Transcript RWC/PPW Cross-site Outcome: Factors associated with

Race Targeting and Race Congruence as
Factors Affecting Outcomes in Long Term
Residential Substance Abuse Treatment
for Women
Xiaowu Chen, M.D., Caliber Associates
Ken Burgdorf, Ph.D., Caliber Associates
Tracy Roberts, M.P.A., Caliber Associates
James M. Herrell, M.P.H., Ph.D., CSAT
130th APHA Annual Meeting & Exposition
Philadelphia, November 12, 2002
*Study conducted under Contract 270-97-7030 funded by the Center for Substance Abuse Treatment,
Substance Abuse and Mental Health Services Administration, U.S. Department of Health and Human Services. Center for Substance Abuse Treatment,
5600 Fishers Lane, Rockwall II, Suite 740, Rockville, Maryland 20857, 301/443-5052. Contents are solely the responsibility of the authors and do not
necessarily represent the official views of the agency.
CSAT’s RWC & PPW Programs
• Demonstration projects funded by the Center for
Substance Abuse Treatment (CSAT). Projects provide
model substance abuse treatment that:
– Is long-term residential (6 months or 12 months)
– Is gender-specific, targeted to pregnant and
parenting women
– Includes provision for in-residence care of clients’
children
– Provides comprehensive, culturally appropriate
services to women and children.
Cross-site Study Design
• The study coordinated collection (by grantees) of
standardized sets of data elements concerning
client and child characteristics at several points:
– At admission to treatment
– Quarterly throughout treatment
– At discharge from treatment
– Six months after discharge from treatment.
• The cross-site data collection system was
instituted in October 1996 and continued through
March 2001.
The Sample
• The sample consists of 32 projects that met the
threshold follow-up rate (50%).
• The follow-up rate for this sample is 70%, and
includes 1181 clients.
• The data are weighted by length of stay (LOS) to
adjust for underrepresented short-stay clients.
• Limitations:
– self-report information mostly not validated
– no “placebo” group is available.
Outcome Measures
• Three summary indices of clients’ response to
treatment:
– Whether clients remained in treatment
beyond the first 30 days
– Whether clients completed their treatment
program
– Whether clients successfully maintained
alcohol and drug abstinence through the six
month follow-up period after discharge
This Analysis
• Examine ways that race may have been a factor
affecting clients’ response to this kind of longterm residential treatment:
– Are there overall differences among race
groups in key outcomes?
– Do clients have better outcomes if treated in
projects that target their race?
– Do clients have better outcomes if they are of
same race as most other clients in the project?
Methods
• Pearson Chi-square tests were performed to
examine overall associations between race
measures and the outcomes.
• Hierarchical Linear Models (HLM) were
constructed to assess the factors associated with
the outcomes at the client level and project level.
Summary of Client Characteristics
(n=1181)
Variable
Race
Education
Level
Af-Amer
White
Other
H.S.
<H.S.
Age (years) 18-25
26-35
36+
Pregnant
Yes
No
Percent
40
32
28
53
47
24
55
21
26
74
Variable
Child in tx
Level Percent
Yes
66
No
34
Primary drug Cocaine
11
Crack
40
Heroin
8
Alcohol 14
CPS involved Yes
55
No
45
Abstinent
Yes
60
post-tx
No
40
Treatment Outcomes by Race
N
Race
African Am. 468
Hispanic
170
Native Am.
99
White
377
LOS>30d
(%)
80
75
87
79
Completed Abstinent
(%)
(%)
45
36
48
44
Results are not significant at .05 level.
62
56
57
59
Race Targeting and Treatment Outcomes
N
Target
Group
African Am.
Hispanic
Native Am.
*p<0.05
468
170
99
LOS>30d
(%)
yes|no
77|82
70|90*
88|77
Completed Abstinent
(%)
(%)
yes|no
yes|no
51|42
30|56*
50|33
64|62
52|68
54|74
Project Majority and Treatment Outcomes
for African Americans (n=468)
Majority
African Am.
Different
No majority
P
LOS>30d
(%)
78
94
82
ns
Completed
(%)
45
47
44
ns
Abstinent
(%)
82
44
58
.03
Project Majority and Treatment
Outcomes for Hispanics (n=170)
Majority
Hispanic
Different
No majority
P
LOS>30d
(%)
71
76
82
ns
Completed Abstinent
(%)
(%)
28
56
43
35
52
69
ns
.03
Project Majority and Treatment
Outcomes for Native Americans (n=99)
Majority
Native Am.
Different
No majority
P
LOS>30d
(%)
87
100
78
ns
Completed Abstinent
(%)
(%)
50
54
53
77
35
67
ns
ns
Project Majority and Treatment
Outcomes for Whites (n=377)
Majority
White
Different
No majority
P
LOS>30d
(%)
89
76
73
.003
Completed Abstinent
(%)
(%)
51
59
43
59
39
59
.006
ns
HLM: Abstinence in Post-discharge
Variable
Race
Description
African Am.
Hispanic Am.
Native Am.
White
Race congruence
no majority
different
same as client
LOS
10 days increase
Self-help group
yes vs. no
Client-counselor ratio 0-8 vs. 8+
OR
1.41
1.00
1.22
1.00
1.40
0.72
1.00
1.04
1.75
1.81
P
.0899
.9941
.5988
.1762
.2075
<.0001
.0003
.0339
Results adjusted for age, education, parenting status, primary drug, CPS or CJS involvement,
therapeutic approach, time spent with clients, family involvement, and project location.
HLM: Abstinence in Post-discharge
• Sixty percent of all clients who entered the program
reported complete AOD abstinence for 6 months
after discharge.
• There was no statistically significant difference in
abstinence between African American, Hispanic,
Native American, and White clients.
• Whether or not the client stayed in projects in which
her race was a majority did not affect her AOD
abstinence in 6 months after discharge.
HLM: Abstinence in Post-discharge
• Variation in treatment success (abstinence vs.
relapse) was predictable. Results suggest
important variables are:
– At the project level: intensity of clinical contact
between clients and counselors.
– At the client level: having a longer length of
stay and participating in self-help groups after
discharge were prognostically positive;
negative prognostic indicator was: having pretreatment involvement with CPS.
HLM: Treatment Completion
Variable
Race
Description
African Am.
Hispanic Am.
Native Am.
White
Race congruence
no majority
different
same
With children
yes vs. no
Age
one year increase
Client-counselor ratio 0-8 vs. 8+
OR
1.28
1.40
2.85
1.00
1.85
1.08
1.00
4.48
1.65
1.61
P
.2317
.2549
.0307
.1896
.7631
<.0001
<.0001
.3093
Results adjusted for education, primary drug, CPS involvement, mental health, therapeutic
approach, time spent with clients, project size, and project location.
HLM: Treatment Completion
• Clients were more likely to complete treatment if
they
– were Native American;
– brought children into treatment;
– were older.
• Whether or not the client stayed in projects in
which her race was a majority did not change the
odds of treatment completion.
• None of the project level variables were associated
with treatment completion.
Summary
• The study did not find any major differences in
treatment retention, completion, or post-treatment
AOD abstinence among race/ethnic groups.
• Clients’ race congruence was not associated with
treatment retention, completion, or AOD abstinence.
• A variety of client level variables were significantly
associated with treatment completion and AOD
abstinence, including bringing children to treatment,
pre-treatment CPS involvement, LOS, and
participating in self-groups.