Dummy pills reveal the hidden power of human expectation in addiction recovery

Patients being treated for addiction routinely experience a reduction in substance cravings simply from taking an inactive dummy pill, particularly those dealing with alcohol dependence. These findings suggest that the medical community might be able to harness a patient’s expectations to improve treatment outcomes. The research was published in The Journal of Clinical Psychiatry.

A placebo is a biologically inactive substance, such as a sugar pill, that is used in medical experiments to simulate an actual drug. Doctors use these inert pills to establish an essential baseline of comparison. By giving one patient group a real medicine and another group an inactive substitute, researchers can isolate the true chemical impact of a new pharmaceutical.

These inactive substances frequently produce genuine changes in human health. When a patient takes a dummy pill, their symptoms often improve in an observable way. This happens in part because the human brain anticipates relief, setting off biological responses that occasionally mirror actual medication.

Repeated behavior also shapes this physical response. The simple ritual of going to a clinic, interacting with a health care provider, and swallowing a daily pill conditions the mind to expect a therapeutic benefit. This combination of expectation and routine can powerfully reduce feelings of distress, physical pain, and overwhelming anxiety.

Medical researchers have extensively documented this phenomenon across a wide array of mental health conditions. A recent analysis of nine different psychiatric diagnoses showed enormous variation in symptom improvement from dummy pills alone. Conditions tied closely to mood, like depression, exhibited massive improvements when patients simply believed they were receiving treatment.

Free daily newsletter

In contrast, conditions grounded in different brain patterns, like schizophrenia, hardly responded to dummy pills at all. That prior analysis left out one major category of mental health struggles entirely.

“Bschor et al. (2024) did a great meta-analysis on placebo treatment across nine psychiatric disorders and showed that symptom severity improved with placebo in all psychiatric disorders, but varied between the disorders,” said Dr. Pim B. van der Meer, a physician and researcher at University Medical Center Utrecht in the Netherlands. “They did not include substance use disorders, despite substance use disorders being such a common and severe problem. This motivated us to conduct our current study.”

Van der Meer and a team of European experts wanted to figure out exactly how much dummy pills might alleviate the core symptoms of addiction. They focused their attention entirely on the sensation of craving.

Craving is defined as a persistent, intense urge to consume a specific drug. It is a central feature of addiction, frequently driven by environmental stress or the initial stages of biological withdrawal. Because craving is a subjective experience reported by the patient on a continuous numerical scale, it can be tested mathematically in a way that is highly similar to measuring feelings of sadness or ambient worry.

The research team conducted a systematic review and meta-analysis to search for broad patterns. This process involves grouping together all the available numerical data from past experiments to find an overarching mathematical trend. They searched eight different global research databases using terms related to addiction, craving, and placebos.

Their initial sweep yielded almost 2,000 unique records. The researchers read through study titles and abstracts to weed out anything that did not meet strict scientific criteria. They gradually whittled the list down to 115 full articles, which they read in full before finalizing a group of nine studies.

“What surprised me the most is the limited number of RCTs that were eligible,” van der Meer said.

All nine of these selected studies were randomized controlled trials. In these types of medical experiments, patients are assigned to medication groups entirely by chance. They were also strictly double-blind, meaning neither the actual patients nor the doctors administering the treatment knew who was receiving the authentic medication.

The included trials focused on oral treatments for three distinct forms of addiction. Five trials looked at individuals recovering from alcohol dependence. Two studies focused on cocaine dependence, and another two studies tracked individuals trying to stop using methamphetamine.

Across the nine studies, the researchers gathered data on 495 patients who were placed into placebo groups. This included 300 patients dealing with alcohol addiction. The stimulant groups featured fewer participants, with 88 patients for cocaine and 107 patients for methamphetamine.

The scientific team extracted specific numbers regarding how strongly the patients rated their urges to use substances before the trial began. They then pulled the exact same numbers collected at least four weeks into the treatment process. They specifically looked at people taking a daily oral capsule rather than a patch or injection to ensure the physical routines were identical.

The researchers noticed a distinct pattern across the gathered data. Following four weeks of daily placebo treatment, craving scores decreased consistently across all three types of addiction. Simply taking a dummy pill and routinely interacting with the medical system eased the intense desire to consume drugs.

For individuals with an alcohol use disorder, the therapeutic benefit was highly robust. The combined data from the five alcohol trials demonstrated a mathematically reliable drop in subjective cravings over time. The patients believed they were getting help, and their subjective urges declined accordingly.

The results for the other substances were less definitive. While the symptom numbers did drop for cocaine and methamphetamine users, the results were not statistically significant. This means the researchers could not definitively prove the drop was caused by the dummy pill rather than normal day-to-day fluctuations in mood.

There are multiple reasons why alcohol cravings might respond differently to dummy pills than stimulant cravings. The biological mechanisms underlying methamphetamine and cocaine addiction may exert a more aggressive chemical hold on the brain, leaving less room for expectation-driven relief. Additionally, alcohol addiction often involves a different long-term recovery trajectory than continuous stimulant abuse.

The small number of available studies also presents a major mathematical hurdle. With only two studies for cocaine and two for methamphetamine, the pool of patients was quite limited. The researchers noted that these particular estimates were captured with a high degree of mathematical uncertainty.

The team faced similar challenges when looking at other common substances. They had to exclude therapies for smoking and nicotine addiction entirely. Most nicotine trials combine pharmacological treatment with some form of behavioral therapy or professional counseling.

When talk therapy is combined with a medical trial, it becomes virtually impossible to untangle the varying effects. Researchers cannot determine whether a drop in craving was caused by a placebo pill or by the emotional support given by the therapist. Sticking strictly to pill-only trials preserved the structural integrity of the data.

Another limitation was that the original trials rarely verified whether the scientific blinding actually worked. In many medical experiments, patients sometimes guess correctly that they are taking a dummy pill because they do not feel any bodily side effects. If patients guess their status correctly, their optimistic expectations drop and the placebo response vanishes.

The research team also pointed out a high degree of variation among the included trials. Differing lengths of treatment, differing patient demographics, and the use of unique craving questionnaires made it harder to compare the studies perfectly. This is a common obstacle when pooling older medical data into a single modern review.

“It is important to note that the observed improvement most likely reflects a combination of the true placebo effect including expectancy, the natural course of the condition, and other nonspecific effects,” van der Meer said. “A meta-analysis of this kind cannot determine the precise extent to which each of these factors accounts for the observed improvement.”

Despite the limitations, the study offers practical insights for both everyday medicine and future laboratory design. When setting up a new trial strictly for addiction medication, scientists must account for a high background rate of improvement. A new drug has to clear a very high hurdle to outpace the sheer power of human expectation.

In everyday medical clinics, doctors might be able to ethically leverage this psychological response. If they prescribe a genuine, evidence-based medication, they can also project a professional sense of confidence and optimism about the treatment. Fostering a hopeful outlook from the start could enhance the biological effectiveness of the real medicine.

“We do now know quite well that symptom improvement occurs with placebo treatment across a large variety of psychiatric disorders,” van der Meer said. “To truly leverage the placebo effect in clinical practice, we need to know better to what extent the symptom improvement with placebo treatment consists of the true placebo effect, the natural course, and other nonspecific effects.”

The study, “Placebo Effects in Alcohol, Cocaine, and Methamphetamine Use Disorder: A Systematic Review and Meta-Analysis,” was authored by Pim B. van der Meer, Jan W. Schoones, Jurjen Luykx, Olaf M. Dekkers, Marta Fiocco, Cornelis Kees Kramers, Arnt Schellekens, and Albert Batalla.