Psychology of Addictive Behaviors © 2017 American Psychological Association

2018, Vol. 32, No. 1, 1–15 0893-164X/18/$12.00 http://dx.doi.org/10.1037/adb0000330

A Randomized Trial of Female-Specific Cognitive Behavior Therapy for

Alcohol Dependent Women
Elizabeth E. Epstein, Barbara S. McCrady, Kevin A. Hallgren, Sharon Cook, Noelle K. Jensen,
and Thomas Hildebrandt
Rutgers, The State University of New Jersey

This study compared Female-Specific Cognitive Behavioral Therapy (FS-CBT) to evidence-based,

gender-neutral CBT (GN-CBT; Epstein & McCrady, 2009) for women with alcohol use disorder (AUD).
Women (N ⫽ 99) with AUD, mean age 48, were randomly assigned to 12 outpatient manual-guided
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

sessions of FS-CBT (n ⫽ 44) or GN-CBT (n ⫽ 55). Women were assessed at baseline and 3, 9 and 15
This document is copyrighted by the American Psychological Association or one of its allied publishers.

months after baseline for drinking and for specific issues common among women with AUD. A FS-CBT
protocol was developed that was discriminable on treatment integrity ratings from GN-CBT. No
treatment condition differences were found in treatment engagement, changes in drinking, alcohol-related
coping, abstinence self-efficacy, motivation to change, or constructs directly targeted in FS-CBT
(sociotropy, autonomy, depression, anxiety). Women in both conditions were highly engaged and
satisfied with treatment, and reported significant reductions in drinking and changes in desired directions
for all other variables except social support for abstinence. In the year following treatment, women in the
FS-CBT but not in the CBT condition reported an increase in percentage of abstainers in their social
networks (0.69% per month, SE ⫽ 0.21, p ⫽ .002). The value and appeal of female-specific programming
in AUD treatment has been established in the wider literature (Epstein & Menges, 2013), and the current
study provides support for the use of the Female-Specific Cognitive Behavioral Therapy (FS-CBT)
manual as an option that may yield outcomes similar to standard gender-neutral CBT for women with
AUD. Future research should examine whether FS-CBT enhances treatment utilization for women.

Keywords: alcohol use disorder, women, female-specific therapy, cognitive– behavioral therapy,
randomized trial

Supplemental materials: http://dx.doi.org/10.1037/adb0000330.supp

Males and females with alcohol use disorder (AUD) differ in
etiology, mortality, and course of the disorder (Epstein & Menges,
2013). Women with AUD also have a clinical presentation distinct
from men with AUD. For instance, women report higher comor-
bidity of mood, anxiety, posttraumatic stress, eating, and person-
ality disorders (Rosenthal, 2013). Women are more likely to drink
This article was published Online First November 20, 2017.
Elizabeth E. Epstein, Barbara S. McCrady, Kevin A. Hallgren, Sharon
Cook, Noelle K. Jensen, and Thomas Hildebrandt, Center of Alcohol
Studies, Rutgers, The State University of New Jersey.
Barbara S. McCrady is now at the Center on Alcoholism, Substance
Abuse, and Addictions, University of New Mexico. Kevin A. Hallgren is
now at the Department of Psychiatry and Behavioral Sciences, University
of Washington School of Medicine. Thomas Hildebrandt is now at De-
partment of Psychiatry, Icahn School of Medicine at Mount Sinai.
Some of the data or ideas in the article were presented at the Research
Society on Alcoholism, 2011 and the Association for Behavior and Cog-
nitive Therapies, 2009 and 2012 conferences. The authors are grateful for
the assistance of research assistants, therapists, and students. This research
was supported by NIAAA Grants R01AA007070 and K01AA024796.
Correspondence concerning this article should be addressed to Elizabeth
E. Epstein, who is now at the Department of Psychiatry, University of
Massachusetts Medical School, Biotech One, 365 Plantation Street,
Worcester, Massachusetts 01605. E-mail: elizabeth.epstein@umassmed
.edu
in response to relationship difficulties and negative emotional
triggers (Abulseoud et al., 2013), and to drink alone, in secrecy,
and daily (Zweig, McCrady, & Epstein, 2009). Women are more
likely to have social support networks that include family members
and romantic partners with SUD (Leonard & Homish, 2008) that
do not support the development and maintenance of recovery
(McCrady, 2004).
Treatment utilization for alcohol use problems is lower for
women than for men and women may be more likely to seek help
if single-gender treatment is offered (Cucciare, Simpson, Hoggatt,
Gifford, & Timko, 2013; Lewis et al., 2016), however, female-
segregated treatments may be efficacious only if they include
female-specific programming (Epstein & Menges, 2013). There
currently are few evidence-based AUD treatment protocols with
female-specific programming (Heslin, Gable, & Dobalian, 2015).
Documented gender differences, increasing rates of AUD in
women (Breslow, Castle, Chen, & Graubard, 2017), and the im-
pact of AUD on women’s health, underscore the need to address
female-tailored treatment options that enhance accessibility, en-
gagement, and efficacy of treatments that were developed with
predominantly male samples.
Treatment Considerations for Women With AUD
Women with AUD face unique barriers to seeking help, and
fewer women (15%) than men (23%) seek AUD treatment in their

