Yes. Placebo effects can influence how you feel and the symptoms you report, especially pain and nausea. Improvement after treatment is not, by itself, proof that the treatment caused it. Expectations and the care setting can contribute, while natural recovery and other treatments can also help. Symptom relief alone does not demonstrate tissue repair or disease reversal [1][2].
What Counts as a Placebo Effect?
A placebo is an intervention used for comparison that leaves out the specific treatment being tested, such as an injection without the study drug. A placebo effect is a benefit associated with expectations or the treatment context, rather than that specific treatment. It does not mean the person is imagining their symptoms [1][3].

The placebo response is broader: it describes the changes observed in people receiving placebo. Those changes can include placebo effects, but also:
- The natural course of illness: Symptoms may ease or fluctuate without the intervention.
- Regression to the mean: Someone who seeks help during an unusually severe flare may have a less severe score later, even without an effective treatment.
- Other care: Changes in medicines, rehabilitation, rest, or daily activity can affect the outcome.
These are reasons a before-and-after comparison cannot isolate the treatment’s specific effect. They should not all be labelled placebo effects [4].
How Can Expectations Affect Results?
Expectations and the Care Setting
Expecting relief can influence symptoms. How a health professional interacts with a patient may also bring a positive response independent of the specific treatment. This can accompany active treatment as well as placebo; receiving a real treatment does not remove the influence of its context [1].
Reporting and Measurement
Feeling less pain is a meaningful outcome, but a symptom score does not tell you why it changed. A Cochrane review with outcome data from 202 trials across 60 conditions found no important clinical effects of placebo interventions in general. In some settings, it found effects on patient-reported outcomes, especially pain and nausea. Pain effects varied substantially, and the authors could not reliably separate symptom effects from biased reporting [2].
Which Results Need More Caution?
The question is not simply whether an outcome improved, but what was measured and whether a suitable comparison supports the explanation.
| Outcome | Examples | What the result can tell you |
|---|---|---|
| Patient-reported symptoms | Pain, nausea, perceived energy | The score alone cannot separate treatment effects from expectations, reporting differences, or natural change. |
| Reported daily function | Walking ability, sleep, everyday activities | Ask how function was assessed and whether the comparison group received similar care. |
| Performance tests | Walking distance or grip testing | Check that instructions, testing conditions, and assessment methods were comparable between groups. |
| Disease measurements | Imaging, laboratory results, structural changes | Use evidence for the particular disease and test. A better symptom score does not establish a change on these measures. |
This is a guide to interpreting measurements, not a ranking of proven placebo effect sizes for every outcome. For chronic pain, keep the symptom result separate from any claim that damaged tissue has regenerated [2].
What Do Stem-Cell Trials Show?
A 2025 Cochrane review of stem-cell injections for knee osteoarthritis included 25 randomised trials with 1,341 participants. Eight studies, involving 459 participants, compared injections into the knee with placebo injections. The cells’ source, preparation, and dose varied, so the findings do not establish the effectiveness of one standardised clinic protocol [5].
For pain at up to six months, seven studies with 445 participants found an average benefit over placebo of 1.2 points on a 0–10 scale (95% confidence interval: 0 to 2.5 points better). For function, seven studies with 432 participants found a 14.2-point benefit on a 0–100 scale (95% confidence interval: 3.1 to 25.3 points better). Both findings were low-certainty evidence. Lower scores represented less pain or better function [5].
The pain interval includes zero benefit at its boundary; it does not cross into harm. The estimates are uncertain and may not translate into a noticeable benefit for every patient. A peer-reviewed commentary discussed commonly used minimum important differences of 1–2 points for pain and 10–25 for function, but these are not universal cutoffs [6].
These are differences between treatment groups, not measurements of how much improvement placebo caused. The review did not assess radiographic disease progression in any included study. Serious adverse events were infrequent, but the evidence about harms was very uncertain; this is not proof of long-term safety. Joint injections also carry an infection risk [5].
Does Registration Prove a Treatment Works?
