Camila Knudsen (not her real name) doesn’t like to recall how her world began changing when she was 14. Background sounds in her suburban neighborhood—lawn mowers, planes, barking dogs—intermingled in a deafening buzz. The cacophony made it hard for her to hear what people were saying or to respond sensibly. “My mind was a blank,” she recalls. “Sometimes I felt like I couldn’t see or hear anything. I’d walk past someone and if they were laughing, I felt like they were laughing at me.” Exhausted, Camila holed up in her room at home and avoided her friends. She wouldn’t shower, and she spent much of her time staring into space. Her father remembers how on 9/11 he and Camila watched television together as a plane flew into the second tower of the World Trade Center. Camila didn’t so much as bat an eye when the building burst into flames. “It was as if she hadn’t seen it,” he says. “She had such a hard time speaking; it was almost to the point that she was catatonic.”
Alarmed by their daughter’s behavior, Camila’s parents began looking for help. But local psychiatrists were booked solid, some of them for the next six months. Meanwhile, her downward spiral accelerated. Then a pediatrician referred the Knudsens to William McFarlane, a psychiatrist in Portland, Maine, who had recently launched an experimental treatment program for early-stage mental illness. McFarlane championed a radical view that psychotic illnesses, including schizophrenia, can be prevented by treatment if caught early enough. His program, Portland Identification and Early Referral, or PIER, was a groundbreaking effort to find and treat patients showing early warning signs of psychosis.
Camila was accepted in September 2001 on the day she was assessed. The PIER staff believed that her symptoms, coupled with a history of schizophrenia on both sides of the family, put her at high risk for a full-blown psychotic break with reality. Quick intervention was crucial, McFarlane and his staff stressed, to prevent the onset of major hallucinations, delusions, and paranoia. Should Camila become psychotic even once, the lingering effects on her brain could diminish chances for recovery. One of the most crippling features of schizophrenia is that if delusional thinking is left unchecked, it takes over, robbing patients of the ability to recognize—and seek treatment for—their illness. Losing touch with reality puts schizophrenic patients at high risk of job loss, illness, homelessness, and suicide.
The Knudsens began PIER’s recommended treatment, a novel mix of psychotherapy and medication. Camila stuck with the program for four years and her symptoms subsided, slowly at first, but steadily. “You couldn’t tell she has a mental illness now if you tried,” her father says. “She’s going to be a productive member of society.”
By taking schizophrenia prevention to the community, McFarlane is charting a bold new course. With nearly $15 million in new funding from the Robert Wood Johnson Foundation—a longtime PIER supporter—his program has been broadened from Portland to four additional sites nationwide (Sacramento, California; Ypsilanti, Michigan; Salem, Oregon; and Glen Oaks, New York), embracing a combined population of more than 1 million people. Similar programs are ongoing in other countries, notably Australia, Denmark, Canada, England, and Germany. McFarlane’s program is by far the largest effort in the United States to prevent psychosis.
Through PIER’s efforts, McFarlane seems to have kept dozens of patients who once teetered on the edge of psychosis from falling into its grip. Some within the medical community, however, regard the preventive approach as promising but still in the experimental stage. Researchers are far from a complete understanding of what causes schizophrenia and how it affects the brain, and some psychiatrists contend that treating a patient for a disease not yet manifest is a clear violation of a basic tenet of medicine: to do no harm. But others support McFarlane in his efforts to block the onset of an illness so disabling that it can make a patient want to take his or her own life. “I don’t know that we have a way to prevent psychosis yet because we don’t have a clear understanding of what causes it,” says psychiatrist E. Fuller Torrey, executive director of the Stanley Medical Research Institute in Chevy Chase, Maryland. “But is it worth it to try? I’d say the answer is yes.”
With his treatment model expanding nationally, McFarlane hopes to show conclusively that his approach is indeed viable. One controversial feature of his program—which might more accurately be termed intervention than prevention—involves offering antipsychotic medication when only the warning signs of schizophrenia and other psychotic illnesses, such as altered perception and paranoia, are present.
“We have every reason to be cautious,” McFarlane says. “But our experience has been that so many of these kids are on the verge of psychosis that we don’t have the time or luxury to try other things. I’m a great proponent of psychosocial interventions,” such as counseling, family psychoeducation, and community support, “but none of them works very fast. What I wish we knew is whether there is a subgroup that might be stable long enough for the psychosocial treatments alone to take effect. But that’s a tricky race, and not one you want to lose.