1
 2 EPSTEIN ET AL.
lifetime (Dawson, 1996). For those who do seek help, women’s
and men’s AUD treatment outcomes generally are similar when
women are in mixed gender or gender-neutral treatment programs
(Greenfield, Brooks, et al., 2007; Greenfield, Pettinati, O’Malley,
Randall, & Randall, 2010). Many relapse antecedents are more
prevalent in women than men, including being alone, negative
affect, interpersonal distress, and relationship distress (Walitzer &
Dearing, 2006). Mediators of treatment on AUD outcomes may
also differ by gender; likely female-specific mechanisms of change
include alleviation of negative affect, enhanced coping skills and
self-care, improved interpersonal functioning (Velasquez & Stotts,
2003), and greater emotion regulation (Ashley, Marsden, & Brady,
2003; Timko, Finney, & Moos, 2005). These differences in relapse
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
antecedents and mediators suggest that treatments tailored to wom-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
en’s unique concerns might yield more positive outcomes than
gender-neutral programs.
Treatment approaches can be gender-customized in modality
(female-only vs. mixed-gender), gender-sensitivity (considering
women’s issues but not designed for women only), and/or female-
specificity (content designed for women only; Greenfield &
Grella, 2009). Components of AUD treatment have been associ-
ated with treatment engagement and positive outcome for women,
including topics such as self-efficacy, life coping skills (Ashley et
al., 2003; Connors & Walitzer, 2001), and interpersonal and emo-
tional cues as antecedents for drinking (Zweig et al., 2009); treat-
ing comorbid internalizing disorders (Haver & Gjestad, 2005);
enhancing a sense of self, personal agency and social support
(Sword et al., 2009); enhancing healthy relationships (Covington,
2002); and increasing social support for abstinence (Litt, Kadden,
& Tennen, 2015). Randomized controlled trials (RCT) of treat-
ments tailored for dual diagnosis samples of women with AUD
and specific comorbid disorders have shown promise (Gamble et
al., 2013; Hien, Litt, Cohen, Miele, & Campbell, 2009), and a
Swedish RCT showed better outcomes for female-tailored inpa-
tient treatment (Dahlgren & Willander, 1989; Gjestad, Franck,
Lindberg, & Haver, 2011). However, there are very few RCTs of
female-specific treatments in outpatient settings or for general
samples (i.e., not dual diagnosis) of women with AUD. One
exception is a recent study comparing a female-only group treat-
ment for substance use disorders, Women’s Recovery Group
(WRG), to mixed-gender group drug counseling. WRG yielded no
treatment condition differences during treatment, but superior
drinking outcomes at the 6-month follow-up in a small pilot study
(Greenfield, Trucco, McHugh, Lincoln, & Gallop, 2007), and
equivalent outcomes in a later Stage II trial (Greenfield et al.,
2014).
The Current Study
Although some elements of gender-neutral CBT approaches for
AUD include interventions that are relevant to female-specific
issues, such treatments do not explicitly focus on the application of
these interventions to female-specific issues or the overall context
of women’s lives, and are also not typically core, high priority
treatment targets. Given the dearth of evidence-based, female-
specific protocols to use in outpatient settings as either stand-alone
treatment or as part of a female-segregated treatment program, the
current study was designed to develop and test a female-specific
cognitive behavioral therapy (FS-CBT) treatment and compare it
to a standard, gender-neutral CBT (GN-CBT). If at least compa-
rable in efficacy, future research could then evaluate whether
FS-CBT enhances treatment access and utilization compared to a
mixed gender or gender-neutral option.
No study to date has used a “pure comparison design” (Sobell,
Sobell, & Agrawal, 2009) to compare GN-CBT for AUD with
FS-CBT that contains the same core CBT components plus addi-
tional female-centric content, themes, and presentation. For the
current study, we modified our evidence-based 12-session gender-
neutral CBT for AUD (Epstein & McCrady, 2009) to incorporate
content and themes uniquely relevant to women, based both on
theory and empirical findings. Core components of the original
GN-CBT manual that were retained for the FS-CBT adaptation
and were not expected to differ in outcome by condition in our
study, included motivational enhancement, coping skills to initiate
and maintain abstinence, general problem-solving, self-efficacy
for abstinence, and relapse prevention. New components were
added to FS-CBT to explicitly address issues for women with
AUD, including social support, self-confidence, interpersonal
functioning, and mood/emotion regulation problems. In FS-CBT,
these issues were addressed not only as triggers for drinking (as
they might be handled in GN-CBT) but also were presented as
separate, explicit, core, and prescribed manual-guided interven-
tions to directly treat deficits in these areas and to address these
skills in the context of women’s lives. In GN-CBT, the treatment
goal was to eliminate drinking; other clinical issues that arose were
addressed in a gender-neutral way and only as triggers for drink-
ing. In contrast, FS-CBT had dual goals—to eliminate drinking as
in GN-CBT, and to directly alleviate problems and address issues
in the female specific areas independent of the drinking-related
skills. FS-CBT fully integrated female specific programming
throughout every session in intervention content, themes, case
conceptualization, treatment goals, and structure (illustrations, vi-
gnettes).
For the current RCT. we assessed whether FS-CBT led to better
treatment engagement than GN-CBT and also tested three hypoth-
eses: Hypothesis 1. FS-CBT would be more efficacious than
GN-CBT in drinking outcomes during treatment and in the 12
months after; Hypothesis 2. outcomes targeted by both treatments
(coping skills to stop drinking and maintain abstinence, self-
efficacy and motivation for abstinence) would not differ between
conditions during and 12 months following treatment; and Hypoth-
esis 3. outcomes directly targeted in FS-CBT but not in GN-CBT
(self-confidence, interpersonal functioning, mood/emotion regula-
tion, and social support) would improve more in the FS-CBT
condition during treatment and follow-up.
Method
Trial Design and Setting
The trial was part of a larger two-arm choice study (McCrady,
Epstein, Cook, Jensen, & Ladd, 2011), in which women with AUD
chose either couple (McCrady, Epstein, Hallgren, Cook, & Jensen,
2016) or individual therapy, then randomized to one of two types
of couple or individual therapy within the chosen arm. Most
women chose the individual study arm during recruitment from
September 2003 to November 2005, after which the individual arm
was closed to allow sufficient time to complete recruitment for the
 FEMALE-SPECIFIC CBT FOR ALCOHOL 3

