Title: Race, Ethnicity, and Gender: Lessons for Culturally Appropriate Care Models for Depression
1Race, Ethnicity, and Gender Lessons for
Culturally Appropriate Care Models for Depression
- Lisa A. Cooper, MD, MPH
- Professor of Medicine, Epidemiology and Health
Policy Management - Welch Center for Prevention, Epidemiology
Clinical Research - Johns Hopkins University
2Overview
- Case presentation
- Epidemiology of depression
- Race, ethnicity, gender, and depression care
- Review of selected studies
- Patient attitudes and preferences
- Patient-clinician communication
- Strategies to improve depression care
- Standard quality improvement
- Patient-centered and culturally-tailored models
- Implications for clinical care, health
professional education, future research, and
health policy
355-year old African American woman with depression
- History of Present Illness
- Mrs. S. is a 55 year old married African American
woman feeling down off and on for at least 6
months. Recently feeling tired and overwhelmed
and not sleeping right. - Past Medical History
- Acid reflux, sometimes its so bad I get chest
pains. - No history of treatment for depression.
- Medication
- Prevacid for acid reflux
- Family History
- No history of depression or other psychiatric
illnesses.
455-year old African American woman with
depression, contd
- Social History
- Habits She does not drink alcohol, smoke or use
illicit drugs. - Occupation She runs a housing program. Work is
hectic, someones always in crisis or needs
something. I cant even get 2 minutes to myself. - Family Her husband had a stroke 2 years ago.
- He has some mobility back now but he still
depends on me a lot, even when he could do things
for himself. - Ive been going through some problems in my
marriage since my husband became ill. - Social She belongs to a womens group at her
church. I have a very supportive family but its
difficult to ask for help. My husband expects me
to always be there for him.
555-year old African American woman with
depression, contd
- Her thoughts about depression
- I think Ive been in a depressed state for
awhile. I started writing in a journal, my church
group suggested it. I tend to get down at the end
of the year, in the winter. This year in
particular was bad. - Treatment by the doctor
- Joan was prescribed antidepressant meds at her
last visit. I didn't like how that Paxil made me
feel. I was feeling drained, not myself. I took
it for about a week and just felt worse so I cut
the pills in half and tried that for a couple of
weeks. I still couldn't function. I didn't tell
Dr. Morales. - I'm not opposed to counseling but right now I
think I'm doing much better and getting support
from the ladies in my group. I'm trying to do
things on my own. - What should she do?
6Major Depressive Disorder
- A very common and serious medical condition that
affects 5 of the population at any time - Affects women at twice the rate it affects men
- Is most common in persons aged 18 - 44 years
- Affects persons of various educational, social
class, and ethnic backgrounds at similar rates - Leads to functional impairment and worsens
physical health (e.g., cardiovascular disease) - Increases risk of mortality from suicide
7Treatment of Depression in Primary Care Settings
- Most individuals with depression in the U.S. who
seek care receive their mental health care in
primary care settings Wang 2000, Kessler 2005 - Treatment of depression in primary care remains
suboptimal - Under-recognition of mental illness is common
Perez-Stable 1990, Balsa 2005 - Primary care physicians use lower doses of
antidepressants and for shorter periods of time
than guidelines recommend Katon 1992,
Magruder-Habib 1989
8Racial and Ethnic Disparities in Mental Health
Care
- African Americans and Hispanics use specialty
mental health services at half the rate of whites
Vernon 1982, Sussman 1987, Hough 1987, Scheffler
1989, Gallo 1995 - Attrition from psychotherapy and pharmaco-therapy
is higher for ethnic minorities than whites Wells
1994, Brown 1999 - Disparities in mental health care not explained
by differences in education or health insurance
Padgett 1994, Charbonneau 2003
9Racial and Ethnic Disparities in Mental Health
Care
- Use of outpatient mental health services in
primary care settings has increased for African
Americans and Hispanics Cooper-Patrick 1999, Vega
1999 - However, ethnic minorities are less likely to be
recognized as depressed, receive
