Bad Therapy Summary and key ideas

Bad Therapy, by Abigail Shrier.

Abigail Shrier begins with a question about what happens when care itself causes harm. Therapy is usually understood as help: someone is struggling, a professional listens, and treatment is meant to improve things. But a good purpose does not guarantee a good result. Shrier asks readers to judge psychological care by its effects, including effects that were never intended.

She draws a boundary around the book’s subject. Some young people have profound mental illness that can prevent them from working or maintaining relationships. They need psychiatric care, and Shrier is not arguing that they should go without it. Her focus is the larger group whose difficulties may be worry, loneliness, sadness, fear, or other ordinary and moderate distress. The question is whether those struggles call for treatment, and what happens when they enter a system built to diagnose and intervene.

Shrier uses the term iatrogenesis for harm that originates with a healer or treatment. It can follow an error or negligence, but it can also arise from ordinary risks, or from treating a problem as though it were something else. The word does not by itself accuse a clinician of malice or incompetence. It names a possible cost of intervention, even when people involved are trying to help.

That possibility matters because treatment is not the same as everyday advice. Encouraging a child to sleep well or eat vegetables does not presume that the child is deficient. An intervention, by contrast, is usually meant to correct a presumed problem or incapacity. For someone who is seriously ill, the chance of improvement may justify the risks. For someone who is generally well, the balance may be different: there may be less to gain, while the risks remain. Shrier’s standard is not to reject care, but to ask whether its expected benefit warrants its costs.

She illustrates the question with a visit to urgent care with her twelve-year-old son, who had a stomachache. The doctor had ruled out appendicitis and suspected dehydration. A nurse then sought to screen him privately for suicidal thoughts, following a written staff script. The script said to ask the questions away from parents and to notify them if safety concerns arose. Shrier says there was no sign her son had a mental illness. She wondered whether a child might try to give an adult the answer he thought was wanted, and whether an affirmative answer could affect whether she could take him home. Her concern is about applying a serious mental-health procedure without an apparent psychiatric reason, not a claim that asking about suicide is never appropriate. The encounter leaves a practical question: does routine screening in this situation protect a child, or could it introduce anxiety and confusion?

Shrier places that episode beside D.A.R.E., a school drug-prevention campaign that used group techniques shaped by therapy. Students discussed personal problems, confessed drug use, and rehearsed refusing drugs. The effort aimed to prevent substance use, but Shrier cites follow-up studies suggesting the program may have increased teen alcohol and drug use. She proposes that inviting children to discuss a problem they did not already have could sometimes introduce it. The example shows why an intervention’s purpose cannot stand in for evidence about its effects.

This scrutiny is not a claim that therapy never helps. Shrier describes her own experiences as mixed: therapy could be comforting or enlightening, and some focused treatments have shown success. She specifically acknowledges evidence that cognitive behavioral therapy can help with phobias. Nor does she portray clinicians as uniformly bad actors; therapists generally want to help. Her argument is that sincerity, protocol, and professional status cannot settle whether a particular treatment helps a particular person.

The question has special weight when the patient is a child. Shrier argues that children and adolescents may not yet have the self-knowledge or confidence to challenge an adult’s interpretation of them or their family. That makes it important to notice not only whether help is offered, but how the child is doing afterward. Immediate reassurance, a sense of being heard, or a brief feeling of relief may matter, but none alone establishes lasting improvement.

Measuring that improvement is difficult. Shrier says therapists often do not track side effects, and the profession does not require such tracking. It can also be hard to agree on what counts as harm: crying or a divorce might signal worsening distress, but could also be interpreted as part of change or progress. If outcomes remain undefined, clinicians and patients may miss cases where treatment made things worse. The difficulty of measurement is a reason for greater care in evaluating treatment, not a reason to assume either that harm is common or that it never happens.

The governing question, then, is not whether help is well meant. It is whether the benefits outweigh the risks for the person receiving it, and whether anyone is checking. Serious illness can make treatment’s risks worthwhile. Targeted therapy can bring real benefit. But when ordinary childhood distress becomes a reason to intervene, good intentions alone are not enough. Care must be judged by what it does, including the harms it may cause.