couple study arm (McCrady et al., 2011). The individual arm
compared standard GN-CBT with FS-CBT and is the focus of this
article. All study procedures were conducted at the Center of
Alcohol Studies at Rutgers University and approved by the Rut-
gers University Institutional Review Board.
Participants
Participants were recruited via media articles, advertisements,
and community outreach. Inclusion criteria were: (a) woman age
18 or older; (b) in a committed heterosexual relationship (married,
separated, cohabiting at least 6 months, or in a committed dating
relationship for at least 1 year) to accommodate the choice design;
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
meetings, role play, and supervised interviewing). Participants
were paid $50 for BL completion.
Randomization. After the BL, simple urn randomization
(Stout, Wirtz, Carbonari, & Del Boca, 1994) was used to counter-
balance conditions on depression (Beck Depression Inventory
scores ⬍14 vs. ⱖ14), personal drinking goal (abstinent or nonab-
stinent goal), and partner drinking status (recovering, abstainer, or
light drinker vs. moderate or heavy drinker).
Study treatments, therapists, and treatment condition ma-
nipulation effect. Both treatments were manual-guided, 12-
session outpatient, individual CBT with an explicit goal of absti-
nence from alcohol, provided over a maximum of 16 weeks.
Session 1 was 90 min; all others were 60 min. The GN-CBT

(c) consumed alcohol in the 30 days prior to telephone screen; (d)

manual was a modified version of a 20-session GN-CBT manual
This document is copyrighted by the American Psychological Association or one of its allied publishers.

