By Shankar Vedantam
Washington Post Staff Writer
Tuesday, June 28, 2005
Last of three articles
John Zeber recently examined one of the nation's largest databases of psychiatric cases to evaluate how doctors diagnose schizophrenia, a disorder that often portends years of powerful brain-altering drugs, social ostracism and forced hospitalizations.
Although schizophrenia has been shown to affect all ethnic groups at the same rate, the scientist found that blacks in the United States were more than four times as likely to be diagnosed with the disorder as whites. Hispanics were more than three times as likely to be diagnosed as whites.
Zeber, who studies quality, cost and access issues for the U.S. Department of Veterans Affairs, found that differences in wealth, drug addiction and other variables could not explain the disparity in diagnoses: "The only factor that was truly important was race."
The analysis of 134,523 mentally ill patients in a VA registry is by far the largest national sample to show broad ethnic disparities in the diagnosis of serious mental disorders in the United States.
The data confirm the fears of experts who have warned for years that minorities are more likely to be misdiagnosed as having serious psychiatric problems. "Bias is a very real issue," said Francis Lu, a psychiatrist at the University of California at San Francisco. "We don't talk about it -- it's upsetting. We see ourselves as unbiased and rational and scientific."
As the ranks of America's patients and doctors become more diverse, psychiatrists such as Lu are spearheading a movement to address the problem. Clinicians need to be trained in "cultural competence," they say, to prevent misdiagnosis and harm.
Psychiatrist Heather Hall, a colleague of Lu's, said she had to correct the diagnoses of about 40 minorities over a two-year period. She estimated that one in 10 patients referred to her came with a misdiagnosis such as schizophrenia, a disorder characterized by social withdrawal, communication problems, and psychotic symptoms such as delusions and hallucinations.
Unlike AIDS or cancer, mental illnesses cannot be diagnosed with a brain scan or a blood test. The impressions of doctors -- drawn from verbal and nonverbal cues -- determine whether a patient is healthy or sick.
"Because we have no lab test, the only way we can test if someone is psychotic is, we use ourselves as the measure," said Michael Smith, a psychiatrist at the University of California at Los Angeles who studies the effects of culture and ethnicity on psychiatry. "If it sounds unusual to us, we call it psychotic."
When hospitals diversified their staffs to include Spanish-speaking doctors, many cases of psychotic behavior were reassessed, he said: "Half the cases were rediagnosed as depression. Some doctors think if you don't make eye contact, you can be diagnosed. In some communities, eye contact is a sign of disrespect."
Zeber and a team of other researchers said they do not know why doctors were more likely to diagnose schizophrenia among blacks and Hispanics. Perhaps diagnostic measures developed primarily with white patients in mind do not automatically apply to other groups, said Zeber, who published his results in the journal Social Psychiatry and Psychiatric Epidemiology.
"Race appears to matter and still appears to adversely pervade the clinical encounter, whether consciously or not," Zeber and his colleagues wrote in their October 2004 report.
Darrel Regier, director of the division of research at the American Psychiatric Association and U.S. editor of the journal, said the study had been carefully conducted. He agreed that cultural differences between patients and doctors could result in misdiagnosis.
"I believe bias exists, and there is a risk a psychiatrist with a different cultural experience than a patient can misinterpret the expression of a psychiatric symptom," he said. "If you have a very religious group of patients and a very secular psychiatrist who thinks beliefs in spirits or hearing the voice of God is not normal, you are going to have misses."
But he added that Zeber's study did not explain what caused the diagnostic disparity among the veterans. Regier also questioned whether the veterans in the study were representative of the general population, or even representative of all veterans. Different ethnic groups seek care in different ways within and outside the VA, he said, and blacks tend to seek care when they are sicker than white patients.
While agreeing that even more comprehensive analyses are possible, Zeber stood by his findings. The study had carefully eliminated a host of confounding variables, he said, and the analysis had not found that black patients were any sicker than whites. "Access issues or selection bias are unlikely to account for our findings," the paper concluded.
"If you have an African American patient presenting with elevated paranoia, that has been referred to in some quarters as healthy paranoia based on how they perceive society," said Zeber, who works at the Veterans Affairs Department's Health Services Research and Development center in San Antonio. "If you base your diagnosis on that symptom, you can be misled."
Zeber's argument is supported by a panel of academic experts who helped draft a research agenda in 2002 for the next edition of psychiatry's manual of mental disorders.
They wrote: "Misdiagnosis due to a different cultural perspective of bizarreness is rather frequent." Inattention to the role that social standards and cultural factors play in diagnosis has caused patients to be stereotyped, they added, "with obvious negative consequences for diagnosis and treatment."Rethinking a Diagnosis
The patient was a young black man named Kevin Moore. He had been picked up by police -- after his mother called 911 and said he was making threats. The police brought him to San Francisco General Hospital.
Moore already had a diagnosis from previous stints at other hospitals: schizophrenia.
But psychiatrist Heather Hall thought something was wrong. Patients with schizophrenia can seem to shrink into their own world.
Superficially, Moore matched that description. He was uncommunicative. But when Hall looked closer, she noticed something else.
"A schizophrenic would be flat, he would be staring blankly into space," Hall said in an interview about Moore's case, given with his permission. "His expression wasn't moving, but he wasn't blank. He looked really, really sad."