No. When considering stem-cell treatment safety, separate evidence for the exact intervention from its legal or administrative status. The FDA warns about unapproved stem-cell therapies, while its licensed cellular and gene therapy list identifies particular products. Approval of one product does not approve unrelated injections or uses [7][8].
In Japan, a regenerative-medicine provision plan falls within a separate regulatory framework. Filing a plan is not proof that an intervention works for your condition [9]. A trial registration or a PubMed listing likewise does not replace an appraisal of study methods and results.
How Do Trials Separate the Effects?
Random allocation and, where feasible, blinding help researchers compare groups receiving similar attention and follow-up. A placebo-controlled trial estimates the treatment’s additional benefit over the comparison intervention. It does not tell an individual exactly which part of their improvement came from expectations [4].
To study placebo effects themselves, researchers can compare placebo with no treatment, while accounting for differences in attention, reporting, and study design. That was the approach of the broad Cochrane placebo review. If an active-treatment and placebo group improve similarly, the trial has not established an additional benefit of the active treatment for that outcome; it has not proved that all improvement was a placebo effect [2][4].
How Should You Judge a Treatment Story?
- Look for a comparison: A testimonial describes one person’s experience, not what would have happened without treatment.
- Identify the endpoint: Less pain is different from repaired cartilage. Ask whether the claim matches what was actually measured.
- Check the match: For cell treatments, compare the diagnosis, disease stage, cell source, preparation, dose, route, and accompanying care with those in the study.
- Check duration and harms: Ask how long benefits lasted, what adverse events occurred, and whether follow-up was adequate to identify delayed problems.
These checks matter particularly when evidence comes from small, varied trials, as in the knee osteoarthritis review. Knee-injection findings should not be applied to intravenous infusions, other diseases, or anti-aging claims [5].
How Can You Track Your Own Experience?
Consider a hypothetical situation: your knee pain improves after an injection, but you also rest more and start physiotherapy. A pain score alone cannot identify which change helped. Bring a record of symptoms, activity, medicines, and rehabilitation to your follow-up visit so your doctor can assess the pattern.
Agree in advance on a symptom measure, a daily activity that matters to you, and a review date. Ask your doctor how often to record symptoms and whether any tests are appropriate. Use consistent recording conditions where possible, and do not stop prescribed treatment to create your own comparison.
A diary documents your experience; it is not a controlled trial and cannot prove that cells repaired tissue or isolate a placebo effect. Ask separately about the exact protocol’s evidence, risks, total costs, and established alternatives.
What Is the Bottom Line?
Placebo effects can influence treatment results, particularly reported symptoms. Relief is worth discussing, but it does not prove the proposed mechanism worked. In knee osteoarthritis, the 2025 review found a 1.2-point average pain benefit over placebo on a 0–10 scale, with low certainty. That between-group difference is not the size of a placebo effect. Judge the exact treatment by controlled evidence, meaningful outcomes, follow-up, and harms, not by improvement alone [5].
If you are weighing a stem-cell offer in Japan, bring your diagnosis and the exact protocol to your own doctor first, and ask the clinic for its plan name, risks, and written costs in advance.
References
- National Center for Complementary and Integrative Health. Placebo Effect. Updated January 2020.
- Hróbjartsson A, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane, CD003974. Public evidence page dated 2022; searches through March 2008.
- National Cancer Institute. Definition of placebo effect.
- U.S. Food and Drug Administration / ICH E10. Choice of Control Group and Related Issues in Clinical Trials.
- Whittle SL et al. Stem cell injections for osteoarthritis of the knee. Cochrane Database of Systematic Reviews, 2025. DOI: 10.1002/14651858.CD013342.pub2. Searches to 15 September 2023.
- Madden K. Cochrane in CORR: Stem Cell Injections for Osteoarthritis of the Knee. Clinical Orthopaedics and Related Research, 2025. Secondary commentary.
- U.S. Food and Drug Administration. FDA Warns About Stem Cell Therapies.
- U.S. Food and Drug Administration. Approved Cellular and Gene Therapy Products.
- Ministry of Health, Labour and Welfare, Japan. Regenerative medicine regulatory information.