guideline-concordant care in primary care
settings, or to be referred to mental health
specialists Wang 2000, Borowsky 2000, Harman
2001, Young 2001, Miranda 2004, Leo 1998
10Gender and Depression Care
- Literature on gender and depression care limited
- Large study of depressed primary care patients
shows little evidence of gender disparities
except women less likely to be counseled about
anxiety and alcohol or drug use Chan 2006 - The associations of shared decision-making and
guideline-concordant care with patient
satisfaction and depression resolution are
similar among men and women Clever 2006, Swanson
2007
11Access to Health Care for Racial and Ethnic
Groups
Modified From Access to Health Care in America
(1993, Millman M, ed). Cooper LA, Hill MN, and
Powe NR. JGIM 2002 477-486
12Patient Attitudes and Preferences Regarding
Treatment of Depression
- Design Qualitative study using focus groups
- Study Subjects
- Patients African Americans and Whites, 18 years
or older with recent depressive episode - Professionals physicians and social workers
involved in care of general medical patients
Cooper-Patrick L et al, JGIM 199712431-438
13Conduct of Focus Groups
- Three groups two patient groups
(African-American and White) one professional
group - Groups led by trained moderator 90 minutes each
- Discussions audio taped and transcribed verbatim
- Focus group questions addressed
- Depression experience/help-seeking behaviors
- Treatment preferences
- Perceived barriers to mental health care
14Content Analysis
- Two investigators independently reviewed
transcripts and mark comments - Third investigator adjudicated differences
- Comments separated into categories with thematic
labels - Themes and comments sent to two other
investigators for independent second review - Final taxonomy developed by consensus
15Sample Comments Made by Patients in Depression
Focus Groups
Spirituality
- I did pray a lot. Im a Christian, and I would
pray and pray and find verses of scripture. - African-American male, age 30
Cooper-Patrick L et al, JGIM 199712431-438
16Sample Comments Made by Patients in Depression
Focus Groups
Stigma
- And I didnt want anyone to know that I was
taking this prescription. I just didnt want to
feel like I was crazy. - African-American female, age 53
Cooper-Patrick L et al, JGIM 199712431-438
17Sample Comments Made by Patients in Depression
Focus Groups
Patient-provider relationships
- This guy my doctor was just a plain old nice
guy, you knowhe was very, very sharpI thought,
whatever this guy tells me for the most part, if
it sounds sensible, Ill give it a try. - male, age 28
Cooper-Patrick L et al, JGIM 199712431-438
18Sample Comments Made by Patients in Depression
Focus Groups
Attributes of treatment Medicines
- If its gonna make me feel good, make me feel
good right away so I can get up and start doing
what I want to do. I dont want it to take a
long time to kick in. - female, age 41
Cooper-Patrick L et al, JGIM 199712431-438
19Patients Opinions Regarding the Importance of
Various Aspects of Depression Care
- Design cross-sectional survey to prioritize
items derived from focus groups - Patients 76 adults (49 white, 27 AA) aged 18-64
years - Setting Urban primary care clinics
- Data Collection demographics, past mental health
treatment, mental health status, opinions about
various aspects of depression care - Instruments Importance Questionnaire, Center
for Epidemiologic Studies Depression Scale (CES-D)
Cooper LA et al, Gen Hosp Psychiatry
200022163-173
20Importance Questionnaire
- Patients were
- Given a brief description of symptoms of major
depression - Told that depression is treatable and given the
major types of treatment - Asked to rate the importance of each item on a
5-point scale of not at all important to
extremely important for good quality depression
care - Items were grouped by category/theme
21Data Analysis
- Ranking of items by mean score and rating item
extremely important (EI) - Items selected for inclusion in Patient Attitudes
and Ratings of Care for Depression (PARC-D)
instrument based on importance ratings - Reliability and validity testing of scales
- Comparisons between AAs and Whites
22Most Important Aspects of Depression Care to
Patients
- 1. Health provider interpersonal skills
- 2. Treatment effectiveness
- 3. Treatment problems
- 4. Patient education, information, and
understanding - 5. Intrinsic spirituality (African Americans)
- 6. Financial access
