Science Watch

Diagnosing mental illness isn't like diagnosing other chronic diseases. Heart disease is identified with the help of blood tests and electrocardiograms. Diabetes is diagnosed by measuring blood glucose levels. But classifying mental illness is a more subjective endeavor. No blood test exists for depression; no X-ray can identify a child at risk of developing bipolar disorder. At least, not yet.

Thanks to new tools in genetics and neuroimaging, scientists are making progress toward deciphering details of the underlying biology of mental disorders. Yet experts disagree on how far we can push this biological model. Are mental illnesses simply physical diseases that happen to strike the brain? Or do these disorders belong to a class all their own?

Eric Kandel, MD, a Nobel Prize laureate and professor of brain science at Columbia University, believes it's all about biology. "All mental processes are brain processes, and therefore all disorders of mental functioning are biological diseases," he says. "The brain is the organ of the mind. Where else could [mental illness] be if not in the brain?"

That viewpoint is quickly gaining supporters, thanks in part to Thomas R. Insel, MD, director of the National Institute of Mental Health, who has championed a biological perspective during his tenure at the agency.

To Insel, mental illnesses are no different from heart disease, diabetes or any other chronic illness. All chronic diseases have behavioral components as well as biological components, he says. "The only difference here is that the organ of interest is the brain instead of the heart or pancreas. But the same basic principles apply."

A new toolkit

Take cardiology, Insel says. A century ago, doctors had little knowledge of the biological basis of heart disease. They could merely observe a patient's physical presentation and listen to the patient's subjective complaints. Today they can measure cholesterol levels, examine the heart's electrical impulses with EKG, and take detailed CT images of blood vessels and arteries to deliver a precise diagnosis. As a result, Insel says, mortality from heart attacks has dropped dramatically in recent decades. "In most areas of medicine, we now have a whole toolkit to help us know what's going on, from the behavioral level to the molecular level. That has really led to enormous changes in most areas of medicine," he says.

Insel believes the diagnosis and treatment of mental illness is today where cardiology was 100 years ago. And like cardiology of yesteryear, the field is poised for dramatic transformation, he says. "We are really at the cusp of a revolution in the way we think about the brain and behavior, partly because of technological breakthroughs. We're finally able to answer some of the fundamental questions."

Indeed, in recent years scientists have made many exciting discoveries about the function — and dysfunction — of the human brain. They've identified genes linked to schizophrenia and discovered that certain brain abnormalities increase a person's risk of developing post-traumatic stress disorder after a distressing event. Others have zeroed in on anomalies associated with autism, including abnormal brain growth and underconnectivity among brain regions.

Researchers have also begun to flesh out a physiological explanation for depression. Helen Mayberg, MD, a professor of psychiatry and neurology at Emory University, has been actively involved in research that singled out a region of the brain — Brodmann area 25 — that is overactive in people with depression. Mayberg describes area 25 as a "junction box" that interacts with other areas of the brain involved in mood, emotion and thinking. She has demonstrated that deep-brain stimulation of the area can alleviate symptoms in people with treatment-resistant depression (Neuron, 2005).

Maps of depression's neural circuits, Mayberg says, may eventually serve as a tool both for diagnosis and treatment. Understanding the underlying biology, she adds, could help therapists and psychopharmacologists decide which patients would benefit from more intensive therapy, and which aren't likely to improve without medication. That would be a welcome improvement, she says. "Syndromes are so nonspecific by our current criteria that the best we can do now is flip a coin. We don't do that for any other branch of medicine," she says.

Yet despite the progress and promise of her research, Mayberg isn't ready to concede that all mental illnesses will one day be described in purely biological terms. "I used to think you could localize everything, that you could explain all the variants by the biology," she says. "I think in a perfect world you could, but we don't have the tools to explain all those things because we can't control for all of the variables."

One of the biggest problems, she says, is that mental illness diagnoses are often catchall categories that include many different underlying malfunctions. Mental illnesses have always been described by their outward symptoms, both out of necessity and convenience. But just as cancer patients are a wildly diverse group marked by many different disease pathways, a depression diagnosis is likely to encompass people with many unique underlying problems. That presents challenges for defining the disease in biological terms. "Depression does have patterns," Mayberg says. "The caveat is different cohorts of patients clearly have different patterns — and likely the need for different specific interventions."

Software malfunction

When it comes to mental illness, a one-size-fits-all approach does not apply. Some diseases may be more purely physiological in nature. "Certain disorders such as schizophrenia, bipolar disorder and autism fit the biological model in a very clear-cut sense," says Richard McNally, PhD, a clinical psychologist at Harvard University and author of the 2011 book "What is Mental Illness?" In these diseases, he says, structural and functional abnormalities are evident in imaging scans or during postmortem dissection.

