Chapter 2

What Causes Mental Illness and Why Does It Matter?

Insanity, madness, anxiety, irrational fear, unrelenting depression, addiction, suicide: Mental illness has been described in every human culture on Earth, stretching back from the present to antiquity. While we have seen that it is on the rise, it is far from a new affliction. And yet the question of what causes it continues to perplex us. Ancient scholars, philosophers, and poets, along with modern-day neuroscientists, physicians, and psychologists, have studied this question relentlessly, and come up with no definitive answer.

Many theories have been proposed over the past few millennia. In ancient times, mental illnesses were largely thought to be the result of supernatural forces. Punishment from God was a common belief. Demonic possession also had its heyday, with exorcism being the treatment of choice. While these kinds of views have persisted and resurfaced throughout history, a more scientific attitude emerged almost as soon as illness in general began to be looked at in a natural rather than supernatural light, and the conception of mental illness as a medical disorder was born. The ancient Greek physician Hippocrates was among those who took mental illnesses seriously; he postulated that they might be due to an imbalance of the body’s four vital fluids, or “humors.” An excess of one of these, black bile, was thought to cause depression—or melancholia; in fact the word “melancholy” comes from the Greek for “black bile.” (Interestingly, bodily substances—particularly feces, as it relates to the gut microbiome—are making a comeback in the theory of mental illness. More on that later.) Just as the birth of medicine transformed the way people thought of mental disorders, so, naturally, did the development of the field of psychology. Sigmund Freud famously theorized that mental disorders were due to unconscious desires or conflicts, and he framed the working of the mind in terms of nonphysical entities or forces—the id, the ego, and the superego. Other psychological theories have since been developed, many attempting to explain mental illness more “scientifically” based upon what we know about behavior and neuroscience. Modern cognitive or behavioral theories, for instance, might view an anxiety disorder as the result of internalized patterns of thought, or advocate changing certain behaviors as a way to change mental experiences. While psychological theories are still used in treatment today, most clinicians and researchers do not believe they can explain all mental disorders. From the mid-nineteenth century to the present day, there has been increasing evidence that mental illnesses have at least some biological components or influences. Chemical imbalances, brain changes, hormones, inflammation, and immune system problems are all believed to have possible roles in causing mental illness. Nonetheless, some authorities in the field feel that a physical model of mental states is too “reductionistic.” They say it reduces the complexity of human behavior, emotion, and experience to chemistry or biology, and that human experience cannot be explained by our mere molecules.

In 1977, Dr. George Engel, an internist and psychiatrist, developed a working model of what causes mental illness that is still widely used today. He called it the biopsychosocial model.¹ It asserts that there are (1) biological factors, including genes and hormones; (2) psychological factors, such as upbringing and rigid beliefs; and (3) social factors, like poverty or a lack of friends, that all come together for any given individual to produce a mental illness. Another popular model is the diathesis–stress model. “Diathesis” means a biological predisposition to becoming ill, something like genetics or a hormonal imbalance. The stress in this model can be anything in the environment, such as getting fired from a job, drug use, or even an infection, that then pushes the already predisposed person to actually become ill. This model assumes that most people who develop mental disorders were likely to do so at some point or another in their lives—they were just waiting to be triggered. Both of these models present a picture of mental illness that attempts to account for many different factors all contributing to the development of a mental disorder.

Because, in fact, we have identified many factors that make people more likely to develop various mental disorders. And today, when we think about what causes mental illness, we often think in terms of these risk factors. They include things like stress, drug and alcohol use, hormonal issues, and a family history of mental illness, among others. The problem is that although we know of many such risk factors, not one of them is present in everyone with a specific disorder, and not one of them is sufficient, in and of itself, to cause any specific disorder.

An obvious example is PTSD. This is the disorder that causes people to experience fear, flashbacks, excessive anxiety, and feelings of numbness for months or years after living through a traumatic event. By definition, anyone with PTSD must have been exposed to a traumatic event, but only about 15 percent of people who experience such trauma end up developing PTSD. Even when two people experience the same traumatic event, one person can end up with severe PTSD while the other is spared entirely. In other words, trauma, on its own, does not “cause” PTSD. OK, you say, that’s because it results from a combination of risk factors. Unfortunately, there is no combination of risk factors “guaranteed” to result in PTSD, either. And so it goes with almost every other mental disorder. Sometimes, it seems easy to understand why someone develops mental illness. For example, a woman who had a horrible, abusive childhood, who has a thyroid disorder, and who was just left by her husband of ten years for another woman might develop clinical depression. Most people can understand why she might be depressed because she has many risk factors for developing clinical depression. However, there are other people for whom mental illness seems to come from out of the blue for no reason at all.