- 7. Primary care provider recognition of
depression -
Cooper LA et al, Gen Hosp Psychiatry
200022163-173
23African Americans rate spirituality as more
important in depression care than whites
All p-values lt0.05
Cooper LA et al, Journal of General Internal
Medicine 200116634-638
24Acceptability of Depression Treatment Among
Primary Care Patients
- Methods cross-sectional telephone survey
- Patients 829 U.S. adult primary care patients
with major depressive disorder by the CIDI - Independent variable patient race/ethnicity
- Outcomes patient self-reported acceptability of
antidepressant medications and counseling - Analysis Logistic regression using weighted data
Cooper LA et al. Med Care 200341479-489
25Quality Improvementfor Depression Project
- Hopkins Quality Improvement for Depression
(HQID), Baltimore/Washington DC (Ford, PI) - Mental Health Awareness Project (MHAP), Los
Angeles, CA (Rubenstein, PI) - Quality Enhancement by Strategic Teaming (QuEST),
10 mid-western states (Rost, PI) - Partners in Care Depression PORT (Wells, PI)
26Views about depression differ among Whites,
Blacks, and Hispanics
- White AA Hisp
- n659 n97 n72
- I believe I need treatment 68 70 68
- Medications are effective 91 69 84
- Medications are addictive 34 56 51
- Counseling is as effective as meds 50 57 74
- Counseling brings up bad feelings 50 71 71
- Prayer heals depression 67 93 67
- Socially embarrassed 24 24 33
- Family would be disappointed 16 15 22
- Prefer same ethnicity/race provider 14 25 13
plt0.05, plt0.01, plt0.001 Cooper LA et al. Med
Care 200341479-489
27African Americans and Hispanics find medications
less acceptable as treatment for depression
Counseling, Hispanics
3.26
Whites, reference group
0.63
0.44
0.30
Counseling, AAs
Meds, AAs Meds, Hispanics
adjusted for patient demographics, illness
factors, treatment experience, social support,
life events, QID site
Cooper LA et al. Med Care 200341479-489
28Access to Health Care for Racial and Ethnic
Groups
Modified From Access to Health Care in America
(1993, Millman M, ed). Cooper LA, Hill MN, and
Powe NR. JGIM 2002 477-486
29Patient-physician communication is related to
important outcomes
- Patient recall of information
- Patient adherence
- Patient satisfaction
- Clinical outcomes
- Glycemic control
- Blood pressure control
- Pain reduction
- Depression resolution
- Roter 1988, Greenfield 1988, Kaplan 1989, Kaplan,
1995, Stewart 1998, Clever 2006
30Ethnic minorities receive poorer interpersonal
health care than whites
- African Americans, Hispanics, and Asians report
- Lower levels of trust in physicians and hospitals
Boulware 2003 - Less respectful treatment Johnson 2004
- Less participation in health care decisions
Cooper-Patrick 1999 - Minorities experience poorer communication with
physicians Johnson 2004 - The only exception is in race-concordant
relationships (where the patient and doctor are
the same race) Cooper-Patrick 1999, Cooper 2003
31Women receive better interpersonal health care
than men
- Female patients receive more information, ask
more questions, and have more partnership-building
with physicians than male patients Kaplan 1995,
McKinlay 1996 - Female physicians engage in more psychosocial
talk, partnership building and have longer visits
than male physicians Roter Hall 1993, 1998 - Studies of gender concordance (where patients and
physicians are the same gender) show mixed
results Roter 1999, Derose 2001, Schmittdiel 2000
32Race, Gender, and Partnership in the
Patient-Physician Relationship
- Design Cross-sectional telephone survey
- Subjects 1816 adults (784 W, 814 AA, 218 Other)
who had seen their MD (n65) within the past 2
weeks - Setting 32 primary care practices, large network
style managed care organization in Washington
D.C. area - Predictor variables race and gender concordant
or discordant status in patient-physician
relationship - Main Outcome patients ratings of their MDs
participatory decision-making (PDM) style
Cooper-Patrick L et al, JAMA 1999282583-589
33Measurement of Physicians Participatory
Decision-Making Style
- Patient is asked
- If there were a choice between treatments, how
often would this doctor ask you to help make the
decision? - How often does this doctor make an effort to give
you some control over your treatment? - How often does this doctor ask you to take some
of the responsibility for your treatment?
Kaplan SH et al, Medical Care 1995331176-1187
Each item contributes 33.3 points. Maximum score
is 100 points.
