an adaptive reinforcement-based treatment (rbt) intervention …€¦ · · 2016-02-22an adaptive...
TRANSCRIPT
An Adaptive Reinforcement-Based Treatment (RBT) Intervention for Pregnant
Substance Dependent Women
Michelle Tuten, Ph.D.
Assistant Professor Department of Psychiatry and Behavioral Sciences and
Johns Hopkins University School of Medicine
HIV Methods and Intervention Science Meeting September 12-13th 2013
2
Acknowledgements
National Institute on Drug Abuse
R01DA14979
Participants
Staff at the Center for Addiction and Pregnancy
Hendree Jones (original study PI: slides
adapted from 2011 CPDD workshop)
Collaborators Susan Murphy, Pierre
Alexander, Margaret Chisolm (current PI)
3
Outline
1. Complexities of the Problem
2. Clinical Setting
3. Role of Behavioral Treatment
4. HOME II Study
1. Complexities of the Problem
Alcohol and tobacco substances are the most commonly abused during pregnancy
Minority but significant number of pregnant women (also use) illicit drugs during pregnancy
Drug addiction begins before pregnancy and is the result of complex past and current factors
National Survey on Drug Use
and Health 2008/9
Past Month Use
4
5
1. Complexities of the problem
Pregnancy is a window of opportunity for behavior change
TREATMENT
6
1. Complexities of the problem
Issues Facing Pregnant Drug Users
• Exposure to violence, trauma
• Generational drug use
• Lack of formal education
• Lack of work history or skills
• Legal involvement
• Multiple drug exposure • Limited parenting skills • History of abuse/neglect • Psychiatric issues • Unstable housing • Lack of social support • Medical issues • Nutrition deficits
• Address Barriers to Care
• Improve maternal and infant outcomes
• Conduct clinical research
7
2. Clinical setting
Center for Addiction and Pregnancy mission statement:
Comprehensive Care
• Interdisciplinary approach – Substance abuse
counseling – Psychiatry – Obstetrics – Pediatrics – Nursing
• Multiple modalities
– Medically-assisted withdrawal
– Methadone stabilization
8
2. Clinical setting
No
CAP Treatment (n=100) (n=46)
Clinical measure:
Prenatal visits 8 4
EGA (mean week) 39 35
% positive at delivery 37 63
Infant birth weight (grams) 2934 2539
Apgar scores (1 minute) 8 7
% NICU use 10 26
Length of hospital stay (days) 7 39
All group comparisons are significant: p ≤ .05 (Data adapted from Svikis et al., 1997) 9
2. Clinical setting
Outcome research: cost effectiveness
Investing in CAP drug treatment services resulted in net savings of $4,644 in NICU costs.
Although CAP is cost-
effective, many women continue to drop out of treatment prematurely and/or do not respond to treatment
Guiding Principles
• Drug addiction occurs in absence
of alternative positive reinforcers
• Drug use “makes sense” in that it
meets certain behavioral functions
for the individual
• Drug use behavior can be modified
using operant reinforcement
– Goal of treatment is for non-
drug using behaviors become
more reinforcing than drug
use
10
3. Role of behavioral treatment
Vocational
Assistance/
Job Club
Peer
Support/
Social Club
Functional
Drug Use
Assessment
Behavior
Graphing
Intensive
Outreach
Patient
Personalized
Feedback
Recreational
Activities
Functional assessment (FA)
Personalized feedback (MI adaptation)
Social reinforcement (in the form of weekly Social Club)
Behavior graphing
Vocational assistance/job club
Recreational activities
Intensive outreach
Tangible reinforcers
3. Role of behavioral treatment
11
Treatment Elements
Treatment Elements
12
Functional
Assessment (FA) of
drug use
Behavior
Graphing
FA: when, where and
why of drug use
Constellation of
behaviors that compete
with drug use
Graph these target
behaviors
3. Role of behavioral treatment
12
Treatment Objectives
Treatment Elements
Treatment
attendance
Recreational
Activities
Vocational
Activities
Medication
compliance
13
Example of Close Behavioral Observation: DRUG FREE DAYS
0
12
3
4
56
7
8
910
11
12
1314
15
1 3 5 7 9 11 13 15
Days
Ab
sti
nen
ce (
yes/n
o)
3. Role of Behavioral Treatment
Allow for connections between
behaviors (if you do this, drug use
occurs, if you do this, drug abstinence
occurs)
Larger goals are broken into
smaller goals
Positive reinforcement given
for successes
Good
Job!