Decoding Depression

Let’s take a look at one of the most clearly defined and understood mental disorders—major depression. Everyone gets depressed, but not everyone gets major depression. People with major depression feel sad or depressed most of the time, and they may experience fatigue, problems concentrating, and disrupted sleep. This disorder can rob people of all ability to experience pleasure and enjoyment from life, and it may leave them with overwhelming feelings of hopelessness and even thoughts of suicide. There are nine symptoms in all, and to be diagnosed with major depression, a person must experience at least five of them for at least two weeks.

There are many clearly established risk factors for the development of major depression. They include genetics or a family history of depression, stress, the death of a loved one, the breakup of a relationship, conflict at work or school, and physical and sexual abuse. Various hormonal problems also make the list, including low thyroid hormone, high levels of cortisol, and the fluctuations in women’s hormones that can contribute to the risk of depression in the postpartum period or around the time of menstruation. In fact, just being a woman doubles the risk of depression compared to being a man. Excessive drug or alcohol use is a risk factor, and even some less obvious prescription medications, like certain antibiotics or blood pressure medications, can increase risk as well. Then there are social issues, like being bullied or teased, having no friends, or simply feeling lonely most of the time. Poverty, malnutrition, and unsafe living environments also increase risk. Sleep disturbance is a big one: either too much or too little sleep puts people at risk for depression. Many physical illnesses make the list of risk chronic pain, diabetes, heart disease, and rheumatoid arthritis. Cancer is another risk factor—but not necessarily the way you might think. Most people would be stressed by a diagnosis of cancer and assume it’s only natural that such a devastating diagnosis might make someone depressed. That does happen for some people. However, some patients get clinically depressed before they even know they have cancer. This is a common occurrence with pancreatic cancer in particular—people find themselves depressed for seemingly no good reason, and then a few months later are diagnosed with pancreatic cancer. Virtually all neurological disorders are associated with higher rates of depression, including strokes, multiple sclerosis, Parkinson’s disease, Alzheimer’s disease, and epilepsy. And interestingly, every other psychiatric disorder puts people at much higher risk for also developing major depression on top of their already existing disorders.

That’s . . . a lot of risk factors. And they vary widely. Not just in the kinds of things they are—biological, psychological, and social—but in how big of a role they are believed to play. For instance, while being a woman is a risk factor for major depression, no one would say that being a woman causes major depression. But there are certain factors that contribute more directly, and in fact many competing theories hold one factor or another to be the root cause of the disorder. Unlike those who subscribe to the biopsychosocial model, some professionals believe that major depression has a purely genetic, purely biological, or purely psychological origin—the rest of the risk factors are just window dressing.

One of the most widely known of these single-cause theories is the chemical imbalance theory of depression—and in fact, of mental illness. This theory suggests an imbalance in brain chemicals called neurotransmitters is the cause of all mental illness. Neurotransmitters are chemicals that relay signals between brain cells. For depression, the most popular belief is that levels of the neurotransmitter serotonin are too low; thus, medications that increase serotonin levels will treat depression. Many of the most commonly prescribed medications for depression belong to a class of antidepressants called selective serotonin reuptake inhibitors, or SSRIs (Prozac, Zoloft, and Paxil are in this category). They often do help resolve the symptoms of depression, which supports the theory that maybe a chemical imbalance is the cause. Other classes of medications that affect different neurotransmitter systems can also alleviate depression, so maybe it’s not just serotonin—maybe it’s various neurotransmitters in different people. Nonetheless, many psychiatrists and researchers believe that at its root, depression always boils down to a chemical imbalance.

However, there are many questions raised by this theory:

These questions are begging for answers, not just in relation to the diagnosis of depression, but for all psychiatric diagnoses. Unfortunately, the chemical imbalance theory doesn’t provide answers.