34Patient race, not gender, is related to
participatory decision-making
P0.007
P0.05
PDM scores range from 0-100. A higher score means
visit is more participatory. Cooper-Patrick L ,
JAMA 1999282583-589
35Physician gender, not race, is related to
participatory decision-making
P0.03
Adjusted for patients age, race, gender,
education, marital status, health status, and
length of patient-physician relationship.
Cooper-Patrick L et al, JAMA 1999282583-9
36Patients in race-concordant, not
gender-concordant, relationships rate their
physicians as more participatory
P-value NS
P0.02
Mean PDM Style Score
Adjusted for patients age, gender, education,
marital status, health status, length of the
patient-physician relationship, physician gender
(race concordant analysis) and physician race
(gender concordance analysis). Cooper-Patrick L,
JAMA 1999282583-589
37Patient-Centered Communication, Ratings of Care
and Concordance of Patient and Physician Race
- Design cross-sectional study using pre-visit and
post-visit surveys and audiotape analysis - Setting urban primary care practices serving
managed care and fee-for-service patients - Participants 458 African American and white
adult patients receiving care from 61 primary
care physicians - Patient recruitment 10 patients per MD
recruited consecutively from waiting rooms
Cooper LA, Roter DL, Johnson RL, Ford DE,
Steinwachs DM, Powe NR. Ann Intern Med
2003139907-915
38Measurement of Patient-Clinician Communication
- Content
- Biomedical exchange
- Psychosocial exchange
- Depression exchange
- Affect (Rapport-building)
- Emotional Talk - Negative talk
- Positive talk - Social talk
- Process
- Orientation (directions or instructions)
- Facilitation (includes partnership-building)
Roter Interaction Analysis System (RIAS) Roter
D, Larson S. Patient Educ Couns 200246243-51
39Examples from Communication Categories
- Biomedical exchange
- Your blood pressure is 100 over 70.
- I was in the hospital last year for ulcers.
- Psychosocial exchange
- You really need to get out and meet more
people. - I guess every marriage has its ups and downs.
- Rapport-building exchange
- This must be very hard for you.
- I hope youll be feeling better soon.
- Facilitation and Partnership-building
- Do you follow me? How does that sound to you?
40Examples of Depression Talk
- Physician
- It might just be a matter of getting you on the
right amount of antidepressant medication. - Maybewe should have you see a professional
counselor to help you sort through some of these
issues. - Patient
- I am feeling tired all the time. I like to look
on the bright side, but it seems like lately,
its harder to do. - Ive been taking Trazodone.
41Measurement of Global Affect or Emotional Tone
- Coders are asked to rate overall emotional tone
of the visit for clinicians and patients - Physician positive affect (assertiveness
interest responsiveness empathy) hurried - Patient positive affect (assertiveness
interest friendliness responsiveness
empathy) - Patient negative affect (anxiety, irritation,
depression, and emotional distress)
All global affect dimensions are coded on a
numeric scale of 1-6 (1low/none, 6high).
42The Patient-Centered Medical Interview
- Visit duration is longer
- Speech speed is lower
- Physicians are less verbally dominant
- doctor talk to patient talk ratio is close to 1
- Patient-centeredness ratio is high more
psychosocial, emotional, and partnership talk
than biomedical talk - More positive emotional tone
43Race-concordant visits are longer with more
positive patient emotional tone
plt0.05. Adjusted for patient age, race, gender,
and health status and physician gender and years
in practice.
Cooper LA et al, Ann Intern Med 2003139907-915
44Patients in Race-Concordant Relationships Rate
Their Physicians Better
Mean Score/Probability
plt0.05, plt0.01 from GEE. Analyses adjusted for
patient gender, race, age, and health status,
physician gender, years in practice, and
patient-centered communication. Cooper LA
et al, Ann Intern Med 2003139907-915
45Physicians are more verbally dominant and have
less positive emotional tone in visits with
African-American patients
Adjusted for patient age, gender, education
level, and self-rated health status and
physician gender, race, time since completing
training, and report of how well he/she knows
each patient. p-value from linear regression
with GEE. Patient and physician affect scores
are derived from audiotape coders impressions of
the overall emotional tone of the medical visit.
Johnson RL, Roter DL, Powe NR, Cooper LA. Am J
Public Health 2004 942084-2090.
