clinical data suggests they are less effective than standard care. The researchers noted that in real-world surveys, a significant portion of the population in various countries uses homeopathy, despite systematic reviews showing its effects are often indistinguishable from a placebo. In their simulations, a treatment with a slower recovery rate can still dominate the cultural landscape if the social signals associated with it are strong enough, or if the population started with a specific distribution of beliefs. The study demonstrates that the coexistence of different medical paradigms is not necessarily a sign of confusion or a failure of science, but a natural outcome of how human groups process information about health. The system is stable, but it is not flexible; it resists returning to a previous state once it has been nudged into a new one.

The researchers also explored what happens when external forces, such as public health campaigns, try to intervene. They found that a temporary intervention that directly changes how many people choose a treatment acts like a permanent shift in the system's trajectory. Because the system relies on a specific mathematical balance that is broken by direct intervention, the population does not drift back to its original state once the campaign ends. Instead, it relaxes into a new, different stable state. This means that short-term shocks, whether they are misinformation campaigns or targeted health education, can leave a lasting structural change in the cultural composition of a society. The long-term outcome is not just about the quality of the medicine, but about the path the society took to get there.

The study relies on a simplified, deterministic model, meaning it describes the average behavior of a large population without random chance. The researchers used this approach to isolate the specific mechanism of social feedback. They proved analytically that the system lacks a single, isolated "best" solution. Instead, it possesses a continuous line of neutral equilibria. This means that for any given starting point, there is exactly one final destination, and that destination is determined by the initial health status and the initial distribution of beliefs. The model shows that the nonlinearity, or the complex behavior, of the system comes entirely from the social feedback loop, not from the biology of the disease. The disease itself is simple and linear, but the way people talk about it and react to its changes creates a complex, path-dependent history.

Ultimately, this work suggests that the persistence of less effective medical treatments is not a mystery to be solved by better data alone. It is a feature of the system itself. The researchers showed that the social visibility of a treatment, driven by the rate of change in health outcomes, is sufficient to sustain a continuum of coexisting health paradigms. Whether a society ends up with a high adoption of evidence-based medicine or a high adoption of alternative medicine depends on the specific history of how that society reacted to health changes in the past. The model reveals that societies carry a permanent record of their prior states, and that temporary interventions can rewrite that record, shifting the long-term cultural landscape in ways that last long after the intervention has ended.