Another widely known theory for the cause of major depression is the learned helplessness theory. In a nutshell, it holds that when people are unable to change adverse circumstances in their lives, they “learn” that they are helpless. This can apply to something like not being able to find a romantic relationship despite numerous attempts, or more direly, an abused child trying to get his father to stop hitting him. In either case, these people begin to feel powerless, and then they get depressed. Eventually they stop trying to do much at all. Why bother? Some experts assert that the cause of these people’s depression is their psychology. They have learned, and believe, that they are helpless. Obviously, getting the abused child out of that environment is of paramount importance. But even years later, that boy may still be depressed. The treatment is often based on cognitive behavioral therapy (CBT), a form of talk therapy that focuses on identifying and changing thoughts, emotions, and behaviors. This therapy is based on the belief that when people are clinically depressed, it is likely because of thoughts that are based not so much on the reality of their current situation as on the helpless mindset developed in the past. The goal is to empower patients to challenge these thoughts and replace them with ones not so dire and hopeless. This will help them feel better and make changes in their lives, which will make them feel less helpless still, and this cycle will reinforce itself. This treatment works, at least for some people, which again supports the theory that this kind of problem might be a cause of depression.

There are many other theories about specific factors believed to be the cause of major depression—biological, psychological, and social. Many have led to the development of specific treatments and interventions that work with real people, at least some of the time. In fact, the theories themselves are often shaped by treatments that are effective for depression, using the logic that if a treatment works, even in some people, then it must be correcting a problem that was causing the illness.

Medications used to treat major depression include those specifically known as “antidepressants,” which are commonly sorted into five different classes. These classes act on different neurotransmitters and receptors, including ones for serotonin, dopamine, and norepinephrine. However, antidepressants aren’t the only medications used to treat major depression. Others include anxiety medications, mood stabilizers, antipsychotics, stimulants, antiepileptic medications, hormones, vitamins, and a wide variety of supplements, such as St. John’s wort. These all work in very different ways, and yet all are used routinely in treating depression, and all have been shown to work for at least some people, some of the time.

Psychotherapy to treat depression also comes in many varieties. Some focus on relationships, others on thoughts and feelings, and others on behaviors; some focus solely on changes in the present, and others on revisiting your past or childhood. Different types of psychotherapy can be very different from each other, yet there is at least some evidence that they can all be helpful to at least some people with depression.

Finally, there are more aggressive treatments, like TMS, ECT, and even surgery, in which parts of the brain are severed or electrodes are implanted to stimulate the brain or the vagus nerve, the main nerve of the parasympathetic nervous system.

That’s a lot of different treatments! It’s difficult to understand how they can all treat the same set of symptoms. However, not one of these works for all people with depression. Why not? Are there just different causes of major depression in different people that require different treatments? And sadly, as I reviewed in the last chapter, there are millions of people who try treatment after treatment without finding even one that works.

On the flip side, it’s important to point out that not everyone who develops major depression even gets treatment—in fact, the majority of sufferers throughout the world do not. Yet major depression will often resolve on its own. The symptoms can come and go, sometimes lasting a few weeks or months and then spontaneously disappearing. What causes some people’s symptoms to go away without any treatment? Why, for others, does depression become a chronic and debilitating illness? If we truly understand what causes this disorder, we should be able to answer these questions.

But the picture gets more complicated still. In addition to risk factors or theories about what causes major depression, we have good evidence of physical changes in the body that are associated with major depression—that is, they are found more often in those with the disease than those without it. These are changes that have been observed in people who already have the diagnosis, but they may also provide clues about the cause of the disorder.

Inflammation is a big one. We know that when compared to those without depression, people with chronic depression, on average, have higher levels of inflammation, as measured by different biomarkers, such as C-reactive protein and interleukins.² At this point, however, we don’t know for sure if the inflammation is causing the depression or if the depression is causing the inflammation. And if the inflammation is causing the depression, what starts the inflammation in the first place? Is it one or more of the risk factors that we’ve discussed so far? Or is it something else altogether that we just haven’t discovered yet? As usual, many people have theories—some speculate that it is a chronic infection, or an autoimmune disease, or exposure to a toxin, or a bad diet, or having a “leaky gut,” and on and on—but these theories are not answers. What’s more, not everyone with chronic depression has higher levels of inflammation, at least not that we can measure. The research showing higher levels of inflammation is based on comparisons between groups of people: When looking at a group of people with depression and a group without, the group with depression has more inflammation . . . but not every individual in the depressed group will have higher levels of inflammation than the individuals in the group without depression. In fact, researchers and clinicians have not identified any parameter for measuring inflammation in the body or brain that will consistently separate people who are suffering from depression from those who aren’t.

