Dropout in Treatments for PTSD and Comorbid SUD ALEXANDER C. KLINE, - - PowerPoint PPT Presentation

Dropout in Treatments for PTSD and Comorbid SUD

ALEXANDER C. KLINE, PH.D. VA SAN DIEGO HEALTHCARE SYSTEM

Do Not Disseminate Without Permission from the Author

Local thank you’s

• Track directors Tamara Wall, Ph.D. and Scott Matthews,

M.D.

• Mentor Sonya Norman, Ph.D. and her lab

Treatments for PTSD+SUD

• Posttraumatic stress disorder (PTSD) and substance use

disorder (SUD) frequently co-occur (Pietrzak et al., 2011)

• Increased severity, greater risk of many negative correlates

(Norman et al., 2018)

• PTSD treatments are effective for patients with comorbid

substance use (e.g., Roberts et al., 2015)

• Trauma-focused treatments regarded as first-line

interventions (Hamblen et al., 2019)

• There is an increasing push to provide first-line

treatments for patients with PTSD and comorbid SUD

(VA/DoD, 2017)

• But patient retention remains a significant challenge

Impact of dropout

• Dropout is problematic for number of reasons
• 1. Limits effectiveness
• Dropout generally a bad clinical outcome
• Sufficient treatment “dose” necessary for optimal outcomes

(e.g., Holmes et al., 2019)

• 2. Limits reach (Becker et al., 2004; Cook et al., 2014; Osei-Bonsu et al., 2017)
• Concerns regarding substance use, symptom exacerbation,

and increased dropout risk have limited their implementation

• Particularly relevant to exposure-based / trauma-focused

therapies

Dropout research in PTSD+SUD Tx

• Dropout not unique to PTSD+SUD
• But often higher in these trials (Roberts et al., 2015) relative to

PTSD alone (Lewis et al., 2020) and other disorders (Swift et al., 2014)

• In PTSD treatment, substance use has been linked to

attendance (e.g., Bedard-Gilligan et al., 2018)

• Limited understanding of precise facets of use that may be

linked to dropout

• Chronicity? Severity? Craving? Frequency? Co-occurrence with
• ther substances?
• What should providers target? Difficult to identify patients at high

risk for dropout

• Patients with comorbid SUD have typically been excluded from

PTSD clinical trials (Leeman et al., 2017)

Study rationale and aims

• Current study used data from a recent randomized clinical trial
• f an integrated exposure therapy vs. integrated coping skills

therapy for treatment of Veterans with comorbid PTSD+AUD

(Norman et al., 2019)

• Questions:
• Can we identify specific SUD-related patient characteristics

associated with attendance?

• Are these characteristics associated with attendance

differentially by treatment condition?

Current study: Method

• 119 Veterans with comorbid PTSD+AUD
• Average age = 41.61, SD = 12.59
• Mostly male (89.9%)
• Varying index traumas, with ~60% combat-related
• Integrated exposure therapy (Concurrent Treatment of PTSD

and Substance Use Disorders Using Prolonged Exposure [COPE]; Back et al., 2015)

• Integrated coping skills therapy (Seeking Safety [SS]; Najavits, 2002)
• Randomized to receive 12 sessions of COPE or SS
• Provided opportunity to extend up to 16 sessions
• In main trial, greater reductions in PTSD symptoms in COPE

compared to SS (d = 0.41)

Current study: Method

• Patients completed baseline measures:
• CAPS-5 (Weathers et al., 2013)
• SCID-IV-TR (First et al., 2002)
• Penn Alcohol Craving Scale (PACS; Flannery et al., 1999)
• Timeline Follow-Back (TLFB; Sobell et al., 1992)
• SUD predictors of interest:
• AUD duration (SCID)
• AUD severity (SCID)
• % of heavy drinking days in 90 days prior to treatment (TLFB)
• % of drug use days in 90 days prior to treatment (TLFB)
• % of abstinence days in 90 days prior to treatment (TLFB)
• Craving (PACS)

Current study: Data analysis

• Covariate checks:
• Predictors of attendance from prior PTSD+SUD trials
• PTSD severity, age, education, trauma type (Belleau et al., 2017;

Szafranski et al., 2017; Zandberg et al., 2016)

• Living arrangements (i.e., controlled environment)
• Regression models with sessions attended as DV
• Treatment condition in all models
• Treatment interaction term entered at step 2
• Sensitivity analyses:
• Excluded early responders from analyses
• Recoded treatment “extenders” (i.e., attendance of 13+

sessions) as attending 12 sessions (i.e., completing treatment)

Results

• Veterans attended an average of 9.13 (SD = 4.76) sessions
• Higher session attendance in SS compared to COPE, d = 0.79
• More veterans elected to extend treatment past 12 sessions in

SS (44.6%) compared to COPE (20.6%), p = .009

• Covariate checks all unrelated to attendance
• Trauma type
• Age
• Education
• PTSD severity
• Living arrangements

Results

• Across both treatments, two baseline SUD characteristics associated

with fewer attended sessions in the trial

• Higher % of heavy drinking days, β = -.23, p = .011
• SCID AUD severity, , β = -.21, p = .019
• AUD duration (p = .56), % days abstinence (p = .57), % drug use days

(p = .83) unrelated to attendance

• Treatment condition moderated relationship between craving and

attendance, β = .17, p = .057

• Higher craving predicted less attendance in SS, β = -.31, p = .02
• No relationship in COPE, β = .14, p = .28
• Sensitivity analyses accounting for early treatment completers and

extension past treatment completion yielded equivalent findings

Discussion

• Significant differences in attendance between the two

treatments, but do not appear to be explained by substance use

• Several facets of baseline substance use (heavy drinking, SCID

severity) were modestly associated with attendance

• These relationships were not exclusive to exposure (COPE)
• Greater craving associated with worse attendance in coping

skills therapy (SS), but not exposure therapy (COPE)

• Notable what was NOT linked to attendance:
• Abstinence
• Living arrangements
• Drug use
• AUD duration

Discussion

• Limitations:
• Generalizability given predominantly male veteran sample
• Despite randomization, higher baseline craving in SS

compared to COPE (d = 0.45)

• SS often delivered in group format or 60-min sessions,

rather than 90-min individual sessions

• Future steps:
• Are there within-treatment signals tied to dropout?
• Investigating the how and why of dropout
• Are changes in session by session variables (e.g., PCL, PHQ,

substance use) during treatment as predictors of dropout?

Discussion

• Clinical targets for providers
• Discussions with patients on how alcohol use may impact

motivation or ability to attend therapy

• Findings reinforce recommendations that SUD should not

impede patients with PTSD+SUD from getting first-line interventions

• Patients with baseline substance use likely to benefit from first-

line PTSD treatments

• Significant reductions in PTSD symptoms and substance use,

and do not show high rates of relapse or symptom exacerbation

(Hien et al., 2010; Norman et al., 2019; Roberts et al., 2015; Tripp et al., 2020; Tripp et al., in press)

• SUD not a contraindication
• Abstinence should not be required before beginning PTSD

treatment