met DSM–IV (American Psychiatric Association, 2000) criteria for

alcohol abuse or dependence. Exclusion criteria were: (a) physio-
logically dependent on drugs other than marijuana or nicotine, (b)
psychotic symptoms in past 6 months, and/or (c) gross cognitive
impairment.
Procedures
Telephone screen and in-person clinical screen. Callers
were screened for initial eligibility and study arm choice, and
scheduled for an individual or couple in-person clinical screen
interview. Four women attended a conjoint in-person screen but
then selected the individual arm. Participants completed self-report
questionnaires, and a research clinician administered a semistruc-
tured interview to assess eligibility, provide a study description,
and obtain informed consent. Intake clinicians had 22 hr of train-
ing.
Baseline research assessment (BL). A bachelors or masters
level research interviewer administered self-report questionnaires
and structured interviews to assess drinking and drug use, psycho-
social functioning, and psychopathology. BL interviewers had 42
hr of training (reading manuals, watching training videos, training
used in a prior RCT (Epstein & McCrady, 2009; McCrady, Ep-
stein, Cook, Jensen, & Hildebrandt, 2009), and included core CBT,
motivational enhancement therapy (MET) and relapse prevention
(RP) components delivered in a nonconfrontational, collaborative
therapist style.
In the FS-CBT manual (see Table 1), the core CBT, MET, and
RP elements were retained. Additionally, two female-specific
themes were highlighted throughout each session via discussion,
psychoeducation, and examples. The first theme emphasized self-
confidence of the woman as an active agent in her own life,
enhancing autonomy and empowerment, viewing herself as com-
petent and capable of managing her life, and being less “sociotro-
pic” (i.e., less emotionally and behaviorally reactive to others’
negative behavior and perceived expectations), as low autonomy
and high sociotropy may correlate with psychopathology (Bieling,
Beck, & Brown, 2000). The second theme emphasized self-care
beliefs and behaviors. In FS-CBT, all language, examples, vi-
gnettes, worksheets, and illustrations were female specific, includ-
ing those topics modified from the GN-CBT manual. The FS-CBT
manual also included several new modules linked to areas that
research suggests are particularly salient for women with AUD: (a)

Table 1
Female-Specific (FS-CBT) and Gender Neutral (GN-CBT) Content in the Manuals