“Our kids keep getting better and better,” McFarlane says. “Our experience has been that after a couple of years, they don’t need a lot more support.”
A bearded, professorial figure, McFarlane has long been on the leading edge of advancements in the psychiatric understanding of schizophrenia. In the 1970s, after completing a residency in psychiatry and a two-year fellowship at the Albert Einstein College of Medicine that focused on family therapy with severely psychotic patients, he began working at a mental health clinic in the South Bronx. But he found that family therapy often made patients worse. It failed, he says, because its methods were based on a now antiquated view that dysfunctional families caused psychosis. “The entire enterprise of family therapy was based on the notion that dysfunctional families brought about the onset of schizophrenia,” he says. “There wasn’t any real science behind that idea, but it was widely accepted and taught. And it turned out to be 180 degrees wrong.”
Toward the end of the 1970s, psychotherapists led by Carol Anderson at the University of Pittsburgh Medical Center launched a movement to rethink that approach. They began by stressing the long-standing evidence for schizophrenia’s biological roots, believing that the more the disease was understood by families, the less fearful it would seem. With a new talk therapy that they called psychoeducation, Anderson and her colleagues urged family members to temper their expectations of improvement, which had been found to stress patients and slow their progress. The best results, families were told, would come from low-key, convalescing interactions, which would allow patients to recover naturally from the shock of a psychotic episode. Even sudden noises and unanticipated changes in routine were to be avoided. Helping families learn how to direct their responses to the illness, rather than to the person suffering from it, is difficult, McFarlane says. But once they understood the process, they were relieved they could participate in the patient’s recovery.
McFarlane added his own twist by bringing families together in small groups. To his delight, he found that they bonded quickly in helpful ways. Patients treated in multifamily groups were more likely to stick with medication, he found, and within a few months, even the sickest among them were no longer psychotic.
Research during the late 1980s began showing that the sooner after a psychotic break patients were treated, the better they did. Symptoms were fewer and less intense. Often the symptoms could be controlled with a lower dose of medication than that used to treat full-blown psychosis. Patients also had less evidence of the loss of brain tissue, a key characteristic of the disease.
“This really captured the field’s imagination,” says Jeffrey Lieberman, who chairs the department of psychiatry at Columbia University College of Physicians and Surgeons and directs the New York State Psychiatric Institute. “Until then, we figured there wasn’t any real rush to treat. As long as patients weren’t hurting themselves or others, it wouldn’t matter if we treated them now or later. But when we recognized that treatments had to be fast, this created an urgency to find people with psychotic illness and to get them on medication quickly.”
With psychiatry turning to “first-episode” patients, those who had only recently suffered a psychotic break, McFarlane wondered if treating patients before psychosis even occurred might prevent schizophrenia altogether. He wasn’t alone in asking. In 1984 the late Ian Falloon, professor of psychiatry at the Auckland Medical Center in New Zealand, created a model mental health service in Buckingham and Winslow, two small towns northwest of London, to identify and treat high-risk patients in the community. Caregivers offered a range of supports, including family psychoeducation, home-based stress-management resources, and low doses of antipsychotic drugs. When the study wrapped up four years later, Falloon found that the incidence of schizophrenia in the treatment zones was one-tenth of what it had been a decade earlier. The program was not a controlled study, however, so there may have been other factors contributing to the decline.
By this time, McFarlane had perfected his own psychosocial technique at Columbia’s College of Physicians and Surgeons. There he came to find that multifamily therapy in schizophrenia worked best when combined with support from social workers, nurses, and psychiatrists to help patients find employment and housing so they could live independently. Using this approach, McFarlane claims he was able to halve rates of psychotic relapse and double employment, even among the sickest patients.
In 1992 McFarlane moved to the Maine Medical Center to become chief of the department of psychiatry. He also began consulting on a large first-episode treatment program in four cities in Norway and Denmark, applying insights from earlier studies that had convinced him that combining multifamily psychoeducation, social services, and low-dose antipsychotics could prevent relapse among young, first-episode patients.
