Access and Adequacy of Depression Treatment for Racial and Ethnic Minorities in the US
Margarita Alegria, PhD Pinka Chatterji, PhD Norah Mulvaney-Day, PhD Zhun Cao, PhD Chih-nan Chen, PhD David Takeuchi, PhD
Academy Health Annual Research Meeting June 2007
Acknowledgements
Robert Wood Johnson Foundation
grant # K23 DA018715-01A2: Improving the Quality of Depression Treatment for Ethnic/Racial Minorities
Data Collection: NIH Research Grant # U01 MH62209
funded by the National Institute of Mental Health as well as the Substance Abuse and Mental Health Services Administration Center for Mental Health Services (SAMHSA/CMHS) and the Office of Behavioral and Social Sciences Research (OBSSR).
Motivation
Major depression is a prevalent and costly illness
& Miller, 1993; Wang, Simon, & Kessler, 2003). (Rice
Worldwide, it is the fourth leading cause of disability
and the leading cause of nonfatal disease burden (Ustun
et al, 2004).
Although well agreed upon indicators for diagnosis
and guidelines for evidence based care of depression exist (Cornwall & Scott, 2000; APA, 2000), a mismatch between diagnosis and treatment continues to exist in the US.
Motivation
Quality improvements in depression treatment using
guideline concordant treatments have been linked to greater effectiveness of care (Wells et al., 2000), with evidence that these improvements can benefit ethnic minority populations (Miranda et al., 2003).
Yet there is only inconsistent data regarding the gap in
quality treatments for depression between ethnic/racial minorities and whites. For example, some investigators have found lower quality for African Americans (Wang, Demler, & Kessler, 2002) while others find no ethnic or racial differences in the quality of depression care (Rollman et al.,
2002).
Objectives
To test for differences To compare adequacy
between ethnic/racial minority patients (Latinos, Asians, African-Americans) and non-Latino whites in the access to depression treatments.
of depression treatment for ethnic/racial minority patients (Latinos, Asians, AfricanAmericans) as compared to nonLatino whites
Hypothesis
We hypothesize that non-Latino whites will continue to
demonstrate greater access and receive higher quality depression treatment than ethnic/racial minorities, in the form of guideline concordant care, and appropriately prescribed medication.
Framework
Study uses a system cost effectiveness framework (Alegria et al., 2005; Frank et al., 1999). We evaluate if the U.S. healthcare system neglects to provide resources to those who would benefit from depression care by either not treating them or treating them ineffectively;
whether it allots resources to those who do not seem to
need depression care by treating them anyhow;
and whether these allocation errors are more likely to
happen for ethnic minorities as compared to non-Latino whites.
Allocation Errors in a Health Care System Unmet Need and Inefficient Use of Resources
A: Need B: Treatment
Unmet Need
Appropriate
Treatment
Inefficient Use of Resources
About one third get appropriate care (Wang et al., 2005)
Framework
Operationalization of our outcome measures:
Access to care as receiving any formal mental health
treatment; seeing a specialty or general health professional for problems with their mental health, nerves, alcohol or drugs in the last year.
Adequacy of treatment if either of the following two
criteria is met during a 12-month period: 1) eight or more visits of at least 30 minutes of counseling, or 2) antidepressant medication with 4 visits of medical supervision as defined by Wang et al. (2005).
Definition of Providers
Provider can be from the general medical sector, specialty
mental health sector, or a counselor or social worker in a nonmental health setting.
General medical sector providers include: general practitioner, family doctor, nurse, occupational therapist, or other health professional for a mental health problem. Specialty mental health sector providers include: psychiatrist, psychologist, counselor or social worker seen in mental health settings, or other mental health professional.
Combined NLAAS/NCS-R Study
A national psychiatric epidemiologic survey conducted to
measure psychiatric disorders and mental health service usage in a nationally representative sample of Asians and Latinos (NLAAS).
We pool data from the NCS-R (conducted in 2001-2002)
to incorporate contrasts to Non-Latino whites and African Americans.