46Physicians engage in less depression talk and
rapport-building with depressed African Americans
P0.07
P0.04
P0.01
P0.30
Ghods BK, Roter D, Ford DE, Larson S, Arbelaez J,
Cooper LA. J Gen Intern Med 2008 Feb 8 Epub
ahead of print
47Patient Race, Visit Length, and Coder Ratings of
Global Affect
plt0.05, plt0.01. Adjusted for patient age,
education and PCS-12 score
48Physician Ratings of Patients Health Status
The sample size is lower than the total number of
patient participants because of physician
non-response to post-visit surveys.
49Summary
- African American patients report less favorable
attitudes regarding depression treatment - Racial disparities in patient-physician
communication and physician recognition of
depression exist - To overcome disparities in depression care,
studies should test the effectiveness of - Physician communication skills training programs
that emphasize recognition and rapport-building - Culturally sensitive patient activation programs
50Blacks Receiving Interventions for Depression and
Gaining Empowerment
- Design Randomized controlled trial
- Population 30 primary care providers and 250
African American patients with depression - Setting Urban, community-based clinics in
Delaware, Maryland, and Washington DC - Interventions
- Standard quality improvement program
(disease-oriented provider academic detailing and
patient case management) - Patient-centered, culturally tailored program
(provider communication skills training and
patient case management focused on cultural
issues)
Supported by AHRQ R01HS013645, 9/30/03-8/31/08
51Bridge Study Outcomes
- Resolution of depression symptoms
- Functional status and disability days
- Guideline concordant depression care
- Health service utilization
- Patient ratings of care
- Patient-provider communication behaviors
Assessed at 6 and 12 months of
follow-up Assessed at baseline and after
provider intervention
52Conceptual Framework for BRIDGE
BRIDGE Study Interventions
Provider Needs assessment CME lunch meetings
(2) Communication skills training Feedback from
case manager
Health Care System Psychiatric consultation Collab
orative care Feedback regarding patients
status Case manager encourages more follow-up and
appointments
Patient Needs assessment Education Activation Supp
ort Active follow-up
Process of Care
Outcomes
53Study Design
Patient-Centered Intervention Physicians N15
Patient-Centered Intervention Patients N125
All clinicians are videotaped with a simulated
patient at baseline, then randomized
Standard Intervention Patients N125
Standard Intervention Physicians N15
Depression Care Manager contacts for active
follow-up up to 12 months
54Clinical Sites Partners
- Baltimore Medical Surgical Associates
- Baltimore Medical System (BMSI)
- Johns Hopkins Community Physicians
- Sinai Hospital
- Henrietta Johnson Medical Center, DE Associates
- Westside Healthcare, DE
55Recruitment
- Clinicians
- Via letter from medical director and PI
- CME credit and individualized feedback on
communication style - Organizations given incentive for MD/NP/PA
participation in research (200/clinician)
- Patients
- Approached onsite by RA
- Screener administered and visit audio-taped
- Baseline interview completed by phone within 2
weeks - Assigned to care manager
- Monetary compensation
(25 per interview3 75)
56Primary Care Clinician Intervention Features
57Examples of Provider Goals
- Improve recognition
- Evaluate depressed patients more thoroughly
example, substance abuse, anxiety - Assess suicidal ideation
- Change the antidepressant I usually use
- Identify patients cultural beliefs regarding
depression and its treatment - Elicit patients treatment preferences more often
58Interactive CD-ROM
59Patient Intervention Features
60Standard Needs Assessment
- Depressive symptoms
- Associated conditions
- Functional Status/Activities affected
- Stressors
- Social Support
- Treatment preferences
61Patient-Centered Intervention Needs Assessment
- Meaning of illness from patient perspective
- Perceptions of racial discrimination
- Literacy and language concerns
- Importance of spirituality in coping and care
- Specific treatment concerns regarding
antidepressants or counseling - Financial concerns
- Role of stigma
- Relationships with health professionals
62Standard InterventionPatient Education Materials
63Patient-Centered Intervention Patient Education
Materials
- Brochure
- Book
- Videotape