The chance to apply that method with at-risk patients finally came in 1997. Aiming to replicate his success in Scandinavia, McFarlane set up a small first-episode pilot program in Portland. Within a few months he started getting referrals for patients who seemed on the verge of psychosis—he calls them “prodromal,” showing early symptoms and signs of a disease—so he decided to add them to the first-episode group. Over the following months, almost none of the handful of at-risk patients converted to psychosis. “It all went incredibly well,” he says. “So we thought, ‘Well, this is worth doing,’ and at that point, we started looking for funding to develop a formal team dedicated to prodromal patients.”
Just as McFarlane’s preventive program was getting under way, Thomas McGlashan, a psychiatrist from Yale School of Medicine, was facing fierce opposition to his plan to try out the antipsychotic olanzapine as a preventive treatment. Attacked by critics who insisted it wasn’t ethical to use the drug on young patients who weren’t yet psychotic, McGlashan says he supplied “about 200 pounds” of supporting documentation to the federal Office for Human Research Protections before he got a green light to continue.
McFarlane says that his program, too, faced strict ethical review from both the Maine Medical Center and the National Institute of Mental Health (NIMH). The difference, he believes, is that PIER fostered local support by engaging consumer advocacy organizations and safeguarding patient privacy. Because of the stigma surrounding mental illness, McFarlane says, confidentiality is crucial. “It’s a big issue that’s important to the success of the program, and we bend over backward to protect the privacy and confidentiality of these young people.”
The PIER program relies on community networks to flag vulnerable patients and bring them in for treatment. Clinicians, social workers, teachers, and others who routinely work with young people are taught to recognize schizophrenia’s early warning signs, which include paranoia, disorganized thought, fleeting hallucinations, social alienation, and a sudden loss of interest in activities. Admission to the program starts with a phone interview. Nearly 60 percent of those who move past that initial screening to a diagnostic assessment in person are accepted and supplied with a two- to four-year treatment regimen for free.
The approach employs a mix of psychosocial measures aimed at limiting the patient’s emotional stress load. Precisely how such stress can lead to a psychotic break isn’t understood, but scientists do know schizophrenia has complex genetic and environmental components. Since the inherited risk is based in family genetics and is therefore beyond clinical reach, McFarlane relies on multifamily group therapy and social supports at school and at work to protect vulnerable patients from environmental triggers—illicit drugs and emotional stress in particular—that might send them over the edge. Over time, most patients learn how to manage their condition, just as patients with chronic diseases such as diabetes learn to monitor symptoms and medications.
McFarlane describes the case of one patient who felt her symptoms returning after a weekend party. At school on Monday, the young woman noticed that “things seemed strange,” he says. “Noises were disturbing her, lights seemed too bright, and she began to worry she was slipping. So she gave us a call, and we told her to take it easy for a few days; we reminded her to take her medication and to get enough sleep, to reduce her stress load below normal. And she did that, and those symptoms started to diminish. So she realized she could use those symptoms as a marker or indicator, almost the way a diabetic might look at blood sugar.”
In this multipronged approach, only the early use of medication worries McFarlane’s skeptics. Antipsychotics, particularly drugs like olanzapine (marketed as Zyprexa), have severe metabolic side effects, which include weight gain and other changes that heighten the risk of diabetes. Long-term use of all antipsychotics can also cause uncontrollable movements known as tardive dyskinesia. The newest antipsychotics, including aripiprazole (Abilify), appear to have far less metabolic risk. Camila, who gained weight on Zyprexa, says she has lost it on Abilify.
Still, many psychiatrists insist antipsychotics shouldn’t be given without a confirmed diagnosis of a psychotic disorder. “We just don’t know enough yet about the risks and benefits of antipsychotics, particularly in children and adolescents,” warns Diana Perkins, a psychiatrist at the University of North Carolina at Chapel Hill. “My concern is that using them for prevention is premature.” Perkins points out that antipsychotics block dopamine receptors in the brain, which can also have the effect of depressing estrogen and testosterone levels. Whether that has any impact on development isn’t known. “What little data we have suggest sexual development might be altered, but the evidence base for that assumption is weak,” Perkins says.
