Are We Over-Diagnosing Our Children? Why Doctors Are Rethinking Mental Health Labels

Mental health diagnoses for children have skyrocketed, but a growing debate among clinicians raises an uncomfortable question: Are we helping kids understand themselves, or are we creating problems where none existed? In England alone, the number of children with active referrals to mental health services jumped from 157,000 in 2013-14 to 958,000 in 2023-24—an increase of over 500% in just one decade. That represents 8% of all children in the country. Yet experts increasingly worry that not all of these diagnoses are necessary, and some may actually do more harm than good.

Why Are Childhood Diagnoses Increasing So Rapidly?

The explosion in diagnoses reflects a complex mix of factors. Some argue it represents long-overdue attention to mental health problems that were previously ignored. Others point to the destabilizing effects of modern life—late capitalism, social media, and pandemic-related stress. But a third camp suggests something different: that shifting attitudes and policy changes have made it easier for families to seek diagnoses they may not actually need.

The problem isn't unique to mental health. Whenever we look harder for a disease, we find more cases of it. This is especially true when screening healthy people. The challenge is knowing when we've crossed the line from identifying real problems to creating unnecessary ones. Clinician and author Suzanne O'Sullivan explores this tension in her book "The Age of Diagnosis," arguing that we've begun treating risk factors and genetic predispositions as if they were actual diseases.

What Does Overdiagnosis Actually Mean?

Overdiagnosis happens when someone receives a diagnosis that, while technically accurate, doesn't benefit them and may cause harm. Consider how breast cancer screening works: among 3,000 women screened, roughly 25 to 30 will be found to have cancer. However, screening prevents untreatable disease in only 2 or 3 of these women. Most of the others would have noticed symptoms while the cancer was still treatable anyway. At least one woman will have a cancer that never would have caused her any problems—she's a victim of overdiagnosis.

The same principle applies to childhood diagnoses. A child might receive a label like autism or attention-deficit/hyperactivity disorder (ADHD) that technically fits their presentation, but the diagnosis itself—and the identity it creates—may cause more problems than it solves. The harm doesn't always come from treatment; it comes from how the diagnosis shapes a child's sense of who they are and what they're capable of.

How Can a Diagnosis Harm a Child?

O'Sullivan interviewed dozens of patients across different diagnoses and found something striking: almost everyone felt their diagnosis was positive. Cancer patients believed they received life-saving treatment. Those diagnosed with autism or ADHD felt they finally understood themselves. Yet O'Sullivan argues that in some cases, these diagnoses did more harm than good—a challenging assertion that raises uncomfortable questions about how we label children.

One powerful example involves two sisters whose mother died from Huntington's disease, a rare genetic disorder. One sister took a genetic test and confirmed she had inherited the disease. The other sister delayed testing but became convinced she was developing symptoms. Every stumble, every moment of clumsiness seemed like an early sign. When she finally took the test, she discovered she didn't have the disease at all. Some of her symptoms had different medical causes; others were products of anxiety and the nocebo effect—the dark mirror of the placebo effect, where people experience side effects from sugar pills because they expect to.

Notably, the Huntington's test has been available for 30 years, yet fewer than 20% of at-risk people have chosen to take it. The majority apparently prefer hope to certainty—a decision O'Sullivan cites as an example of an informed community recognizing that just because we can test for something doesn't mean we should.

Why Is Genetic Testing Making This Problem Worse?

When the human genome was sequenced in 2003, scientists hoped to unlock the genetic basis of inherited disease and usher in a new era of drug discovery. They already knew some diseases involved alterations in 10 to 12 genes. But the reality proved far more complicated. The most common diseases are associated with mutations scattered throughout thousands of genes across the human genome.

Here's the key insight: a core set of genes code for proteins central to disease, but because genes work through densely connected networks, a change in almost any gene can influence the behavior of the core. This means most inherited susceptibility to disease can't be attributed to a specific gene or even a small number of genes. Yet we're increasingly treating genetic risk factors as if they were diseases themselves—testing for them, labeling people based on them, and sometimes treating them before any symptoms appear.

Steps to Evaluate Whether Your Child Needs a Diagnosis

• Ask about alternatives: Before accepting a diagnosis, ask your child's doctor whether the symptoms could be explained by other factors—stress, developmental variation, environmental changes, or simple anxiety. Not every behavioral difference requires a label.
• Consider the benefits and harms: Understand both sides of the diagnosis. Will treatment actually help your child function better, or will the label itself become a self-fulfilling prophecy? Some children benefit from understanding their diagnosis; others are harmed by it.
• Seek a second opinion: Diagnostic criteria have loosened for conditions like autism and ADHD in recent years. Different clinicians may reach different conclusions. Getting a second perspective can help you make a more informed decision.
• Question genetic testing: If your doctor recommends genetic testing for risk factors, ask whether knowing the result would actually change treatment or management. If not, consider whether the test is worth the psychological burden.

What Should Parents Actually Do?

The goal isn't to dismiss all diagnoses—many children genuinely benefit from understanding why they struggle and receiving appropriate support. Rather, it's to approach diagnosis thoughtfully, recognizing that labels carry both benefits and risks. O'Sullivan's work suggests we should ask harder questions before accepting a diagnosis, particularly when it's based on genetic risk rather than actual symptoms, or when the criteria for diagnosis have recently expanded.

The surge in childhood mental health referrals reflects real suffering, but it may also reflect a cultural shift toward medicalizing normal variation and risk. As more children receive diagnoses, we should pause to ask: Are we helping them understand themselves better, or are we creating identities that limit their possibilities? The answer likely depends on the individual child, the specific diagnosis, and whether treatment actually improves their life. That's a conversation worth having before accepting any label.