GN-CBT FS-CBT
Treatment element Interventions session session

FS themes Self-care and self-confidence none all

Treatment goals Abstinence from alcohol all all
Increased self-confidence, interpersonal functioning, and social support none all
Format Gender neutral language all none
Gender neutral illustrations/clip art all none
Females in illustrations/clip art half all
Female-specific vignettes and completed worksheet examples none all
Core skills for both GN Motivational enhancement interventions 1, 4 1, 4
and FS-CBT Alcohol-specific coping skillsⴱ 1–12 1–12
General coping skills 9 9
Alcohol relapse prevention skills 9–12 9–12
Additional female-specific Psychoeducation, female alcoholism none 1
core interventions Heavy drinkers in social network (social support) none 3
Emotion regulation, coping with anxiety/depression none 5, 6
Connecting with others (social support, interpersonal functioning) none 7
Assertiveness training (interpersonal functioning) none 8
Anger management, emotion regulation none 9
ⴱ
To make room for FS interventions, some core alcohol related coping skills from the CBT protocol were removed or shortened in FS-CBT (see study
treatments in Method section). Core alcohol related coping skills were obtained from Epstein and McCrady (2009).
 4 EPSTEIN ET AL.
social support: coping with heavy drinkers in the social network,
and increasing social network support for abstinence; (b) interper-
sonal functioning and self-confidence: connecting with others,
increasing autonomy, decreasing sociotropy, and assertiveness
skills; (c) coping directly with mood problems and negative affect:
anxiety, depression, and emotion regulation; and (d) psychoedu-
cation about women and alcohol: discussion of unique risk factors,
consequences, and clinical correlates of excessive drinking for
women. Each of these elements of the FS-CBT manual was pre-
sented in a way that emphasized the female-specific nature of the
intervention. For instance, FS-CBT session three included exam-
ination of the structure of the woman’s social support network and
involvement of alcohol or drugs among its members, and skills
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
training to cope with heavy drinking social network members and
This document is copyrighted by the American Psychological Association or one of its allied publishers.
to strengthen connections with members supportive of abstinence.
In Session 7, interpersonal functioning was explicitly addressed by
evaluating and identifying characteristics of new or existing social
connections (such as people who treat the woman with respect, do
not abuse her, support her emotionally, nurture her, etc.) and
discussing strategies to connect with supportive people. Interper-
sonal functioning was enhanced in session eight via interventions
to learn about passive, aggressive, and assertive behaviors/expec-
tations and practicing cognitive and behavior changes for more
effective, empowering, and assertive interpersonal interactions.
Sessions 5 and 6 in FS-CBT included interventions to identify
specific symptoms of depression or anxiety and subtypes of anx-
iety the woman might be experiencing. Thought logs were used to
identify and track depressive and/or anxious thoughts, as was a
cognitive restructuring intervention to identify and replace think-
ing errors related to depression or anxiety. Emotion regulations
techniques in addition to cognitive restructuring were taught, in-
cluding relaxation training, paced breathing, time out, exercise,
and mindfulness. An anger management functional analysis inter-
vention covered additional strategies for regulating emotional re-
sponses. All interventions were linked to the female-specific
themes, for instance, anger management was linked to psychoe-
ducation on reduction of emotional reactivity to others (i.e., so-
ciotropy); assertiveness trained highlighted self-care beliefs and
autonomy in women’s interpersonal interactions. To keep session
length and number of sessions (dose) equivalent across treatment
conditions, three GN-CBT interventions were removed in the
FS-CBT protocol, including: developing a hierarchy of high risk
situations, reviews of skills and progress, and rearranging behav-
ioral consequences. Two treatment elements were shortened:
weekly check-in and seemingly irrelevant decisions.
Therapists (n ⫽ 14; 11 female) included seven doctoral- and
seven master’s-level psychologists, clinical social workers, or
counselors, all cross-trained to administer both therapy protocols.
Training took 39 hr and included reading the manuals and back-
ground material, attending workshops by the first two authors, and
reviewing and role-playing each session. Every session of each
therapist’s first two cases in each condition was audiotaped and
reviewed by one of the first two authors. A weekly team clinical
supervision was held; also, each case was assigned a clinical
supervisor who met with the therapist and listened to sessions as
needed. Clients were assigned to therapists based on mutual avail-
ability.
To evaluate therapy integrity, masters- or doctoral-level psy-
chologists (n ⫽ 16) rated full-session audiotapes of 42% of the
total number of sessions (about four sessions per participant). Two
independent raters double-coded 46 (15%) sessions. Therapy in-
tegrity was evaluated using (a) Therapist Checklist of each session,
on which therapists reported whether specific they delivered spe-
cific session elements; (b) Therapist Checklist, rater version, a
parallel form on which independent raters coded the degree to
which each item in the session was delivered; and (c) Treatment
Integrity Rating Scale (TIRS) developed by the first two authors.
Each TIRS item assessed quantity (adherence) and quality (com-
petence) of delivery of each treatment component, anchored from
1 ⫽ not at all to 5 ⫽ extensively for quantity, and from 1 ⫽ very
poor to 5 ⫽ excellent for quality. Thirty-eight items covered six
areas: GN-CBT interventions, FS-CBT interventions, female-
specific themes, common factors in psychotherapy, common fac-
tors in addiction-specific psychotherapy, and general adherence to
manual.
Follow-up. Participants were evaluated at the end of treatment
(3 months postbaseline; or at the end of treatment if a woman was
still in treatment up to 4 months postbaseline), and 9 and 15
months postbaseline. Women were paid $50 for the 3-month and
$75 for the 9- and 15-month interviews. Follow-up rates were
93%, 87%, and 81%, respectively. There were no differences in
follow-up rates between conditions, all ␹2(df ⫽ 1) ⱕ 0.13, p ⱖ .77.
Follow-up attrition was not associated with any baseline variables
except the SOCRATES Recognition scale, with higher scores
predicting a lower likelihood of completing a follow-up (Odds
Ratio [OR] ⫽ 0.72 to 0.77, p ⫽ .01).
Study Measures Administered at Baseline
Telephone screen. A brief structured telephone interview was
used to provide study information and to screen for initial eligi-
bility.
Clinical screen interview. This semistructured intake as-
sessed demographic characteristics, alcohol and other substance
use, AUD diagnosis, level of care determination, psychotic symp-