Good
Job!
Nu
mb
er
of
Ap
pli
cati
on
s
14
0
2
4
6
8
10
12
14
16
wee
k 1
wee
k 2
wee
k 3
wee
k 4
Number of Job
Applications
Example Of Behavior Tracking
_______________
Employment
obtained
Goal= 9
applications/week
3. Role of Behavioral Treatment
15 15
Examples: Participant Treatment Progress
D
ays D
rug F
ree
3. Role of Behavioral Treatment
0
20
40
60
80
100
RBT Control
Opioid and Cocaine Negative
0
20
40
60
80
100
1-month 3-month 1-month 3-month%
Po
siti
ve U
rin
e S
amp
les
RBT (n=72)
Control (n=128)
16 16
*
Opioids Cocaine
3. Role of Behavioral Treatment
(Gruber et al., 2000; Jones et al., 2005; Jones 2011; Tuten et al., 2012 * All treatment condition comparisons
are significant at p ≤ .05
3 randomized trials showing
RBT’s efficacy in nonpregnant
patients (1 month, 6 months, 12
months follow-up)
1RBT adaptation and test in
pregnant patients
As with any intervention, not all
RBT-treated participants
attended treatment or reduced
drug use, suggesting room to
improve patient response
0
20
40
60
80
100
RBT Control
Opioid and Cocaine Negative
0
20
40
60
80
100
RBT UC
Me
an D
ays
Delivery
*
*
*
*
* *
17
3. Role of Behavioral Treatment
We identified two important questions regarding RBT
for pregnant women:
Would variants of RBT in terms of their intensity and
scope be effective?
Would patients who differed in terms of their treatment
compliance be differentially responsive to these variants
of RBT?
18
4. A SMART trial at HOME
Inclusion Criteria:
Evidence of cocaine and/or opiate use
Treatment entry at or before 32 weeks EGA with singleton fetus
Completion of the eight-day residential detoxification stay Exclusion Criteria:
Age 17 or younger
Geographical Constraints
Severe medical or psychiatric concomitant condition interfering with treatment or needing hospitalization
HOME II Eligibility Criteria
19
• Birth weight
• Head Circumference
• Length of Hospital/NICU Stay
• Urine Toxicology at Delivery
• Physical Birth Parameters
• Neonatal Complications
Primary Outcomes- Maternal: Treatment Completion (Delivery)
Heroin Use
Cocaine Use
Secondary Outcomes- Neonatal :
4. A SMART trial at HOME
Decisions for Designing the Trial:
How to choose a design structure to answer
questions of interest:
Initial randomization conditions?
Second randomization to what?
Which treatment response characteristics:
Was strongly related to outcome?
Would allow reasonable tailoring?
Length of time to allow for assessment of the
tailoring variable
Pregnancy time-limited window
Solutions:
Initial Randomization: Compare
TAU RBT to a potentially more cost
effective form of RBT intervention
Treatment Compliance as the
tailoring variable: Failure to comply
with treatment soon after treatment
entry the biggest reason for failure to
complete treatment
Assess compliance during the first
two weeks following treatment
entry
20
4. A SMART trial at HOME
21
• HOME II intervention adjusts in intensity or scope following patients’ initial treatment compliance or non-compliance.
• It is expected that RBT in an adaptive intervention format will optimize maternal treatment outcomes for both early compliant and early non-compliant participants by matching treatment to patient needs.
