An Introduction to CBT Research

Sarah Rakovshik

• Publisher: SAGE Publications Ltd
• Year: 2019
• ISBN:
  9781529700541
• Edition: 1
• Topic: Psicologia
• Subtopic: Psicologia cognitiva e cognizione

This highly practical book will guide students through the different levels of research within CBT by addressing the general principles, how to understand statistics and tackling with general principles. It also highlights how to critically engage with, interpret and evaluate research so that it can be used to shape practice. This important book will help readers see the relevance of research in their working lives and empower them to become active and keen researchers.

Part I Why does research matter in CBT?

1 What is research and why do we bother?

Case study

Hannah closes the door behind her last patient of the day and wonders if she is doing something wrong. She has been trained in CBT and endeavours to provide evidence-based treatment to the patients in her service. Some seem to respond well, and she is, at times, surprised herself at how quickly someone who has been nearly housebound for years for fear of having a panic attack in public can return to a normal life. With others, she feels it is three steps forward and two steps back. It seems to Hannah that many of these patients have a combination of symptoms and difficulties that don’t neatly fit into one diagnostic category. She wonders if having comorbid disorders makes it harder to sustain progress in therapy.

Hannah’s service can only offer up to 8 sessions, maybe 12 at a stretch, and she increasingly feels that therapy is ending with these patients before they have really been able to shift key beliefs and ways of coping. Is it because she is a bad therapist? She knows that since these therapies have been tested in RCTs, they have a very good chance of working. The reason she chose to train as a CBT therapist is that she really liked the idea of offering an evidence-based treatment. She does wonder if the trials the protocols are based on had the same kind of patients as she sees in her service. She was trained to follow the evidence base for each particular disorder, but she’s not sure how she should proceed when the presentation isn’t diagnostically so clear-cut. She’s had several supervisors who gave her different advice: one said to prioritise the problem that was most important to the patient, another said to look for commonalities, and the third emphasised the need to formulate each patient individually.

Hannah isn’t sure where to start looking for the answers to these questions and knows that she doesn’t really know enough about research to evaluate the quality of a study, even if she does find a relevant paper. Is she actually offering evidence-based therapy to these patients and is there any research that can tell her the most effective way to approach treatment when a patient experiences psychological difficulties characteristic of more than one disorder?

One of the central themes of this book is that there is – or should be – a natural ‘fit’ between cognitive behaviour therapy as a model of human psychology/psychotherapy and the general scientific approach known as ‘empiricism’ (roughly speaking, the idea that scientific theories should be tested through gathering observations and evidence – i.e., through research). CBT has always tried to be scientific in this sense, both as an approach to therapy and as a model of how mental health problems arise and are maintained.

How is a CBT therapist like a researcher?

Case study

Lela has been told her whole life that she is not good enough. Not only did both her parents point out her mistakes and imperfections, but some of her teachers also commented that she wasn’t living up to her potential. She has mostly avoided being close to other people because she is sure that if they get to know her, they’ll see how deeply flawed she is and won’t want to keep up the relationship. Now, she has just been made redundant and comes to you convinced that all the facts in her life demonstrate what she has always known: she is useless. The lack of motivation she currently feels just confirms this conclusion. When you ask if this belief may understandably leave her feeling that it is futile to keep on trying, she insists that this is not a thought, it is a fact, and she has a lifetime of evidence to prove it.

Do you agree with Lela? She has evidence and some good reasons for her opinions – does this make them facts? How might you find out together?

As a therapy, CBT reminds of mottos such as ‘Thoughts are not facts’, ‘Don’t believe everything you think’ and similar calls to adopt a sceptical approach. We ask patients to consider that thoughts, like opinions, may be accurate or helpful, but they may also be biased and unconstructive; we ask them to use their reason to think through alternative explanations of what happens to them; we use tools such as behavioural experiments to help them find out whether their ideas are accurate or helpful. All of these are essentially scientific, empirical procedures, aimed at gathering evidence that will support or refute our and our patients’ ideas – in effect, we ask our patients to conduct research. Indeed, Beck and colleagues coined the phrase ‘collaborative empiricism’ to describe the basic working style of cognitive therapy (Beck et al., 1979).

Why does CBT care about research?

As a general set of theories and models of therapy, CBT has also always had a commitment to a scientific approach. Behaviour therapy, cognitive therapy and the hybrid we nowadays know as CBT, have all consistently taken the view that it is not enough to come up with a plausible theory or a treatment approach that sounds convincing. We want to gather evidence that will help us decide whether the theories are true, whether the therapy is actually effective, and so on. Again, we need research to answer these questions.

Of course, none of this means that CBT is perfectly consistent in its commitment to these principles. Like all human communities, we sometimes respond more to what is fashionable or exciting than to what has the best evidence (‘I went to a great workshop on this technique last week, so now I’ll suggest it to all my patients’); we can still take a biased view of the evidence (‘anything that supports my pet theory is obviously first-class research; anything that opposes it was clearly done by people who are incompetent, if not fraudulent’); and we still sometimes respond to appeals to authority (‘this must be true just because Beck says it is’). But despite all these numerous imperfections, there is at least a commitment to try to do better: to try to pay serious attention to the state of the evidence, rather than just what we would like to believe or what feels right.