toms in the last 6 months (Psychotic Screen, SCID I; First, Spitzer,
Gibbon, & Williams, 2002), and gross cognitive deficits (Folstein,
Folstein, & McHugh, 1975). Women were asked to commit ver-
bally to an abstinence goal for the duration of the treatment.
Personality Diagnostic Questionnaire for DSM–IV (PDQ-
4ⴙ; Hyler, 1994). The PDQ-4⫹ is a 99-item self-report measure
for DSM–IV personality disorders to aid in description of the
clinical presentation of the sample. A total score above 25 was
used to indicate possible personality disorder of any type (see
McCrady et al., 2016). Cronbach’s alpha in the sample was .91.
Study Measures Administered at Baseline and
Follow-Up
Structured Clinical Interview for DSM–IV Disorders
(SCID-I; First et al., 2002). The SCID-I alcohol and drug use
module was administered at the clinical screen and each follow-up
to assess for alcohol dependence or abuse, as well as current/
lifetime mood and anxiety disorders (American Psychiatric Asso-
ciation, 2000). Interrater reliability for alcohol diagnoses is re-
ported at ␬ ⫽ .75; for other substance use disorders, ␬ ⫽ .84, and
for mood/anxiety disorders reported kappa’s are .84 –1.00 (Wil-
liams et al., 1992).
 FEMALE-SPECIFIC CBT FOR ALCOHOL
Timeline Followback (TLFB; Sobell & Sobell, 2003). The
TLFB captured daily drinking and drug use in the 90 days before the
last drink prior to baseline interview, and for all days since the last
interview at each follow-up. The primary drinking outcome vari-
ables were percent drinking days (PDD), percent days with heavy
drinking (PDH, defined as four or more standard drinks in a day,
Greenfield et al., 2010), and the percentage of the sample abstinent
from alcohol.
The Stages of Change and Treatment Eagerness Scale (SO-
CRATES; Miller & Tonigan, 1996). The SOCRATES is a
19-item self-report measure of readiness to change with three
subscales: recognition (range 7–35), ambivalence (range 4 –20),
and taking steps (range 8 – 40), and has adequate internal consis-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
tency, test–retest reliability, and predictive validity.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Beck Depression Inventory (BDI-II; Beck, Steer, & Brown,
1996). The BDI-II is a 21-item self-report measure of depression
symptoms over the last two weeks. Scores range from 0 – 63 and
are categorized as 0 –13 ⫽ minimal; 14 –19 ⫽ mild; 20 –28 ⫽
moderate; and 29 – 63 ⫽ severe depression. Reported Cronbach’s
alpha is .91.
Beck Anxiety Inventory (BAI; Beck, Epstein, Brown, &
Steer, 1988). The BAI is a 21-item measure of anxiety symp-
toms in the last week with scores ranging from 0 – 63 points, with
0 –7 ⫽ minimal anxiety, 8 –15 ⫽ mild anxiety, 16 –25 ⫽ moderate
anxiety, and 26 – 63 ⫽ severe anxiety. The BAI has high reliability
and internal consistency (Cronbach’s alpha ⫽ .94).
The Important People Interview (IPI; Longabaugh, Wirtz,
Zweben, & Stout, 1998). The IPI is a structured interview to
assess social network structure, drinking, and response to absti-
nence. Percent of network accepting/encouraging abstinence and
percent of network abstainers/in recovery were IPI variables de-
rived for the current study.
Situational Confidence Questionnaire-8 (SCQ-8; Breslin,
Sobell, Sobell, & Agrawal, 2000). The SCQ-8 is an eight-item
self-report measure of self-efficacy to abstain from alcohol in
high-risk situations, rated on a scale from 0% to 100% confi-
dence.
The Coping Behaviors Inventory (CBI; Litman, Stapleton,
Oppenheim, & Peleg, 1983). The CBI is a 36-item self-report
measure of strategies used to cope with drinking situations, with a
total score range of 0 to 108. Cronbach’s alpha coefficient for the
total scale is .93.
Sociotropy-Autonomy Scale (SAS; Bieling, Beck, & Brown,
2000). The SAS is a 27-item self-report measure (0- to 5-point
scale) with two subscales. Sociotropy reflects an individual’s con-
cern with others’ opinion of him/her and is described as “the
person’s investment in positive interchange with others” and “de-
pendence on social feedback for gratification and support” (Biel-
ing et al., 2000, p. 763). Autonomy reflects an individual’s self-
confidence and “the person’s investment in preserving and
increasing his [sic] independence, mobility, and personal rights”
(Bieling et al., 2000, p. 763).
Dyadic Adjustment Scale—short form (DAS-7; Hunsley,
Best, Lefebvre, & Vito, 2001; Spanier, 1976). The DAS-7 was
used to measure relationship satisfaction over time. Criterion,
convergent, and discriminative validity of the instrument are good
(Hunsley et al., 2001).
5
Study Measures Administered Within Treatment
Daily drinking logs (DDLs). Women kept daily records of
drinking and urges to drink during treatment. At each session, the
therapist reviewed the week’s DDLs to assess treatment progress
and guide skills training. Within-treatment drinking variables were
computed using DDL data supplemented with 3-month follow-up
TLFB data when DDL data were missing. Such DDL data are
highly correlated with retrospective TLFB data (McCrady, Ep-
stein, & Hirsch, 1999).
Homework record. Therapists recorded the completion of
assigned homework.
Study Measures Administered Only at 3-Month
Follow-Up
Working Alliance Inventory–Short Form–Client (WAI-S-C;
Busseri & Tyler, 2003). The WAI-S-C is a 12-item self-report
client’s perception of the therapeutic alliance with a possible range
of 1– 84. The WAI-S-C exhibits strong internal consistency, inter-
rater reliability, and predictive validity.
Client Satisfaction Questionnaire-8 (CSQ-8; Attkinsson &
Zwick, 1982). The CSQ-8 is a self-report measure of satisfaction
with services; scores range from 0 – 4.
Data Analysis Plan
Drinking outcomes were examined using continuous and
classification-based measures, and a modified intent-to-treat ap-
proach in which all participants completing at least one treatment
session (55 in GN-CBT; 44 in FS-CBT) were included (Witkie-
witz, Finney, Harris, Kivlahan, & Kranzler, 2015). For continuous
measures, growth curve models of PDD and PDH were generated
to test differences in change over time during and after treatment
with multiple trajectories (e.g., linear, quadratic, piecewise
growth). Based on raw descriptive data, model-fit indices, and
convergence patterns, a two-piece linear growth-curve model was
most suitable for modeling within-treatment drinking, with one
trajectory for Weeks 1– 8 and one for Weeks 9 –16, each with
random linear time effects and random subject-level intercepts.
Posttreatment drinking was modeled separately and included a
single linear growth term (i.e., intercept and time) with random
intercepts and slopes. Growth curve models assumed Gaussian
outcomes despite the non-normal drinking variable distributions.
Other approaches were considered that make fewer distributional
assumptions (e.g., generalized estimating equations, generalized
linear mixed models) but these have been shown to provide similar
effect estimates and significance tests for testing treatment out-
comes in growth curve models as multilevel models that assume
normal distributions (Hallgren, Atkins, & Witkiewitz, 2016).
Gaussian models were chosen instead of multilevel zero-inflated