- Prayer card
- Bridge Study calendar
only if patient is spiritually oriented and/or
receptive
64Primary Care Clinician Enrollment(June 2004
March 2006)
108 Clinicians Contacted
9 No Response
99 Clinicians Responded
63 Refusals
36 Clinicians Randomized
27 Clinicians with patients enrolled in study
65Demographic Characteristicsof Primary Care
Clinicians (N27)
66Primary Care Clinician Elicitation of Depressive
Symptoms at Baseline
67Primary Care Clinician Depression Proficiencies
at Baseline (N27)
- Treatment of Depression
- Proficiency of clinicians who did
- Recommend counseling/
- referral or future referral 81
- Discussed psychotropic
- medication use 89
- Asked knowledge/ beliefs
- regarding depression 15
- Asked treatment/ concerns
- regarding depression 96
- Ask for commitment to plan 74
68Primary Care ClinicianPost-Intervention
Proficiencies
- Standard Patient Centered
- (n31) (n25)
- Depression Proficiency elicited
elicited - gt 5 depression symptoms 63 81
- Depressed Mood 52 80
- Sleep disturbance 55 76
- Appetite or weight change 35 60
- Fatigue 26 36
- Suicidal Ideation 68 64
- Diminished interest 45 44
- Impaired concentration 29 28
- Psychomotor 10 12
- Guilt 13 8
plt0.05
69Primary Care ClinicianPost-Intervention
Proficiencies
- Standard Patient Centered
- Depression Proficiency (n31) (n25)
-
- Asked knowledge/beliefs 0 4
- Asked treatment concerns 23 12
- Recommended counseling 77 64
- Discussed medication use 77 44
- Ask for commitment to plan 3 20
plt0.05
70Patient Enrollment(October 2005 - August 2006)
1,486 patients Approached in waiting rooms
873 (59) Agreed to be screened
457 not depressed 46 getting SMH care 68 CAGE
positive
231 (26) Eligible by screener
22 Unable to locate
132 (90) Eligible by CIDI and randomized
71Demographic Characteristics of Patients at
Baseline (n132)
72Clinical Characteristics of Patients at Baseline
(n132)
73Patients Treatment Preferences
- If you had to choose between taking
antidepressant medicine daily for 6-9 months or
going to counseling weekly for at least 3 months,
which would you prefer?
74Patient Attitudes and Ratings Towards Depression
Care
75Satisfaction with Care and Trust in Providers
76Patients Decision-making Preference
77Patient Follow-up Status
78Patient Intervention Contacts
Refused during initial call or at some point
during the 12 month intervention
79Treatment status at 6 and 12 months by
intervention status
80Depression status at 6 and 12 months by
intervention status
81Conclusions
- Recruiting primary care clinicians and depressed
African American patients from busy urban
community-based clinics to participate in quality
improvement interventions for depression is
challenging, but feasible - Both intervention approaches show similarly
modest improvements in clinician assessment of
depression and patient depression symptoms
82Conclusions
- Standard QI approaches may lead to greater use of
antidepressant medications among clinicians and
their African American patients - Patient-centered care may lead to greater
long-term improvements in depression status - We are conducting an 18-month follow-up
- Future analyses will focus on processes of care
and patient experiences and make comparisons to
clinical effects seen in other randomized trials
83Implications
- Research Examine more links among clinician and
patient attitudes, behaviors, and health
outcomes - Health Professional Education - employ
patient-centered communication skills programs
that emphasize rapport building and affective
dimensions and enhance awareness of bias and
intercultural skills - Clinical Practice - implement patient activation
programs improve scheduling, increase time to
build rapport and develop continuity of care - Policy - increase numbers of underrepresented
ethnic minorities among health professionals
parity in mental health coverage reimbursement
for case management and collaborative care
84Research and Clinical Team
- Principal Investigator
- Lisa A. Cooper, MD, MPH
- Co-Investigators
- Daniel E. Ford, MD, MPH
- James M. Gill, MD, MPH
- Susan Larson, MHS
- Annelle B. Primm, MD, MPH
- Debra L. Roter, DrPH
- Elias K. Shaya, MD
- Nae-Yuh Wang, PhD
- Research Program Coordinator
- Bri Ghods
- Depression Case Managers
- Karen Kemp, MSW
- Dawna McGlynn, LCSW-C
- Data Collection Unit
- Battelle Memorial Institute
- Other collaborators
- Junius Gonzales, MD
- Joseph Gallo, MD
- Charlotte Brown, PhD
- Jeanne Miranda, PhD