Yet for other conditions, such as depression or anxiety, the biological foundation is more nebulous. Often, McNally notes, mental illnesses are likely to have multiple causes, including genetic, biological and environmental factors. Of course, that's true for many chronic diseases, heart disease and diabetes included. But for mental illnesses, we're a particularly long way from understanding the interplay among those factors.

That complexity is one reason that experts such as Jerome Wakefield, PhD, DSW, a professor of social work and psychiatry at New York University, believe that too much emphasis is being placed on the biology of mental illness at this point in our understanding of the brain. Decades of effort to understand the biology of mental disorders have uncovered clues, but those clues haven't translated to improvements in diagnosis or treatment, he believes. "We've thrown tens of billions of dollars into trying to identify biomarkers and biological substrates for mental disorders," Wakefield says. "The fact is we've gotten very little out of all of that."

To be sure, Wakefield says, some psychological disorders are likely due to brain dysfunction. Others, however, may stem from a chance combination of normal personality traits. "In the unusual case where normal traits come together in a certain configuration, you may be maladapted to society," he says. "Call it a mental disorder if you want, but there's no smoking-gun malfunction in your brain."

You can think of the brain as a computer, he adds. The brain circuitry is equivalent to the hardware. But we also have the human equivalent of software. "Namely, we have mental processing of mental representations, meanings, conditioning, a whole level of processing that has to do with these psychological capacities," he says. Just as software bugs are often the cause of our computer problems, our mental motherboards can be done in by our psychological processing, even when the underlying circuitry is working as designed. "If we focus only at the brain level, we are likely to miss a lot of what's going on in mental disorders," he says.

The danger in placing too much attention on the biological is that important environmental, behavioral and social factors that contribute to mental illness may be overlooked. "By over-focusing on the biological, we are doing patients a disservice," Wakefield says. He sees a red flag in a study by Steven Marcus, PhD, and Mark Olfson, MD, that found the percentage of patients who receive psychotherapy for depression declined from 53.6 percent in 1998 to 43.1 percent in 2007, while rates of antidepressant use stayed roughly the same (Archives of General Psychiatry, 2010).

A nuanced view

The emerging area of epigenetics, meanwhile, could help provide a link between the biological and other causes of mental illness. Epigenetics research examines the ways in which environmental factors change the way genes express themselves. "Certain genes are turned on or turned off, expressed or not expressed, depending on environmental inputs," McNally says.

One of the first classic epigenetics experiments, by researchers at McGill University, found that pups of negligent rat mothers were more sensitive to stress in adulthood than pups that had been raised by doting mothers (Nature Neuroscience, 2004). The differences could be traced to epigenetic markers, chemical tags that attach to strands of DNA and, in the process, turn various genes on and off. Those tags don't just affect individuals during their lifetime, however; like DNA, epigenetic markers can be passed from generation to generation. More recently, the McGill team studied the brains of people who committed suicide, and found those who had been abused in childhood had unique patterns of epigenetic tags in their brains (Nature Neuroscience, 2009). "Stress gets under the skin, so to speak," McNally says.

In McNally's view, there's little danger that mental health professionals will forget the importance of environmental factors to the development of mental illness. "I think what's happening is not a battle between biological and non-biological approaches, but an increasingly nuanced and sophisticated appreciation for the multiple perspectives that can illuminate the etiology of these conditions," he says.

Still, translating that nuanced view to improvements in diagnosis and treatment will take time. Despite decades of research on the causes and treatments of mental illness, patients are still suffering. "Suicide rates haven't come down. The rate of prevalence for many of these disorders, if anything, has gone up, not down. That tells you that whatever we've been doing is probably not adequate," Insel says.

But, he adds, there's good reason to hold out hope. "I think, increasingly, we'll understand behavior at many levels, and one of those will be physiological," Insel says. "That may take longer to translate into new therapies and new opportunities for patients, but it's coming."

In the meantime, according to Insel and Kandel, patients themselves are clamoring for better biological descriptions of mental disorders. Describing mental illnesses as brain malfunctions helps minimize the shame often associated with them, Kandel says. "Schizophrenia is a disease like pneumonia. Seeing it as a brain disorder destigmatizes it immediately."

Certainly, Kandel adds, social and environmental factors are undeniably important to understanding mental health. "But they do not act in a vacuum," he says. "They act in the brain."

It's too soon to say whether we'll someday have a blood test for schizophrenia or a brain scanning technique that identifies depression without any doubt. But scientists and patients agree: The more we understand about our brain and behavior, the better. "We have a good beginning of understanding of the brain," says Kandel, "but boy, have we got a long way to go."


Kirsten Weir is a freelance writer in Minneapolis.