In addition to differences in inflammation levels, we have identified differences in the brains of people with chronic depression. Some people with depression have shrinkage, or atrophy, of specific brain regions, and this can progress over time. Because these types of changes are often seen in neuro-degenerative disorders, some researchers speculate that depression might be a neurodegenerative disorder as well, or that it could represent the early stages of another neurodegenerative disorder such as Alzheimer’s or Parkinson’s.³ Other researchers speculate that these changes might be the result of the increased inflammation associated with depression. We know that inflammation over a prolonged period of time can cause damage to tissue. For example, when a person’s knee is inflamed by arthritis, we know that permanent damage can result; the longer the inflammation lasts, the more the damage progresses. Maybe something like that is happening in the brain—the inflammation comes first and causes damage to these brain regions.

Research has also found a number of differences in the way the brains of depressed people function. When comparing functional MRI scans of those with major depression to those without, depressed people seem to have decreased activity in some brain regions and increased activity in others, as well as differences in the way brain regions communicate with each other.⁴ However, as with all of these brain changes we’ve discussed, the studies have only shown relative differences between groups. And once again, we don’t know if these changes are the cause of depression or a consequence of it. Could another process be causing both the depression and these brain changes? We just don’t know yet.

Finally, let’s throw another wrench into the works—the gut microbiome. The human digestive system contains trillions of microorganisms, including bacteria, viruses, and fungi. They produce hormones, neurotransmitters, and inflammatory molecules that get released into our gut and then absorbed into our bloodstreams. Research suggests that these microbes play a role in obesity, diabetes, cardiovascular disease, depression, anxiety, autism, and even schizophrenia.⁵ But microbiome research is relatively new, and we don’t yet know the details of which specific microorganisms might be beneficial and which ones harmful, or in fact if it’s about the mere presence or absence of certain organisms at all; it could also be that the key lies in the balance of different types of organisms. More to the point, although some research in mice has found changes in depressive symptoms mediated through changes in the gut microbiome, we don’t yet know how to use this information to effectively treat depression or most other disorders.⁶

So, that’s a whirlwind tour through major depression—many of the risk factors, an overview of some theories about what causes it and the treatments aimed at those causes, and some of the biological and brain changes that are seen in people with the disorder. So, given all that, how do we answer the question: “What causes major depression?”

This is why the biopsychosocial model makes sense—there are biological things, psychological things, and social factors that may come together differently in different people to result in major depression. Put another way, there are different causes in different people. Some researchers and clinicians claim that there must be different types of depression—maybe one type that’s caused by social stressors and another type that’s caused by biological factors. Maybe there are dozens of different types of depression—all caused by these various risk factors. Maybe certain factors are responsible for certain symptoms, and if we did a better job of clustering these symptoms, we could identify these types and get a better handle on these causes. Unfortunately, this doesn’t appear to be the answer. Clinicians and researchers have struggled with this for decades, and the same set of symptoms continues to come up over and over again across categories, regardless of the risk factors or perceived causes of the depression, whether they are biological, psychological, social, or some combination. The same constellation of symptoms has been found in countless people, in countless varieties of circumstance. In fact, the symptoms of major depression have been described in the Bible, historic texts, literature, poetry, and medical records dating back to Hippocrates. So what causes it? There must be an answer—one that ties together all the facts about the different risk factors, the treatments that work, and the brain and body changes that we see time and again.

Is it possible that there are different processes leading to the same set of symptoms in different people, completely independent of each other? Well, it’s possible, but it would be highly unlikely. You may have heard of Occam’s razor—it’s a general rule or guideline, also known as the law of parsimony. It is usually held to mean that the simplest, most unifying explanation is most likely to be correct. For example, all things being equal, if a patient comes in with a high fever, sore neck, and headache, it is less likely that the patient has a headache due to a brain hemorrhage and a sore neck due to a pinched nerve and a fever due to an infection than it is that the patient has meningitis—one diagnosis that explains all three signs and symptoms. In short, when faced with a situation like the one we have outlined for major depression, a unifying theory that can connect all the evidence in a logical and plausible way is most likely to be correct. Before we get too far down the road toward the answer, though, it is worth considering why it matters in the first place.

