The placebo effect can still work, even if people know it's a placebo

"The placebo effect is real – even if you know the treatment you've been given has no medical value, research has concluded," the Mail Online reports. The study in question aimed to further understand how placebos – inactive or dummy treatments – work.

The research involved 40 volunteers who took part in a series of experiments where a heat sensor was applied to their arm. Before the heat application, petroleum gel (Vaseline) was applied to the skin. The researchers added a blue dye to one of the batches and told the volunteers it was a pain relief gel.

The researchers ran a series of conditioning tests where they applied the blue gel or the plain gel to the skin before the heat. What they were actually doing was applying low heat after the blue gel and high heat after the plain gel.

The longer this "conditioning" went on, the greater effect it had. Even when the dyed blue gel was revealed as an identical inactive gel, some pain relief was still experienced by those who had four days of this conditioning, compared with people who had only one day.

While interesting, the study has limited direct applications. The results cannot easily inform the effect a placebo may or may not have in real-life situations.

However, the results reinforce the notion the psychological can have just as big an impact as the physical when it comes to coping with chronic pain

Where did the story come from?

The study was carried out by researchers from the University of Colorado Boulder and the University of Maryland Baltimore in the US, and was funded by the National Institute of Mental Health.

It was published in the peer-reviewed Journal of Pain.

The Mail has a simplistic take on what was quite a complex experimental study and analysis. Its reporting could benefit from recognising the limitations of this experimental research. 

What kind of research was this?

This experimental study aimed to investigate how placebo (inactive) painkillers work.

The researchers explain how recent research has suggested placebo pain relief is mediated by expectations. "Expectancy theory" implies a belief in the placebo is essential for it to work.

This study aimed to see whether placebo painkillers would work if the person was aware they were only receiving placebo, by testing the effects before and after use.

The researchers believed it was all to do with expectancy – if there was enough prior conditioning, the placebo's effect would still persist, even if was later revealed as a placebo. 

What did the research involve?

This experimental study recruited 54 adults (30 men and 24 women aged 18 to 55) via university advertisements.

They were given an initial test to assess their pain response to a thermal stimulus that would be used during the experiments. Those that did not find it sufficiently painful were excluded, leaving 40 participants (27 women and 13 men).

The participants were told they were taking part in a test comparing the painkilling effects of a cream containing an active painkilling ingredient (the placebo) with a cream containing no active ingredients (the control).

Both creams were in fact the same petroleum jelly containing no active ingredients – the only difference being the placebo was blue.

The experiments were in four stages: calibration, placebo manipulation, conditioning, and testing.

Calibration phase

Sixteen different temperature stimuli were given on eight sites of the volunteers' forearms. They were asked to respond on a visual analogue scale from 0 (no pain) to 100 (worst pain imaginable).

From this, six temperatures were derived for each individual for the remaining experiment: two low, two medium, and two high pain stimuli.

Placebo manipulation

The participants were told about the composition of the placebo cream, the active ingredients it contained and possible side effects.


This involved sessions where the person was given either the placebo or control cream before having the heat stimulus applied.

The difference was each time they gave "the placebo" the researchers followed this up by applying a low-heat stimulus, whereas when they gave "the control" they followed this up with a high-heat stimulus.

The participants had been divided into two groups of 20: a short group, who had only one conditioning session, and a long group, who had this conditioning given on four separate days.


This began after the last conditioning session. Participants were given a few runs with the placebo and control creams, each time being asked to assess on the visual scale how much pain relief they expected to receive with the coming heat stimulus.

The placebo was then revealed to be inactive and identical to the control cream. After a 15-minute delay, they were again tested with the placebo and control creams. 

The researchers compared differences between the creams in expected pain relief before and after the reveal, and with the effect of the short or long conditioning.  

What were the basic results?

The analysis of this study was in-depth. In brief, before the reveal, expected pain relief was higher for placebo than the control cream. This was not significantly different between conditioning groups.

After the reveal, the expected pain relief from the placebo varied among the long conditioning and short conditioning groups. There was some pain relief expectation in the long conditioning group, but there was none in the short conditioning group.

Expected pain relief for the control cream ratings did not change after the placebo reveal, and was no different between the short and long conditioning groups. 

How did the researchers interpret the results?

The researchers concluded their study "demonstrates a form of placebo analgesia that relies on prior conditioning rather than current expected pain relief".

This, they say, "highlights the importance of prior experience on pain relief and offers insight into the variability of placebo effects across individuals".  


This experimental study suggests reinforcing an expectation of a positive outcome – as with the long conditioning in this study – can create a placebo effect. Some pain relief seemed to be experienced, even when the placebo was finally revealed to be as inactive as the control.

In terms of any implications from these findings, there are a few points to bear in mind.

  • This was a fairly small, select group of healthy adults. In fact, they were preferentially selected from the group of volunteers as being people who experienced sufficient pain response to the heat stimulus. They are not representative of everyone, and the results could have been different in other groups.
  • This was a very experimental scenario involving a heat sensor applied to the skin. The participants knew the cause of the pain, their health was not under threat, and they were in a safe environment. This cannot be applied to real-life pain scenarios such as illness or trauma, which can obviously involve widely different forms of pain and severity, and may also involve other symptoms and emotional effects. Placebo pain relief – either applied to the skin, or taken in other forms, such as a tablet or injection – may be completely ineffective in real-life pain situations, regardless of how much the person is conditioned or manipulated to believe it will have an effect.
  • The study's results also cannot be applied to placebos used in other circumstances aside from pain relief – for example, when used in trials as an inactive comparison group to a new drug used to treat disease.   

Overall, this experimental study will be of interest in the fields of psychology and pharmacology for understanding how placebos may have effects through the expectation they will work.

If you are being troubled by chronic pain, you should contact your GP. The NHS runs pain clinics that can provide both physical and psychological advice.  

NHS Attribution