models, to reduce complexity and enhance interpretability. We
also modeled categorical abstinence outcomes separately from
continuous drinking outcomes.
Time variables were scaled with 0 as the eighth week of treat-
ment (the transition point between the two linear pieces) for
within-treatment analyses and with 0 as the first week after the
treatment period for the posttreatment analyses. Treatment condi-
tion was coded as 0 for GN-CBT and 1 for FS-CBT and was then
 6 EPSTEIN ET AL.
mean-centered. Growth curve models were analyzed in R using
lmer (Bates, Maechler, Bolker, & Walker, 2014) with restricted
maximum likelihood, which handles missing data with less bias
than other approaches (Hallgren & Witkiewitz, 2013).
Classification-based measures labeled participants as continu-
ously abstinent (“abstainers,” PDD ⫽ 0), engaging in some non-
heavy drinking (“light-to-moderate drinkers,” PDD ⬎ 0, PDH ⫽
0), or engaging in any heavy drinking (“heavy drinkers,” PDH ⬎
0) throughout the treatment and follow-up periods examined. Dif-
ferences between conditions in rates of abstention (vs. any drink-
ing) and absence of heavy drinking days (vs. any heavy drinking
days) were tested using logistic regression during the full treatment
and follow-up periods, as well as during each week of treatment
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
and each month of follow-up.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Nondrinking outcomes (e.g., coping, depression, autonomy)
were compared with baseline at each of the three follow-up time
points (3, 9, and 15 months after baseline) to test changes within
and between conditions using mixed models with restricted max-
imum likelihood. Overall trends for change over time during the
posttreatment period also were tested using growth curve models
to examine possible gains attained after treatment or deterioration
of gains. Treatment condition differences were tested using linear
regression with multiple imputation, and mean-centered baseline
values of the outcome and propensity scores (to control for base-
line differences, see below) were entered as covariates to control
for baseline differences between conditions.
Distributions of variables were examined for anomalies and no
outliers were detected. Effect sizes were estimated using Cohen’s
d and 95% CIs. For hypotheses predicting no differences between
conditions, we used traditional null-hypothesis significance testing
as well as a stricter form of equivalence testing (Rogers, Howard,
& Vessey, 1993), where if the 95% CI of the differences between
the treatment conditions was contained entirely within a specified
interval (effect size of Cohen’s d ⫽ ⫾ 0.5), we concluded a
likelihood of equivalence between conditions. If any part of the
95% CI was outside the interval, we retained the null hypothesis of
nonequivalence.
A sample size of 118 in the individual arm was set as a goal for
the recruitment period providing power to detect significant dif-
ferences between conditions at a medium effect size (power ⫽
0.77 for Cohen’s d ⫽ 0.50). Based on our actual sample of N ⫽ 99,
we achieved an estimated power of 0.69 to detect an effect size of
d ⫽ 0.50.
Results
Participant Description
Ninety-nine women consented to study involvement, were ran-
domly assigned and attended at least one session of FS-CBT (n ⫽
44) or GN-CBT (n ⫽ 55). See Figure 1 for study flow. Sample
characteristics are presented in Table 2, with no significant base-
line differences between conditions on demographic characteris-
tics or drinking variables. Women were, on average, 48 years of
age and about half were employed full- or part-time. The sample
was primarily Caucasian. On average women drank on 71% of the
90 days prior to last drink before the baseline interview and drank
heavily on 53% of the days. All women met DSM–IV criteria for
current alcohol dependence; 94% met physiological dependence
criteria. Psychiatric comorbidity was substantial.
Descriptive Statistics and Randomization Equivalence
Unexpectedly, at baseline, women randomly assigned to the
FS-CBT condition were less depressed (t ⫽ 2.68, df ⫽ 96.57; p ⫽
.009), had lower self-efficacy (t ⫽ 2.18, df ⫽ 81.99, p ⫽ .033),
fewer coping behaviors (t ⫽ 2.42, df ⫽ 96.66, p ⫽ .02), and lower
SOCRATES taking steps (t ⫽ 1.98, df ⫽ 93.32, p ⫽ .05) scores
(see Table 2). Because these variables also were significantly
correlated with outcome variables (e.g., PDD, PHD), propensity-
score based covariate adjustment (Austin, 2011) was used to
control for baseline differences between conditions. Propensity
scores were computed for each participant to reflect her statistical
likelihood of being in GN-CBT versus FS-CBT, conditioned on
her baseline values of these covariates. Propensity scores were
then entered as covariates in all subsequent analyses to control for
these baseline differences in randomization (Austin, 2011). Similar
findings were obtained as those reported here when alternative
approaches were tested (e.g., adjusting for all confounding cova-
riates simultaneously, propensity-score based matching; see online
supplemental Table 1 for a detailed description of propensity score
estimation and illustration of its effect on controlling for baseline
differences).
Pretreatment reduction in drinking during the assessment period
was found across the whole sample (t ⫽ ⫺6.28, df ⫽ 86; p ⬍ .001)
with no treatment condition effects. We tested effects of pretreat-
ment drinking reduction on study outcomes; greater pretreatment
reduction in drinking was associated with lower within and 12-
month posttreatment PDD at trend levels (.08), with no treatment
condition interaction effect. Because pretreatment drinking change
did not significantly predict treatment outcome and did not differ
by treatment condition at prebaseline or in regard to effect on
outcome, we did not control for pretreatment change in drinking.
Therapy Integrity Results
Objective ratings indicated that therapists in the GN-CBT con-
dition completed a significantly lower percentage of session ma-
terials (86.9%) than in the FS-CBT condition, 91.8%, t(79.9) ⫽
⫺2.48, p ⫽ .02. A manipulation effect of treatment condition on
therapist behaviors was supported. Independent raters accurately
classified treatment condition for 94% of GN-CBT and 88% of
FS-CBT sessions. For the TIRS items, t tests (all p ⬍ .001)
indicated that FS-CBT was rated higher in quantity on FS themes
(M ⫽ 2.70, SD ⫽ 0.65) than GN-CBT (M ⫽ 1.26, SD ⫽ 0.31), and
higher quantity ratings on FS interventions (M ⫽ 1.89, SD ⫽ 0.45)
than GN-CBT (M ⫽ 1.41, SD ⫽ 0.37). Note that quantity ratings
on FS core interventions are averaged across all rated sessions,
including those in which FS interventions are not prescribed, that
is, not just for the seven sessions in which core FS interventions
are delivered (see Table 1), thus lowering the overall expected
average quantity rating for the FS intervention variable. Quality
ratings for female-specific themes in FS-CBT were high (M ⫽
3.69, SD ⫽ 0.39). There were no differences between treatment
conditions in quality or quantity of common factors, general ad-
herence to manual, or gender-neutral CBT interventions.
 FEMALE-SPECIFIC CBT FOR ALCOHOL 7