Why the Cause Matters—Treating Symptoms Versus Disorders

When diagnosing someone with a disorder, we rely on signs and symptoms. People often use the term “symptoms” as a catchall, but the difference between signs and symptoms is a crucial one. Signs are objective indicators of an illness that can be observed or measured by someone else. Signs can include things such as a seizure, a blood pressure measurement, a laboratory value, or an abnormality seen on a brain scan. Symptoms are subjective experiences that a patient must tell someone about. Symptoms can include things such as moods, thoughts, or experiences of pain or numbness. There are very few signs in psychiatry. Instead, most of our diagnoses are based on symptoms, such as irritability, anxiety, fear, depression, abnormal thoughts or perceptions, and impaired memory. Mental disorders can also include things that seem more “physical” than “mental,” like sleep disturbances, slowed movements, fatigue, and hyperactivity. Some of these can be observed, but clinicians often rely on patients to tell them about these, too, putting them into the category of symptoms as opposed to signs. Unfortunately, there are no laboratory tests, brain scans, or other objective tests that can accurately diagnose any mental disorder.

Psychiatric diagnoses are all based on the concept of syndromes. A syndrome is a cluster of signs and symptoms that commonly occur together, with a cause that is not yet known. One medical example that began in the 1980s was the syndrome of unusual infections and rare cancers that we called AIDS—acquired immunodeficiency syndrome. Before we knew that it was due to a virus, it was a syndrome. In psychiatry, every diagnosis is a syndrome. This is inherent to the definition of a psychiatric disorder. When mental symptoms are caused by a medical or neurological condition, that alone excludes classifying them as a psychiatric disorder. Neurological illnesses, cancers, infections, and autoimmune diseases can all affect the brain. When people with these conditions have mental symptoms, they are not necessarily diagnosed with a psychiatric disorder. If a patient comes in suffering from irritability, depression, and memory loss, and further evaluation reveals that these symptoms are the result of an infection or cancer, they are diagnosed with that condition and treated by a medical specialist outside of psychiatry, even if their mental symptoms are indistinguishable from those of a patient who “just” has depression. Psychiatrists and other mental health professionals are left with everyone else—the ones for whom we don’t know the exact cause.

This is at the heart of the difficulty we’ve had in making progress in mental health care. Without a clear cause, we end up treating symptoms as opposed to disorders.

Some treatments are designed to attack the root cause of an illness. The best example is an infectious disease. A bacterial infection can cause many signs and symptoms—fever, changes in blood cell counts, chills, pain, cough, and fatigue, to name a few. Definitive treatment for the infection is an antibiotic that eliminates the bacteria from the body. This type of treatment is sometimes referred to as a disease-modifying treatment. In this case, the treatment will cure the illness; after the course of antibiotics, the person will no longer have the infection. But there is another type of treatment commonly used in the medical field; treatments in this category are known as symptomatic treatments. They are designed to reduce symptoms, which can help people feel better, but they don’t directly change the course of the illness. For example, people with bacterial infections are commonly given symptomatic treatments like Tylenol to reduce fever. Symptomatic treatments can reduce suffering and allow people to work and function normally, but they are not addressing the root cause. In the end, with or without Tylenol, either the body will fight off the infection on its own, the person will get antibiotic treatment, or the infection will progress and the person will die. Tylenol won’t really make much of a difference in which of these outcomes occur.

In the mental health field, the reality is that most of our treatments are symptomatic. For most people, psychiatric medications, ECT, and TMS are usually symptomatic treatments. They don’t appear to address the root cause of the illness. For some, they can markedly reduce symptoms. In others, they can put the illness into remission, meaning that all the symptoms get completely better. There are people, like John, who can use antidepressants or other medications for a year or two and then live happily ever after without them. Does this mean that the medications were disease modifying? In some cases, like John’s, it’s possible they are. However, given the extremely high rates of continuing symptoms and relapses in most people with mental disorders, our treatments don’t appear to be modifying the diseases themselves.

As for psychotherapy and social interventions, there are those who believe that these treatments are addressing root causes. In some cases, this makes sense. For example, if a woman is in a physically abusive relationship and is clinically depressed as a result, helping her leave that relationship and build a new and better life may resolve her depression. Many psychotherapists would argue that the root cause of the woman’s depression was being in an abusive relationship. However, we know that because she has gone through the experience of an abusive relationship and developed clinical depression, she is now at increased risk for developing depression again at some point in the future, even if she is never in another abusive relationship. Given this fact, it seems there may be more to the depression than just the abuse, and that treating this factor alone did not definitively address the root cause. What causes her to remain at higher risk of developing depression in the future? If we truly understand what causes mental illness, we should be able to answer this question.