NLAAS was conducted in 2002 and 2003 in English,
Spanish, Chinese, Tagalog and Vietnamese, based on the respondents’ language preference
Contains detailed information on psychiatric disorders
using the Composite International Diagnostic Interview (CIDI) and chronic conditions to do health adjustments.
Analytical Strategy
We classify the NLAAS/NCS-R combined sample into four
groups using the World Health Organization Composite International Diagnostic Interview (WMH-CIDI; Kessler & Ustun, 2004):
(Group 1)- Rs with last year DSM-IV diagnosis of major
depression or dysthymia;
(Group 2)- Rs who do not meet criteria for last year Dx of MDD
or dysthymia but who are sub-threshold cases of depression or have minor depression;
(Group 3)- consists of Rs who do not meet criteria of need, as
assessed by any psychiatric or substance use Dx;
(Group 4)-those with other Dx than depression who excluded
from analysis due to focus on depression treatment.
Analytical Strategy
Use a series of regression analyses to evaluate the main effect of
ethnicity/race on access and adequacy of treatment as well as interactions between ethnicity/race and depression.
In these models, we will combine Groups 1 and 2, to maximize our
power to detect effects. We will use data from all racial/ethnic categories to fit the best empirical model, reporting differences in quality care for depression in African Americans, Asians, Latinos, and Other group relative to Whites.
Demographic and socioeconomic status (SES) covariates included in
the regression model will be poverty level, age, gender, marital status, education level, region, metropolitan/rural area, nativity, insurance (uninsured, private insurance, Medicare, Medicaid/other public programs)
Results
Characteristics of NLAAS/NCS-R Respondents
Total combined sample n = 8,962 Age Category 18-34 years 35-49 years 50-64 years 65 years or more College Education No Yes Type of Insurance Not insured Non-Latino White n = 3,523 Latino n =2,776 Asian n = 2,075 African American n = 588 Chi-square test of difference (P value) 0.000
30.4% 30.2% 21.4% 18.0% 76.4% 23.6% 13.1% 54.4% 4.8% 20.1% 4.9%
26.0% 29.4% 23.3% 21.2% 73.6% 26.4% 9.1% 58.6% 4.8% 22.9% 2.5%
47.8% 30.6% 13.4% 8.3% 89.9% 10.1% 32.9% 40.8% 2.8% 9.8% 11.5%
40.1% 33.4% 17.1% 9.5% 58.8% 41.2% 12.9% 58.7% 8.8% 9.8% 4.9%
35.3% 33.1% 19.7% 11.8%
0.000
87.3% 12.7%
0.000
17.5% 41.2% 4.9% 17.8% 13.4%
Private through employer
Private purchased Medicare Medicaid
Other
2.7%
2.2%
2.2%
4.9%
5.3%
Characteristics of NLAAS/NCS-R respondents
Total Noncombined Latino sample White Latino n = 8,962 n = 3,523 n =2,776
Disorder Category
Asian n = 2,075
African American n = 588
Any Depressive Disorder in the past 12 months Any Subthreshold Depressive Disorder in the past 12 months No Depressive Disorder in the past 12 months
9.5%
10.1%
9.3%
5.6%
7.5%
1.2%
1.4%
0.8%
1.0%
0.8%
89.3%
88.5%
89.9%
93.5%
91.7%
*Any Depressive Disorder diagnostic category includes DSM-IV dysthymia and major depressive episode.
Any Depressive Disorder or Any Subthreshold Depressive Disorder in the past 12 months (N=1,247), age-gender adjusted
Group Non-Latino ID Group description White 1 No Treatment, 44.0% No antidepressants 2 No provider† visits, 6.9% Antidepressants 3 1-3 provider† visits, 10.4% Antidepressants 4 4+ provider† visits, Antidep. 2.0% <30 days ‡ 5 4+ provider† visits, 18.3% Antidep. 30+ days 6 1-7 provider† visits, 12.5% No antidepressants 7 8+ provider† visits, <8 Specialty visits 30+ mins, 2.4% No antidep. ‡ 8 8+ Specialty visits 30+ 3.6% mins, No antidep.
