Psychotherapy: Gaps in Our Knowledge

Our understanding of psychotherapy is as fragmented and distorted as the proverbial blind men’s understanding of the elephant. For one thing, we lack reliable data about how many people practise psychotherapy, let alone how they practise it. The World Health Organisation collects survey data on the numbers of mental health practitioners, but does not ask respondents to specify whether they practice psychotherapy. Some countries, including Germany, Sweden and Australia, collect data on the number of people who practise psychotherapy within their national health care programmes, but not on those with private psychotherapy practices. Some countries collect data only on psychotherapy services that are reimbursed by insurance—which are only a small fraction of the services provided. And many countries collect no data at all. For example, in Poland, a country of 38 million people, no one has even an approximate idea of how many people are currently practising psychotherapy.

Furthermore, there is an enormous variety of practices that people define as psychotherapy. In a 2018 paper, researchers identified at least 600 different modalities. New approaches are constantly emerging, and modalities that have been discredited often re-emerge under new names. Only about a dozen approaches have been systematically tested for their effectiveness, and most of those have been tested only with respect to a small subset of psychological problems.

In many parts of the world, although psychologists and psychiatrists must be licensed, the practice of psychotherapy is completely unregulated: anyone can call herself a psychotherapist. Some branches of psychotherapy do require credentials, but the amount of training they demand varies wildly depending on the modality—from a few dozen hours of online coursework to several years of intensive training. And, after being credentialed, some therapists are subjected to ongoing supervision and quality controls; some are not.

In addition, therapists’ and patients’ reports are influenced by a host of confounding variables, making them no more reliable than those of the blind men describing the elephant. Some patients are delighted with their therapy, and some are not, but there is no way to evaluate the extent to which this is affected by the type of modality used, the quality of the therapist herself or the patient’s specific circumstances.

It is thus impossible to accurately estimate how many people have undergone (or are undergoing) psychotherapy, impossible to compile reliable statistics on patient experiences and impossible to assess the overall quality, efficacy or harmfulness of particular psychotherapy services. Personal opinions of psychotherapy are therefore necessarily based on individual encounters with the field and on the anecdotal experiences of a miniscule fraction of patients, each of whom, like the blind men in the parable, has touched only a fragment of reality. As things currently stand, our assessments of psychotherapy simply cannot be complete, and the surer people are of their assessments, the more sceptical of those assessments we should be.

There has, of course, been some objective research in this area, but it has usually raised more questions than it has resolved. Amid the flood of publications reporting on the effectiveness of individual therapeutic modalities, there have been occasional comprehensive reviews and meta-analyses—and these tend to cast doubt on the accuracy and reliability of the individual studies. For example, a 2017 meta-analysis looked at over 5,000 studies on the effectiveness of psychotherapy, selected because they met the highest methodological standards. While most of the study reports spoke favourably of psychotherapy, only 7% of them contained results that confirmed the effectiveness of the modality or modalities studied.

A meta-analysis published in the April 2021 issue of Nature—which looked at over 400 studies with a combined total of more than 50,000 participants—found some evidence of the partial efficacy of some interventions, particularly those based on mindfulness, positive psychology, cognitive behavioural therapy, acceptance and commitment therapy and reminiscence therapy—but the researchers caution that both the effect sizes and the quality of the evidence were generally only low to moderate. So, this is the current state of play—in a field that was established over a hundred years ago.

This is a castle built on sand. Why is this not more widely recognised? One reason may be conflicts of interest. If a researcher who is employed by a pharmaceutical company publishes research on the efficacy of that company’s drugs, she must add a disclaimer to the article. But in research on psychotherapy, the researchers are often psychotherapists who work in the modality whose effectiveness they are researching. A 2015 study of over 900 papers on the effectiveness of psychotherapy—published between 2010 and 2013—found that, although this was the case for many of the studies involved, few of them contained a declaration of a conflict of interest.

In addition, as several independent analyses have confirmed, therapists may advertise themselves as using modalities that are known to be relatively effective, such as cognitive behavioural therapy, but, in practice, employ quite different approaches. And even when therapists use the recommended modalities, research shows that they may depart from validated protocols. The gap between the treatments that research has shown to be relatively effective and the treatments patients actually receive appears to be growing.