People in the mental health field often use circular logic to support their theories about what causes mental illness. For example, they may claim that if something works to relieve symptoms, it must have been the cause to begin with. The fact that the woman in the example above found relief when her situation changed is used as evidence that this situation was the root cause of her clinical depression. The fact that many psychiatric medications help relieve the symptoms of mental illness is used as evidence that the root cause of mental illness must be a chemical imbalance. As logical as this may seem, it isn’t always true.

Here’s an example to illustrate some of the flaws in this line of reasoning. Let’s go back to an infection that is causing a fever. If we didn’t know anything about infections or the causes of fever, and we were trying to figure it all out, we might do brain scans on people with fevers, looking for clues. Guess what we would see? We would see that the hypothalamus is overactive—that’s the part of the brain that controls the fever response. If we already knew that Tylenol works to reduce fevers, we might then do scans to research how Tylenol affects the brain. Lo and behold, we would see that Tylenol decreases this excessive activity in the hypothalamus! Based on this, we might logically conclude that the cause of the fever is a brain disorder involving the hypothalamus. We would have proof that a fevered patient’s brain activity is abnormal and that Tylenol works to reduce that abnormal activity. But it would be deeply misguided to conclude that we’d identified the fever’s cause. What we really did was identify a part of the brain that is involved in the production of a fever, and prove that a treatment that reduces fever affects that part of the brain as well. But Tylenol doesn’t treat infections. Reducing fever with that treatment wouldn’t change the course of the illness. Our fever and Tylenol brain scans only identified one aspect of the body’s response to an infection. We would better understand one symptom of the illness, or one part of its mechanism. It’s useful information, but it doesn’t help us understand the root cause of the fever at all—an infection.

Correlations, Causes, and Common Pathways

To answer the question of what causes mental illness, it is important to think a little about how we go about asking such a question, and the tools and principles we use to explore it. When medical researchers are doing their detective work to determine what causes an illness, they often study groups of people with and without the illness to look for correlations. A correlation is a relationship or connection between two things, or variables. If two variables are correlated, it might imply a cause-and-effect relationship, which is ultimately what the researchers are looking for. There are many types of studies designed to look for correlations. Researchers might perform brain scans on groups of people with and without depression and look for differences—the association with inflammation mentioned earlier is a correlation, the result of noticing that two variables (depression and inflammation) seemed to occur more often together, implying a relationship. One common type of study is an epidemiology study, which assesses variables in large populations of people and looks for correlations that way. For example, researchers might measure people’s weight, follow them for ten years, and record how many suffer heart attacks over that ten-year period. They would then look at the rates of heart attacks for different groups of people based on their starting weights to see if weight was correlated with heart attacks. If they found that obese people had higher rates of heart attacks than thin people, they would conclude that there was a correlation between obesity and suffering a heart attack. Notice that I said “correlation.” Based on this study alone, they can’t say that being obese causes heart attacks. This is one of the tricky things about correlational research. People often misinterpret the findings and make assumptions that aren’t warranted.

Correlation does not equal causation. Almost everyone has heard this. It means that a correlation does not necessarily tell us anything about cause and effect. Unfortunately, while most people are aware of the principle, they don’t apply it when interpreting research. If the study in the example I just discussed were released today, the headlines would likely read “Obesity Proven to Cause Heart Attacks,” further perpetuating the incorrect interpretation of studies like this. This may seem like semantics. You might be thinking, “Of course obesity causes heart attacks. What’s your point?” Well, as a matter of fact, obesity, in and of itself, does not cause heart attacks. It is a strong risk factor for having a heart attack, but it’s not a definitive cause. What’s the difference? Not all obese people experience heart attacks. If obesity causes heart attacks, all obese people should have them, and probably have them often. Also, there are plenty of people who are stricken by heart attacks who are not obese. If obesity is the cause of heart attacks, why would a thin person ever have one? Clearly, there must be more to heart attacks than obesity. So, what does cause a heart attack? The correct answer might be something along the lines of “the heart’s arteries develop atherosclerosis (thickening or hardening) and, at some point, become obstructed, resulting in some of the heart muscle dying or being damaged due to a lack of blood flow.” What causes those things to happen? That’s where obesity comes in as a risk factor, but other risk factors also contribute to this process, such as genetics, cholesterol and lipid levels, blood pressure, a lack of exercise, stress, poor sleep, and smoking. There is a cascade of events that leads to a heart attack; this cascade of events occurs over years. Understanding the entire cascade of events is important, as it offers numerous opportunities to intervene with different treatments. If we assume the cause is obesity and focus all our treatments on this risk factor, we will fail to prevent heart attacks in many people. How we define the cause of illness matters. Everyone likes simple answers. The way I just defined the cause of a heart attack is a complicated answer. As you will learn, this is also the case when answering the question: “What causes mental illness?”