The debate underscores how little is known about the biological origins and onset of schizophrenia itself, as well as how best to treat its early stages. McFarlane’s approach raises two basic questions. The first is whether a set of symptoms can reliably predict the onset of full-blown psychosis. Some scientists use the term prodromal to describe someone who seems headed toward the disorder, but the medical community has not yet formally accepted this as a diagnosable, treatable condition. Symptoms considered prodromal (the term comes from the Greek for “running ahead”) vary from patient to patient. Camila displayed some classic prodromal indicators: An otherwise healthy kid, she was suddenly victimized by minor hallucinations and paranoia and withdrew from social interaction. But schizophrenia often strikes during adolescence, a time when—as any parent knows—bizarre behaviors are all too common.
Research released just this past January delivered the most recent evidence that prediction of psychosis is possible. According to a nationwide study of 291 patients coordinated by the NIMH, prodromal symptoms predict the onset of psychosis 35 percent of the time, which is comparable with the prediction rate for type 2 diabetes. But Robert Heinssen, who is chief of the Adult Treatment and Preventive Intervention Research Branch at the NIMH and who led the study, claims that prodromal symptoms, when considered in the context of a full medical history, can produce a much more accurate prediction. “We found it can soar to 81 percent if additional features are present, like a family history of psychotic illness, odd or weird thoughts, and severe social deficits, such as having no friends or withdrawing from contact with others,” he says.
The second question is whether antipsychotic medication is an appropriate treatment for preventing the onset of psychosis. While Perkins expresses concerns about the risk of the drugs, Cameron Carter, a psychiatrist and brain researcher at the University of California at Davis, sees the problem differently. If doctors can foretell schizophrenia, he says, then the benefits of preventive drug treatment outweigh the risks of side effects. “It’s likely the side effects would be far less damaging than the psychosis itself,” he says.
McFarlane has not published any results from his program, and other data relating to antipsychotics in prevention are not definitive. McGlashan’s clinical trial with olanzapine to prevent psychosis remains the only one yet to investigate antipsychotics as the sole intervention. Conducted from 1999 to 2004 in a population of 60 patients (29 treated and 31 controls), that study produced statistically inconclusive results, in part because too many drug-treated patients dropped out complaining of weight gain. An earlier study performed in 2002 by Patrick McGorry, from the University of Melbourne in Australia, found that an antipsychotic called risperidone plus psychotherapy was more effective at preventing psychosis than supportive care alone. After his six-month trial ended, however, several patients in the treatment group became psychotic, raising the question of whether the treatment was preventing schizophrenia or simply controlling its symptoms.
During interviews for this article, neither McGlashan nor McGorry backed the use of antipsychotics in prevention outside of clinical trials. Likewise, Heinssen says that “current evidence does not support the use of antipsychotic medications as a prevention strategy in at-risk individuals.” Heinssen adds that NIMH is not preparing any clinical trials to investigate the preventive use of antipsychotic medication. The research effort he directs focuses on entirely new compounds that might slow the loss of brain cell connections, which may play a role in schizophrenia biology.
McFarlane continues to collect data on PIER patients, but he says that of the more than 100 patients treated by PIER in Maine so far, less than 10 percent have developed a major psychosis, compared with 30 to 40 percent who may have converted without treatment. If true, those results represent many lives saved from the devastation of severe mental illness. “What we hope is that the benefits of treatment will be lifelong,” McFarlane says. “We don’t have any empirical evidence to support that yet, but what we’ve seen is that young people who still haven’t converted to psychosis after about three years of our treatment don’t seem to be at much risk. As to when or if they can go off medication, that’s hard to say. I think many of our patients don’t feel a need to stop; they certainly don’t feel oppressed by it. At a certain point it becomes a personal choice.”
The debate underscores how little is known about the biological origins and onset of schizophrenia itself, as well as how best to treat its early stages.
At the moment, the most powerful evidence for PIER may lie in the stories of its participants. Josef Bruno (not his real name) showed up at the age of 18 after struggling for years with a nonverbal learning disorder, multiple psychiatric hospitalizations for anxiety, and a growing sense that he was disconnecting from himself. Treating him has not been easy. He requires numerous medications, frequent contact with clinicians and social workers, and lots of caring attention from his parents. Yet two years into the program, he has made important strides: His mental state has improved, and he is living independently for the first time, sharing an apartment with a roommate. What’s more, he has a job at a convenience store and aspirations to go to college.