Even more troublingly, many psychotherapy researchers neglect to publish or pay attention to negative findings. For example, of all the research papers on the effectiveness of psychotherapy published in 2010 found that only 3% contained an exhaustive description of the negative effects of psychotherapy, 18% contained only rudimentary information on the subject and the remaining 79% contained no information on possible negatives at all. If nearly four out of five papers on the effectiveness of psychotherapy don’t include any data on instances in which it was harmful, it is difficult to believe that they are balanced presentations of the situation.

A further obstacle to assessing the effectiveness of psychotherapy is that so many patients discontinue therapy prematurely. A 1993 meta-analysis found that nearly half (47%) of all patients ended their therapy early. While more recent studies have shown a slight reduction in the rates of patients discontinuing therapy, they still range from around 18% for patients suffering from dissociative disorders to nearly 30% for patients suffering from schizophrenia and psychotic disorders. We don’t have data on what is happening to these patients. Do they go to other therapists; does their condition deteriorate; or do they find other, better ways to cope with their problems?

As I’ve detailed elsewhere, I have spoken to many people who blame psychotherapy for damaged health, wasted years of life and broken family relationships. I believe they deserve special attention, because they have paid the highest price for participating in the experiments that people have decided to call psychotherapy. Perhaps they simply had the bad luck to work with incompetent therapists; perhaps they were simply expecting too much. However, we will not discover what went wrong without listening to their stories.

One of the most significant issues such people raise is the imbalance of power between therapist and patient. Patients seeking therapy are often depressed or distressed. They have often already tried many other ways of resolving their problems. They often feel profoundly helpless and have lowered or even shattered self-esteem. Then they encounter someone who specialises in solving their problems, and who may have impressive credentials—so they try to conform to that therapist’s expectations. As a result, therapists often end up wielding a great deal of power—power some are reluctant to ever relinquish. Therapists may persuade their patients to meet them weekly for years, without any concrete plans as to how to help them, while patients remain convinced that they are moving towards a goal, some kind of catharsis, after which everything will change for the better.

Unfortunately, as former psychoanalyst Jeffrey Masson has written, things may look completely different from the therapist’s perspective:

Many times I sat behind a patient in analysis and became acutely and painfully aware of my inability to help. Many times, indeed, I did feel compassion. But at times I also felt bored, uninterested, irritated, helpless, confused, ignorant, and lost. At times I could offer no genuine assistance, yet rarely did I acknowledge this to the patient. My life was in no better shape than that of my patients. Any advice I might have had to offer would be no better than that of a well-informed friend (and considerably more expensive). I must assume that none of this was unique to me. Everything I experienced in the situation must have been felt by other therapists as well.

Like religion, therapy promises that we can become better versions of ourselves. The more time and energy someone invests in therapy, the more difficult it becomes to persuade him that he is chasing a mirage. A psychotherapist can easily assume the role of a priest: telling us what is good for us and which path in life to follow. The unsubstantiated, dogmatic assumptions of some therapeutic schools seem like replacements for religious doctrine—in which the object of worship may be self-realization, self-development, self-fulfilment, the here and now, work-life balance or some other currently fashionable shibboleth.

Cigarettes and alcohol come with warning labels. But no one warns patients about the risks of psychotherapy.

Many therapists abuse their power over patients—and they often do so with impunity. Patient complaints—even criminal complaints—are often dismissed with the circular argument that the accuser must be deluded because otherwise he or she would not have been in treatment.

Some sources have estimated that therapists may win up to 80% of liability cases. If the therapist has not committed a criminal offence during the therapy session, there to plausibly assign responsibility for what happens to the patient outside the therapy room. The most severe punishment imposed is often merely expulsion from a professional association, or a professional reprimand. But even these penalties are rarely used and do not have any practical consequences. The American Psychological Association reports that only about 2% of psychologists face malpractice suits. This either means that 98% of therapists are trustworthy—or that many abusive therapists are never brought to justice. Following a malpractice suit, the therapist is usually free to continue practising.

And yet in 2012, the council of the largest and most influential association of psychologists in the world—the American Psychological Association, which has nearly 120,000 members—voted to adopt a resolution proclaiming the effectiveness of psychotherapy, which is now cited by most similar associations around the world. This seems like hubris, to say the least.