*p < 0.05, **p < 0.01, ***p < 0.001
African Wald test Latino Asian American p value 61.4% 75.2% 70.4% ***
3.0%
5.0% 1.5%
3.8%
2.2% 0.5%
4.7%
1.7% 1.1%
*
***
12.8%
10.5%
4.6%
8.3%
10.8%
4.1%
***
*
2.1% 3.8%
1.5% 4.0%
0.7% 6.5%
No Depressive Disorder in the past 12 months (n=7,715)
NonLatino White 87.8% 4.0%
2.1%
Group ID Group description 1 2 3 4 5
‡
African Wald test Latino Asian American p value
93.6% 95.5% 1.9%
0.7%
No Treatment, No antidepressants No provider† visits, Antidepressants
1-3 provider† visits, Antidepressants 4+ provider† visits, Antidep. <30 days 4+ provider† visits, Antidep. 30+ days 1-7 provider† visits, No antidepressants 8+ provider† visits, <8 Specialty visits 30+ mins, No antidep. 8+ Specialty visits 30+ mins, No antidep.
93.9% 1.4%
0.2%
***
*** ***
0.8%
0.5%
0.1%
1.3% 3.7%
0.2%
0.4% 2.6%
0.0%
0.7% 2.1%
0.0%
0.4% 3.6%
*
** *
6 7
8
‡
0.1%
0.2%
0.3%
0.3%
0.9%
0.4%
0.1%
0.3%
***
*p < 0.05, **p < 0.01, ***p < 0.001
Odds of accessing depression treatment (n=1128)
OR Race/Ethnicity 95% CI
With depression, Latino - White With depression, Asian - White
0.636 0.325
(0.463 - 0.874)** (0.167 - 0.633)** (0.143 - 0.490)** (0.476 - 0.808)**
With depression, African American - White 0.265 No depression, Latino - White 0.621
No depression, Asian - White
No depression, African American - White
WHO-DAS Disability Assessment No impairment
0.470
0.495
(0.333 - 0.661)**
(0.273 - 0.896)*
1
Days out of role >0
Cognition>0 Mobility>0 Self-care>0 Social Functioning>0 Role Functioning>0
3.425
2.061 1.299 0.812 0.788 1.415
(2.046 - 5.733)**
(1.492 - 2.847)** (0.905 - 1.864) (0.476 - 1.384) (0.558 - 1.114) (1.087 - 1.843)*
Odds of receiving adequate treatment among individuals who received any care (n=417)
OR Race/Ethnicity 95% CI
With depression, Latino - White With depression, Asian - White With depression, African American - White No depression, Latino - White No depression, Asian - White No depression, African American - White
WHO-DAS Disability Assessment No impairment Days out of role >0 Cognition>0 Mobility>0 Self-care>0 Social Functioning>0 Role Functioning>0
0.739 0.411 0.594 0.367 0.344 0.226
(0.416 - 1.312) (0.176 - 0.961)* (0.297 - 1.187) (0.210 - 0.640)** (0.176 - 0.674)** (0.081 - 0.634)**
1 4.365 2.181 1.292 0.707 0.567 1.185 (2.724 - 6.996)** (1.430 - 3.325)** (0.909 - 1.837) (0.384 - 1.303) (0.353 - 0.911)* (0.891 - 1.578)
Summary of Results
Data suggests that depressed ethnic racial minorities continue to:
Have decreased access to mental health care, even in the presence of
comparable levels of depression and disability
Access to Depression Tx for minorities - 36-73% reduced likelihood as compared to non-Latino whites.
Females, those ages 35-49, those with college education, the divorced/
separated and never married, those with Medicare, Medicaid or Other insurance (e.g. Veterans), and those disabled are more likely to access care, while males, the uninsured and the elderly are less likely to get access to depression care, adjusting for clinical profile. Relative undersuse of depression treatment is roughly the same for those w/ depression as for those w/out depression. No greater SCE.
Summary of Results
In terms of Adequacy of Depression Tx
Only significant difference was found for Asians, but not for Latinos or African Americans. This suggests the importance of dealing with the access problem. Even when taking into account the potentially mediating factors of insurance, education, language, region and poverty, our results do not change. Our data emphasize the need to make the system of care more targeted in reducing disparities in depression Tx for ethnic and racial minorities.