Camila and her family stuck with PIER for the four-year treatment program, which ended formally in 2005, and still keep in touch with counselors there. Though her improvement was steady, Camila’s treatment posed its own challenges. Her mother admits she was overprotective and too involved in her daughter’s illness. It was hard to back off and give Camila space to recover, she says. Her father admits to being frustrated by how Camila’s condition made parenting more complicated. “Sometimes it felt like she was manipulating us,” he says with a chuckle. “And that was difficult because we didn’t know if what we were dealing with was part of the illness, or just her being an adolescent.”
Meanwhile, Camila’s health still hinges on antipsychotic medication. In the summer of 2007 she went off the drugs for a spell and her strange feelings returned. She says the medication doesn’t bother her, though. “Sometimes I feel so good, I just forget about it and don’t think I need it,” Camila explains. “But then I start to go downhill.”
Camila’s reliance on antipsychotics does raise a couple of issues. On the one hand, it shows that the threat of psychosis hasn’t really been removed, it’s just been held in check. While McFarlane suggests that over time, some patients may be able to go off medications, he acknowledges that PIER hasn’t developed a plan for managing that process. On the other hand, PIER’s intervention may have given Camila a chance to lead a normal life. Without the medications she might have plummeted toward full-blown schizophrenia, a life-threatening outcome with tremendous costs to society. Indeed, by limiting the debate to whether PIER does or doesn’t prevent schizophrenia, it’s possible to miss a larger point, namely, that most of the patients enter the program because they are in serious need of help, without which they could succumb to psychosis, violence, or suicide. According to Anthony Lehman, chairman of psychiatry at the University of Maryland School of Medicine, “The real value here is that the program is bolstering public awareness about prodromal symptoms, and it’s providing venues for kids and their families who have these problems to get help.”
Researchers hope that a risk-free way of blocking psychosis is around the bend. Recent imaging studies—some by neuroscientist Tyrone Cannon, from the University of California at Los Angeles—are providing clues to schizophrenia’s underlying pathology. These studies suggest that prodromal symptoms emerge from a progressive loss of synapses in the brain. “It’s not that whole brain cells are lost, as is the case with Alzheimer’s disease,” Cannon explains. “It’s the elements that brain cells use to communicate with each other that are lost.” This hypothesis might help explain why prodromal symptoms occur: As connections between neurons dwindle, prodromal patients might lose, for example, the ability to filter out irrelevant sensations, which then flood the brain without any processing. That may be why Camila couldn’t deal with everyday sounds; they just came at her in a wave that would ordinarily have been picked apart and filed: a barking dog here, a car engine there.
Ideally, Heinssen says, research can lead to ways to protect the prodromal brain. The methods might include drugs to slow the loss of connections between neurons, as well as cognitive and behavioral therapies designed to strengthen imperfect circuits. “That is how we see prevention unfolding,” he says. “We want to combine neuroimaging and cognitive assays with clinical symptoms to enhance prodromal diagnosis, identify neuroprotective agents to target underlying disease mechanisms in the brain, and develop behavioral and cognitive exercises that will increase the patient’s ability to adapt.”
Remarkably, two compounds that seem to exert these neuroprotective effects—both of them a focus of intense interest in schizophrenia research—aren’t sophisticated drugs but simple compounds found in nature. Supplementing diets for three months with daily two-gram doses of fish oil, a source of omega 3 fatty acids, was shown in a forthcoming study by University of Melbourne researcher Paul Amminger to reduce conversion to psychosis among 81 high-risk patients. Another promising compound, according to Scott Woods of Yale School of Medicine, is D-serine, an amino acid. D-serine may boost the growth of synapses, he says, and has therefore attracted NIMH funding for clinical studies.
In the meantime, McFarlane treats patients with the imperfect tools science has supplied so far. Of the need, there can be no doubt. Schizophrenia afflicts roughly 1 percent of the population—some 2.4 million Americans—destroying personal relationships and setting up conditions for high rates of joblessness and drug addiction. Untreated, many patients wind up homeless, incarcerated, or dead. “I think there will be an intervention based on a biological mechanism at some point in the future,” McFarlane says. “We aim to have an identification system in place by then, and the treatments we can use in the meantime are good enough. Schizophrenia is as devastating in my mind as cancer, maybe more because it typically strikes at earlier ages and wipes out someone’s whole life. So, yes, we’re trying to forge ahead with a public health intervention, even if we don’t have the best possible treatments. I think it’s silly to wait. Wait for what?”