Correlations, or relationships between two variables, can exist for several reasons. The most common interpretations are cause or consequence: One variable causes the other or is a consequence of the other. In other words, if A and B are correlated, it might be because there is a cause-and-effect relationship, where A causes B or B causes A. However, there is another possibility—one that some people have trouble understanding. Correlations can also reveal a common pathway or, sometimes, a common root cause.

Let’s assume we don’t yet know anything about the cold virus. All we know is that lots of people are showing up at the doctor with runny noses and sore throats. Some people also have other symptoms, such as headaches or fatigue, along with their runny noses and sore throats. Some have only one or the other—a runny nose or a sore throat—but many, many people have both. Researchers notice that there is a correlation between runny noses and sore throats. Since there is a correlation, there must be a relationship. But what relationship? Is it cause and effect? If so, which causes which? Many people seem to get their sore throats first and then develop runny noses, but not all—in fact, sometimes it is the opposite. So does the sore throat come first and cause the runny nose? Is it the other way around? Or are they both simply consequences of some unidentified illness that can cause both symptoms, and maybe even others?

Even though this is a straightforward example of an infection with the cold virus, at one point in time this all needed puzzling out. One source of confusion might have been people with allergies developing runny noses and sore throats when pollen levels were high. These people would present with the same, or similar, symptoms, but from a different root cause—allergies instead of the cold virus. The researchers would have to work hard to tease these two groups apart, trying to sort patients in various ways. At the end of the day, the symptoms in question would likely be indistinguishable: a runny nose is a runny nose, whether it’s from allergies or a cold. Researchers might have more luck by noticing things like seasonal patterns, or that some people seem to spread their symptoms to others (those with the cold virus) while others don’t (allergies). Looking for and combining patterns would give the researchers important clues about how to distinguish the two groups. In the end, they would have to address this important question: Are the runny noses and sore throats in these two different groups of people related in some way? They are, after all, the same symptoms. Why?

The answer is that they share a common pathway—inflammation. Inflammation is part of the body’s process of healing tissue and/or fighting off an attack, and it occurs whenever the immune system is activated. Whether the body is mounting a defense against a cold virus or an allergen, inflammation is causing the runny noses and sore throats. Inflammation is the common pathway or process that produces the symptoms for both groups of patients, but that pathway is downstream from the root cause. To get to the root causes, the researchers will need to determine what’s causing the inflammation.

Another way researchers might try to understand the symptoms of runny noses and sore throats and their causes might be to look at them separately. Not everyone has both symptoms, and some people with both have mostly one or the other. Our researchers might sort people into a group of those with primarily or only runny noses and another of those with primarily or only sore throats. This could make sense. After all, noses are different than throats. The treatments for the two symptoms are also different. Tylenol might help soothe the pain of sore throats, but it wouldn’t help runny noses. The most effective treatments for runny noses would primarily be things like pseudoephedrine or phenylephrine, ingredients found in Sudafed and cold and flu medications. There might be some treatments that relieve both symptoms in some patients—for instance, an antihistamine in those with allergies—but Tylenol would help almost all sore throats, and pseudoephedrine almost all runny noses, while neither would affect the other symptom. This stark difference in treatments might support the categorization of people into a runny nose group or a sore throat group. The researchers might label these distinct disorders—runny nose disorder and sore throat disorder.

Given that the treatments for these disorders are so distinct as well, they might instead think of these disorders as they relate to those treatments. The sore throat group might be thought of as having a Tylenol deficiency disorder: sore throats must be due to people not having enough Tylenol in their system, as correcting this deficiency seems to rectify the problem. The other group might be called pseudoephedrine deficiency disorder, as the effectiveness of the treatment clearly indicates a pseudoephedrine imbalance in the body.

As facetious as this seems, this is the very same logic that we use to conclude that depression is due to a serotonin deficiency and psychosis is due to too much dopamine. It makes sense, until you think about it with an example like the cold virus, which we understand well. In this context, it seems ridiculous. Yet that’s what we are doing in the mental health field today. We look at the treatments that work and assume that this tells us the story of what’s causing the disorders. And the disorders themselves are simply clusters of symptoms we have labeled disorders—the diagnostic labels mean nothing in terms of cause and effect, or what is happening in the body or brain.