After a thorough evaluation, Hall changed Moore's diagnosis to depression and reconfigured his medication regimen. She spent hours with Moore, coaxing him to talk. Within weeks, he began opening up about a host of interpersonal problems.
Moore said he was first diagnosed with schizophrenia when he was 16 or 17. In an interview at the San Francisco hospital, he was dressed in a baggy sweat shirt and sported his hair in cornrows. His braces made him seem younger than his age -- 24 at the time.
He said the police had picked him up because he had talked of getting a gun. A quarrel with a friend had escalated into a fight, and his mother had called the police. Moore thought his mother had overreacted, and he was sullen and uncommunicative when the police forcibly took him to the hospital.
"I probably didn't want to talk to any people," he said. "I didn't want to be there."
The particularly close attention that Hall paid to Moore was not the only unusual thing about his treatment. Moore was treated at one of the hospital's "focus units" -- inpatient psychiatric centers that focus on how culture and ethnicity influence psychiatric diagnosis and treatment.
The units pay attention to everything -- the decor as well as the treatment: The Black Focus Unit, for example, had African and African American art and icons on the walls. The occupational therapy room had photos of Vanessa Williams, Maya Angelou and Oprah Winfrey. The hospital also had an HIV-AIDS and a Lesbian, Gay, Bisexual and Transgender Focus Unit, as well as a Latino/Women's Focus Unit. The Asian Focus Unit had bulletins printed in multiple Asian languages.
The specialized units have been hailed as an innovative way to put patients at ease, but they have also faced criticism.
Psychiatrist Sally Satel described them as a type of "apartheid." Satel, who is affiliated with the American Enterprise Institute and is the author of "PC, M.D.: How Political Correctness Is Corrupting Medicine," said such divisions can prompt patients to avoid examining the real source of their mental problems.
"In its worst form, it is not really counseling," she said of what she called multicultural therapy. "It is a support group between two people who want to blame the outside world."
Moore and psychiatrist Hall are both black, but the hospital does not match doctors and patients by ethnicity. Every unit had staff members and patients from diverse backgrounds, and psychiatrist Lu said the wishes of patients and special needs such as language and prior history determined where patients were assigned. Every unit had specialized training -- staff members at the Asian Focus Unit, for example, spoke 14 languages.
Hall said Moore was a perfect example of why the Black Focus Unit is important: "Maybe because I am an African American psychiatrist, maybe he was able to show me a little more of himself for me to make an accurate diagnosis and change his treatment to a more accurate treatment."
She added, "Because the people who work on our unit are sensitive to the issues of African Americans, we are much more likely to look at our patients with eyes that aren't clouded by preconceived notions."
The psychiatrist recalled another case of a black man diagnosed as delusional. The man had talked about going to another city and getting revenge on people who had killed his son.
"The treatment plan was filled out by someone who was not part of our focus unit," she said. "She assumed it was a delusion -- she said, 'This man has a delusion that his son was killed in a hate crime.' " Hall checked out the man's account. It turned out to be true.
"People say minorities don't follow up" in psychiatric care, Hall said. "Maybe on their first session they are not heard. Why would they come back? If I tell a therapist I am being brutalized and he thinks I am delusional, why would I come back?"
Other clinicians echo such views. UCLA's Smith, who speaks Spanish, said that while making rounds with residents, he once asked an interpreter to check whether a Spanish-speaking patient wanted to commit suicide: "The patient said, 'I feel so bad I could die.' " But rather than convey the sense that the patient was in distress and felt terrible, Smith said, the interpreter told the residents, "She's suicidal."
Carl Bell, a Chicago psychiatrist, said he once went through the medical records of minority patients at Jackson Park Hospital in Chicago and found many misdiagnoses. One 30-year-old woman was talking fast, was calling people at all hours and did not seem to need sleep -- classic symptoms of bipolar disorder, or manic-depression. But her charts showed she had been hospitalized for schizophrenia and treated with injectable medications, which suggested that her doctors thought her schizophrenia was particularly severe.
"How does a woman with a college education, a job . . . she has euphoria, pressured speech, decreased need for sleep -- how do you get schizophrenia, chronic schizophrenia?" asked Bell, still incredulous.
Advocates for cultural competence say both clinicians and patients are unwilling to acknowledge that race might matter: "In a cross-cultural situation, race or ethnicity is the white elephant in the room," said Lillian Comas-Diaz, a psychotherapist in Washington, who added that she always brings up the subject with patients as a way to get hidden issues into the open -- and increase trust.
"I say, 'You happen to be Pakistani, and I am not -- how do you feel about that?' Sometimes they say, 'Oh, it's not important,' but when certain things happen [later] in therapy, people remember you opened the door and they come inside," she said.
Tina Tong Yee, a psychologist in charge of ensuring San Francisco's mental health services are culturally competent, said Western medicine's secular notions of normality are sometimes an uneasy fit in a deeply religious and increasingly diverse America.
"Seeing ghosts in my family was part of growing up," she said. "If I brought it up in therapy, you don't want someone to make that delusional."
Behavioral problems are different than other kinds of ailments, she added: "What you are reading is not a pulse, but how people act and behave and how you react to it. In a cross-cultural setting, it's ripe for misunderstanding."