Planned Sample Size 4. A SMART trial at HOME
22
tRBT: Treatment-as-usual RBT
• All elements of treatment-as-usual RBT
rRBT: reduced RBT
– Key RBT elements are provided at a reduced scope to examine a version of RBT that might be more in line with community practice limitations
All
non-methadone
CAP Patients
tRBT
rRBT
Random
Assignment
5X/week: Individual sessions
3X/week: Recreation
3X/week: Job Club
3X/week: Individual sessions
1X/week: Recreation
1X/week: Job Club
4. A SMART trial at HOME
23
Tailored Treatment Randomization
Two-week window for treatment response
• Early compliant participants
randomized to either the same intensity of treatment or a decreased intensity or scope of RBT treatment
• Early non-compliant participants randomized to receive either the same treatment or a greater intensity or scope of RBT
Early treatment non-compliance:
(a) a missed unexcused treatment day,
(b) a positive opioid or cocaine urine specimen,
(c) or self-report of use of either drug.
4. A SMART trial at HOME
24
Increased intensity
eRBT: Enhanced RBT
All tRBT elements, plus:
Home visits to re-engage participant/deliver
therapy
Immediate re-admission to the residential
unit or community recovery housing as “time
out” from drug use
Decreased intensity
aRBT: Abbreviated RBT
Most reduced version of RBT:
Graphing drug abstinence
Outreach limited to follow-up call
Individual sessions once a week
Recreation and social club once a month
4. A SMART trial at HOME
25
Primary Aim:
Relative efficacy of providing
continued treatment-as-usual RBT in comparison to continued reduced RBT to both early-compilers and early non-
compliers throughout the trial.
►Addresses the question of whether it is necessary to provide treatment-as-usual RBT, or if it is possible to
successfully treat patients with reduced RBT, regardless of the
patient’s level of compliance.
4. A SMART trial at HOME
26
Secondary Aims:
Early non-compliers: Relative efficacy
of transitioning to a more intensive level of treatment for the TAU RBT and the reduced RBT conditions.
►Answers questions regarding the
relative importance of the initial level of care in determining the efficacy of transitioning to a higher level of care
for early-non compliers.
4. A SMART trial at HOME
27
Secondary Aims (cont.):
Early-compliers: relative efficacy of
transitioning to a less intense level of treatment for within the initial TAU RBT and reduced RBT treatment
conditions.
►Answers questions about the importance of the initial level of RBT treatment in determining success in
subsequently reducing the level of RBT treatment in early-compliers.
4. A SMART trial at HOME
28
Secondary Aims (cont.):
Relative efficacy for early-complier and
early-non-compliers who begin and continue in treatment-as-usual RBT or who begin and continue in reduced
RBT, respectively.
►Answers the question about the relative decrement in treatment success that occurs as a result of an
early failure to comply with the initial demands of treatment.
4. A SMART trial at HOME
Decrease
Intensity Decrease
Scope
All non-methadone CAP Patients
N= 230
tRBT n =114
rRBT
n =116
Early
Compliant n=48
Early
Non- Compliant
n=61
tRBT
n=30
Early
Non- Compliant
n=53
Early
Compliant n=56
rRBT
n=28
aRBT
n=28
rRBT
n=24
tRBT
n=24
tRBT
n=26
rRBT n=27
eRBT
n=31
Tailoring Variable Randomization:
Treatment Compliance
Random Assignment
Increase
Intensity
Increase
Scope
29
Current Enrollment
4. A SMART trial at HOME
30
Challenges in Conducting the Trial
• Recruitment
• Participant adherence to assessment protocol • Maintenance of distinct treatment conditions
• Adherence of the clinical staff to the protocol
• Changing clinical programming
New Insights Resulting from the Trial
• Patient acceptance of the tailoring treatment randomization
• Better-than-expected distribution of patients into tailoring treatment randomization
4. A SMART trial at HOME
31 31
What we’ve learned from process:
• Feasible and desirable
• Well-defined tailoring/response definitions
• Distinct and theoretically sound conditions
• Procedures to maintain conditions (and documented deviations)
4. A SMART trial at HOME: SUMMARY