Three trials provide guidance on race, ethnicity: in STEP-BD, response to treatment for psychosis was 40% for white patients and 0 for black patients.

PHOENIX--Teasing out any differences in the success of psychiatric treatment by race or ethnicity is hampered by a dearth of data on minorities, but secondary analyses of major studies on depression, bipolar disorder, and schizophrenia provide some insights.

Across the three studies, blacks and Hispanics in the United States were more socially disadvantaged and had more comorbidities at study enrollment, compared with whites. Even after investigators controlled for those differences, blacks had poorer outcomes, compared with whites, in the studies of bipolar disorder and depression but not for schizophrenia, Jodi M. Gonzalez, Ph.D., said at a meeting of the New Clinical Drug Evaluation Unit sponsored by the National Institute of Mental Health.

These kinds of analyses start to fill an important void in medical data, but more research that includes ethnic/racial minorities is needed, she said. A previous review of 379 clinical trials published between 1995 and 2004 found that less than half provided complete ethnic/racial breakdowns of subject data, and 70% had no potential for subgroup analyses by race or ethnicity.

Minorities who do enter clinical trials of ten mirror differences between groups that are seen in the larger society. For example, 8% of U.S. whites have incomes below the poverty level, compared with 22% of blacks, 21% of Hispanics, 26% of American Indians or Alaskan natives, and 11% of Asian Americans or Pacific Islanders.

The three studies presented in the same session at the meeting tried to control for some of these differences.

The STEP-BD Study

Dr. Gonzalez and her associates analyzed data on a subset of 2,035 patients aged 15 years and older from the multicenter, 5-year-long STEP-BD (Systematic Treatment Enhancement Program for Bipolar Disorder) study, the largest treatment study of bipolar disorder. They matched 174 black patients by gender, income, education level, and insurance status with 800 non-Hispanic white controls, and did a similar match with 163 Hispanic patients with a separate control group of 898 non-Hispanic white patients.

A significantly greater proportion of blacks (31%) had psychotic symptoms at enrollment, compared with whites (18%). Blacks were less likely to be married than whites (19% vs. 36%) and less likely to have private insurance (39% vs. 52%). Recovery during the first year of treatment was significantly more likely in whites (45%), compared with blacks (25%), as measured by the Global Assessment of Functioning, reported Dr. Gonzalez of the University of Texas Health Science Center, San Antonio.

A trend for greater likelihood of recovery among Hispanics (59%), compared with whites (50%), did not reach statistical significance.

"Psychosis likely contributes to poorer outcomes" in the black group, Dr. Gonzalez said, adding, "Is there really greater psychosis, or a greater perception of psychosis" by the mostly white clinicians and researchers? Among black and white patients with baseline psychosis, 40% of the white patients and none of the blacks met criteria for response to treatment in her analysis, she said.

No significant differences were found between groups in recovery rates for depression, as measured by the Montgomery-Asberg Depression Rating Scale (21% for blacks, 29% for whites, and 44% for Hispanics) or for recovery from mania as measured by the Young Mania Rating Scale (29% for blacks, 41%-42% for whites, and 44% for Hispanics). The percentage of symptom-free days in the first year also did not differ significantly between groups (41% for blacks, 45%-48% for whites, and 44% for Hispanics).

Blacks and Hispanics/Latinos account for 12% and 13% of the U.S. population, respectively, but accounted for only 4% each in the STEP-BD study, she noted. The Hispanics in the study primarily spoke English, but 40% of U.S. Latinos primarily speak Spanish.

The STAR*D Study

The language spoken may have played a role in outcomes of the STAR*D (Sequenced Treatment Alternatives to Relieve Depression) study, Dr. Ira M. Lesser said. Many clinical trials will not accept monolingual patients who do not speak English, he added.

He and associates analyzed data on 14 weeks of outpatient treatment for major depression in adults with citalopram (Celexa), the first level of care in the study. At enrollment, the 1,853 white patients in the analysis had higher education levels, employment rates, income, and access to private insurance, compared with 495 black patients and 327 Hispanic patients.

Whites had a significantly lower rate of chronic depression (22%) at enrollment, compared with blacks (31%) or Hispanics (33%), and a higher rate of recurrent depression (78% for whites, 74% for blacks, and 65% for Hispanics).

Blacks were significantly less likely to achieve remission (19%) on treatment, compared with whites (30%); Hispanics (24%) fell between the two groups, as measured by the Hamilton Rating Scale for Depression, reported Dr. Lesser, professor in residence at the University of California, Los Angeles Harbor-UCLA Medical Center, Torrance.

After adjustment for baseline variables, however, remission rates on this scale were not significantly different among the groups. Blacks took longer to achieve remission than whites, but the difference disappeared after adjustment for baseline factors.

A higher proportion of blacks (33%) dropped out of the study for reasons unrelated to medication effects, compared with whites (21%) or Hispanics (25%).

A separate comparison by Dr. Lesser and his associates of 121 English-speaking Hispanics and 74 Spanish-speaking, monolingual Hispanic patients in the STAR*D trial found that the monolingual Spanish-speaking patients were older, less educated, and poorer; they were also treated more often in primary care clinics rather than by mental health specialists.

The monolingual Spanish-speaking group had significantly lower rates of remission (12% vs. 26% of English-speaking Hispanics) and of response to treatment (31% vs. 50%, respectively), and they were slower to respond to treatment, but these differences disappeared after adjustment for baseline variables.

The CATIE Study

The multicenter CATIE (Clinical Antipsychotic Trials of Intervention Effectiveness) study, an 18-month, randomized, double-blind study of 1,430 adults with chronic schizophrenia who were treated with one of five antipsychotic medications, included an impressive enrollment of Hispanics (12%) and black patients (35%) in the cohort.

Black patients at enrollment had lower scores on the PANSS (Positive and Negative Syndrome Scale), lower rates of extrapyramidal symptoms and akathisia, and higher self-rated quality of life, compared with whites. Whites had higher incomes and performed better on neurocognitive tests, and Hispanics were more likely to be married than the other groups.

No significant difference was seen in the primary outcome, which was the time to discontinuation of treatment (8 months for each group). After controlling for baseline variables, investigators found no differences between groups by the end of the study in PANSS scores, which surprised them, said Dr. Jose M. Canive.

There were no differences by race or in the number of daily medication capsules prescribed (and thus, presumably, in medication doses), which also was "different than what we had expected," yet a multidimensional measure described blacks as less adherent to their medication regimen, added Dr. Canive, who is a professor of psychiatry and neurosciences at the University of New Mexico, Albuquerque.

The main outcomes did not change after adjustment for adherence.

BY SHERRY BOSCHERT

San Francisco Bureau