How can you tell if a treatment works?

Sorry for the delay in writing this post – I had a bit of a break for finals and moving back home, but here I am with something I found interesting during revision: How can you tell if a treatment works?

At first glance, it might seem like this question can be easily answered, but it is not enough to give a group of a patients the treatment, and then see if their symptoms got better. For example, what if they would have got better anyway? Or it was something about the act of talking about their problems which caused them to feel better, not that the specific aspects of that therapy worked? I will now outline the method needed to conclude whether a psychological treatment is effective or not:

Randomised Controlled Trials (RCTS)

These are a type of experiment, known as the ‘gold standard’ for psychological experiments. The main feature is that participants are randomly assigned to different groups, for example, an experimental group, which receives the treatment, and a control group, who do not. Ideally, the patients in the control group are matched to patients in the experimental group e.g. same age, level of education etc. This is so that the effects of these other variables can be minimised, and so any difference in outcome can be attributed to the treatment. The control group is important to show that patients wouldn’t have got better anyway. For example, Mayou et al (2000) studied the effects of debriefing after trauma and found an objective drop in symptoms 3 years later. However, a control group who received no debrief had almost no symptoms 4 months after (see graph below). This shows the importance of a control group who received no intervention.

mayou et al

As well as finding out if a treatment is more beneficial than no treatment, RCTs can also be used to compare the effectiveness of different therapies for a psychological disorder. For example, Clark et al (1994) compared found cognitive therapy was the most effective treatment for panic disorder, compared to exposure therapy, applied relaxation, or imipramine (a drug treatment).


RCTs also include a follow-up some time after treatment, which enables researchers to tell if the treatment can cause long-term benefits. For example, in the Clark et al study shown above, you can see from the diagram that they carried out a follow up one year after treatment, and that patients in the cognitive therapy group still showed the largest reduction in panic symptoms.

RCTs are the method used to compare therapies, and in order to tell whether a treatment is effective, they need to feature:

  • A valid measure of symptoms at pre-treatment and post-treatment e.g. Body Sensations Interpretation Questionnaire (used by Clark et al to assess misinterpretations of body sensations in panic disorder patients).
  • Broad assessment e.g. patient and independent assessor. (Needed because patients have a tendency to report feeling better than they actually are to the person who’s been treating them).
  • Assess significant pre-treatment to post-treatment change.

It is important to tell whether a treatment works, as if it is shown to be effective, it is more likely to secure funding, and be used on patients within the NHS.

Thank you for reading, I’ll be able to get back into a routine with blogging again now my exams are over so check back soon for new posts!



Following on from the post about my dissertation.. now on to the next thing that’s been taking up quite a lot of my time this year – my research project. For this part of my degree, I had to carry out my own experiment on something which had not been shown before, and analyse the results.

I chose to look into nostalgic memories, and in particular, do we feel nostalgic from reading someone else’s nostalgic memories? It seemed like there was a bit of a gap in the research that had been done so far: although we know the functions of nostalgia (e.g. self esteem) and features of nostalgic memories (e.g. loved ones), not much work had been done on ways of making people feel nostalgic.

To start, I should probably make clear what nostalgia is: it is defined as a
“sentimental longing or wistful affection for a period in the past”

The problem I identified with existing research was that most experiments manipulated nostalgia by asking participants to write down a nostalgic or ordinary memory, reading words that describe nostalgia and writing a memory based on them, or by listening to a nostalgic or ordinary song. These methods are fine if you then want to find out about effects of nostalgia, but are they equally effective in inducing nostalgia?
Nostalgia is also a very social emotion, and a study has shown that thinking of a nostalgic memory involving an out-group member leads to fewer feelings of prejudice towards that person (Turner et al, 2012). If nostalgia is a social emotion, then I hypothesised that reading someone else’s nostalgic memories could cause you to feel nostalgic. There had only been one study which found that nostalgia could be induced by reading someone else’s old love letters, or looking at their old photos. I decided to present participants with an actual nostalgic (or ordinary) memory narrative from someone else, and told participants the memory was from someone similar or dissimilar to them in age (to see whether similarity affected results). I compared how effective this method was with a previous method of making people feel nostalgic – giving participants a list of words that described nostalgia, or were more general, and asked them to write a memory based on these features.

My hypotheses:
1. That reading the nostalgic memory would make people feel nostalgic (compared to reading an ordinary memory).
2. That participants who were told the nostalgic memory was from someone similar to them would feel more nostalgic than those who were told it was from someone different.
3. That this method would be effective, but more nostalgia would be induced from writing your own memory in the comparison condition.

The reason I thought that similarity would effect results was because of principles shown in social psychology – it has been found that when people are categorised into groups, they will automatically perceive themselves as being more similar to other in-group members, and therefore more dissimilar from out-group members (Tajfel & Turner, 1979). Therefore, the perceived similarity of the reader to the person who wrote the memory could increase the amount of nostalgia transferred to the reader of the narrative. I called this a similarity-based “transfer effect” of nostalgia.

Design of Experiment:
– 121 participants, all between the ages of 16 and 24 (so I could manipulate similarity by age)
– Participants split into 6 groups: nostalgia similar, nostalgia dissimilar, ordinary similar, ordinary dissimilar, central features (write memory) and peripheral features (write memory).
– At the top of the memory narratives, a sentence explained this was ‘an actual memory from someone aged 20 (similar condition) or 60 (dissimilar condition).
– After participants had either read the memory, or read the features and written their own memory, they then completed a questionnaire to assess how nostalgic they felt.

What I found:

Hypothesis 1: Reading someone else’s nostalgic memory did make people feel more nostalgic than those who read an ordinary memory.

Hypothesis 2: Similarity had an effect on the amount of nostalgia people felt: participants in the nostalgia similar condition felt more nostalgia than participants in the nostalgia dissimilar condition – EVEN THOUGH the memory was the same, the age of writer differed.This is shown by the graph below.


Hypothesis 3: The participants who wrote their own memory based on central features of nostalgia felt more nostalgic than participants who had read a nostalgic memory. However, there was no difference between the central and peripheral conditions, which differed from the original study (Hepper et al, 2012), who found peripheral features did not induce nostalgia. The graph below shows the results for this, and hypothesis 1: more nostalgia felt by participants who read someone else’s nostalgic memory than someone’s ordinary memory.


The results of my experiment are the first to show that people can be made to feel nostalgic by reading someone else’s nostalgic memory, and that the amount of nostalgia felt depends on how similar the writer of the memory is to the reader. Therefore, it sets the basis for more research to be done on different ways of manipulating similarity and other ways of inducing nostalgia.

Hope you found this interesting and let me know if you’ve got any questions – I know this is a really complicated experiment!