Let’s return to our hypothetical researchers. These researchers have identified two distinct disorders—runny nose disorder and sore throat disorder. The disorders have different symptoms and different treatments, so the researchers are feeling pretty confident about this classification system. The problem is that, while there are people with just one disorder or the other, comorbidity is common—in other words, there are a lot of people who have both disorders. People diagnosed with runny nose disorder commonly also develop sore throat disorder. But it’s also true in reverse. Runny noses and sore throats are a good example of a bidirectional relationship. It means that if you have either disorder you are at much higher risk for developing the other one. It doesn’t matter which one starts first. A bidirectional relationship often means that the two things share some common pathway. For runny noses and sore throats, as I’ve already discussed, the common pathway is inflammation. Sometimes, in addition to a common pathway, a bidirectional relationship can also imply the same root cause. In this example, we already know that there is a common pathway (inflammation) and different root causes (the cold virus and allergies, among others).

Comorbidity aside, given that the symptoms and treatments are different, researchers and clinicians might advocate for keeping runny nose disorder and sore throat disorder as separate diagnoses. But once someone comes along and identifies the common pathway or a root cause that produces both disorders, this should change. Why? Back to Occam’s razor—the law of parsimony. If there is a simpler explanation for something in medicine, that explanation is more likely to be true. In this case, the explanation of the cold virus (a root cause) causing both disorders is much simpler than that of people developing both sore throat disorder (due to a Tylenol deficiency) and runny nose disorder (due to a pseudoephedrine imbalance) at the same time. Identifying that allergies (a different root cause) can cause both disorders would be an equally valid reason to change the medical field’s approach to diagnosis based on these symptoms. And of course, identifying the common pathway (inflammation) would be especially helpful, as it would allow the development of more effective treatments—and also explain why the symptoms of two different root-cause disorders, a cold and allergies, can be identical.

But the same root cause can also result in different symptoms in different people . . . especially when preexisting vulnerabilities come into play. The flu is a great example. People infected with this virus usually experience a predictable array of signs and symptoms—fever, muscle aches, lethargy, and so on. Yet even though they all have the same illness, different people can have different symptoms to varying degrees. And in people with preexisting conditions, this difference can be magnified. A healthy twenty-year-old might spend a miserable weekend feeling achy and feverish and then bounce back quickly. On the other hand, a child with preexisting asthma might develop severe airway inflammation and end up in the hospital on a ventilator. A frail eighty-year-old man might experience devastating effects resulting in organ damage and death. These people’s suffering stemmed from a single root cause—infection with the flu virus—but that cause produced very different consequences.

At this point, you can probably see both why the question of what causes mental illness matters and why it has remained so persistently difficult to answer. We in the mental health field are working with syndromes defined by symptoms and symptomatic treatments. Right now, we are treating infections with Tylenol. The goal is to understand the physiology of mental disorders, enabling us to develop effective treatments, and ideally, prevent these disorders before they occur.

Causation is proving that one thing causes another. Correlational studies on their own simply cannot do this. They can suggest causation, or at least provide clues, but proving causation requires something more. One type of study that can prove causation is called a randomized controlled trial. For example, to prove that the cold virus can cause a runny nose, researchers could take a group of people who are not sick, expose half of them to the cold virus by squirting the virus up their noses and the other half to a placebo (by squirting plain water up their noses). They could then record the number of people in each group who develop runny noses over the following five days. If the cold virus causes runny noses, the group exposed to the cold virus should have a much higher rate of runny noses than the placebo group. In fact, these studies have been done, and this is true.

One of the challenges in proving causation of a serious or life-threatening disorder in humans is that randomized controlled trials are unethical. Thus, even if we do have a plausible theory of what causes cancer, or mental illness, it will be unethical to expose people to this cause to test the theory definitively. So, what can be done in situations like this? Researchers are sometimes allowed to do equivalent experiments on animals. In the mental health field, this can play a role, but it has some limitations given the nature of mental disorders. Another alternative would be to develop a scientific theory of what might be happening in the body or brain from beginning to end—the cascade of events that leads to mental illness, like the cascade of events we talked about earlier as leading to a heart attack. Once established, researchers can then study people who have already been exposed to different risk factors and look for evidence of this cascade of events occurring in them. As you will learn, all of this research has already taken place; the evidence has been gathered. It’s just that no one has put it all together. That’s what this book does.