What is research and why do we need it?

There is no single, universally accepted definition of research, but most centre on the idea of trying to gain knowledge through systematic investigation. For the purposes of this book, I will use the following as a working definition of what we are talking about.

Research is the attempt to answer some question (practical or theoretical) based on systematically gathering evidence of some kind (not just personal opinion, intuition or guesswork), so that (a) there are some reasonable grounds for us and others to believe the conclusions; and/or (b) there are some reasonable grounds not to believe alternative explanations.

The idea of ‘evidence’, which gives us good grounds to believe a conclusion, is central here. But why do we need it? One of the fundamental ideas behind most scientific approaches to research is that human judgement is inherently fallible. We jump to conclusions, we make mistakes, we see patterns that are not really there – in an endless variety of ways, we can get things wrong. If we want to attain reliable knowledge, therefore, we need to have ways of gathering information that minimise the risk of making such mistakes – and that is what research is for.

The need for scepticism

Case study

Zaid has been experiencing episodes during which he finds it hard to breathe, feels dizzy and can feel his heart racing. Sometimes he also feels a tingling sensation in his arms and hands. Understandably, Zaid believes that he may be experiencing warning signs of a heart attack and therefore he has stopped any kind of physical exertion that might bring on these symptoms. He is scared and frustrated because the more he restricts his activity, the more often these symptoms seem to appear.

The GP has assured him that there is nothing wrong with his heart, but surely it isn’t unreasonable to think that a busy GP might be missing something serious?

What would you say? Does Zaid have evidence for his belief? Does he have a theory that explains his symptoms? How do you go about assessing his evidence and the conclusions he has drawn together? How does the theory that Zaid holds about his experiences matter in how he will react to them?

A crucial aspect of the collaborative empiricism we use in testing out patients’ beliefs is scepticism: a refusal to believe any theory (even our own) until it has been thoroughly tested. As clinicians, we know that there can be different explanations for the same set of facts or data, and that this matters tremendously in whether our reactions are helpful or not. The same is true as we think about conducting research to test hypotheses and theories.

Let’s illustrate the need for scepticism and research with a couple of examples.

Clever Hans

‘Clever Hans’ was a performing horse that astounded audiences around the beginning of the 20th century. When his owner, Mr von Osten, posed various questions to Hans, such as mathematical calculations, Hans would ‘answer’ by tapping his hoof on the ground an appropriate number of times. Although not perfect, he had an astonishingly high success rate, which von Osten took as proof of his theory that animals were as intelligent as humans. Of course, some people were sceptical and suspected von Osten of cheating in some way, but initial investigations could not see any evidence of cheating and ended up endorsing Hans’s abilities.

Others, however, remained sceptical, and eventually a group led by Oskar Pfungst retested Hans and finally arrived at the true explanation. They found no deliberate cheating, but they identified two crucial factors that determined Hans’s performance.

1. He could only answer when he could see the questioner.
2. He could only answer questions to which the questioner already knew the answer.

Their conclusion (Pfungst, 1911) was that von Osten was not deliberately cheating, but that he was unconsciously cuing Hans’s responses through tiny behavioural changes. For example, if the answer to the question posed was ‘6’, von Osten would tend to lean forward expectantly when the horse’s taps reached 5. Without those visual cues, Hans could not perform. Hans was indeed clever, but he was clever at picking up small visual cues, not at mathematics. This process of unconscious cuing by the experimenter is still sometimes called the ‘Clever Hans effect’ and needs to be taken into account as a potential bias in many psychology experiments. It is one of the reasons for doing ‘blind’ or ‘double blind’ experiments (Chapter 4).

So, we shouldn’t trust a counting horse, but surely our clinical intuition mostly points us in the right direction?

Critical Incident Stress Debriefing

Critical Incident Stress Debriefing (CISD) was originally described by Mitchell (1983). It was developed as a way of helping people involved in traumatic incidents to avoid long-lasting traumatic reactions. It was supposed to be delivered soon after the incident occurred, often in natural groups (e.g., a fire-fighting team involved in a traumatic incident). It was not seen as therapy for specific mental health problems, but rather as a way of helping people with normal stress reactions to cope better. CISD was typically delivered in a single session and included participants (a) describing the events; (b) going through their thoughts and emotions with the aim of processing them; and (c) receiving help to find good coping strategies. This and similar approaches, often collectively known as ‘psychological debriefing’, became very widely used – perhaps most famously after the 9/11 attacks on New York.

From a theoretical perspective, CISD seemed to make sense. One of the most well-validated CBT models of PTSD (Ehlers et al., 2003) indicates that avoidance of thinking about a very difficult and distressing experience plays a powerful role in maintaining symptoms of psychological trauma. Neurological findings support the hypothesis that avoidance of the memory (and often anything that would trigger it) impedes processing of the memory from short-term storage in the amygdala to long-term storage in the upper cortices, where it would receive a ‘time stamp’ telling the individual that ‘that was then and not now’ when the memory is retrieved. This distinction is thought to be the difference between remembering a difficult experience, but knowing that it was in the past, and the kind of ‘it’s happening now; it is happening again’ experience typical during a traumatic flashbac...