Telephone Screened Unable to Determine Eligibility (N=12)

Ineligible (N=38):
(N=286) Same-sex partner (N=5), no partner (N=15),
no alcohol use in past 30 days (N=13);
woman calling for her partner (N=2);
Daily cocaine or heroin use (N=1);
Other (N=2)
Eligible
(N=236) Did Not Schedule Intake (N=54)
Partner refused to participate (N=1)
ENROLLMENT

Practical barriers (N=13)

Scheduled Clinical
Interview (N=182)
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Said would call back and didn’t (N=8)
Said would check with partner (N=11)
Did not call back (N=6)
Not interested (N=14)
Other (N=1)

Cancelled/No-show (N=48)
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Entered higher level of care (N=1)

Completed individual
interview (N=109)
Completed couple interview
then switched to individual
condition (N=3)
Completed Baseline (BL)
(N=102)
Randomization
ALLOCATION
Not interested in our treatment (N=1)
Did Couple Clinical Interview (N=23)
Ineligible at Clinical Interview (n = 4)
No alcohol abuse/dependence (N=1)
No alcohol in past 30 days (N=2)
Didn’t meet partner criterion (N=1)
Eligible, dropped out before BL (N=5)
Not interested (N=3); Practical barriers
(n=1); No show to BL (N=1)
Ineligible at Baseline (N=1)

Allocated to CBT (n=57) Allocated to Female Specific CBT (n=45)

• Received CBT (n=55) • Received FS-CBT (n=44)
• Did not receive CBT (n=2) • Did not receive FS-CBT (n=1)
Reason: Issues unrelated to study Reason: No Show/lost contact

Lost to follow up Lost to follow up

FOLLOW-UP

• 3-month (n=3, refused; n=1 lost contact) • 3-month (n=1, refused; n=1, lost contact)
• 9-month (n=5, refused; n=2, lost contact) • 9-month (n=3, refused; n=2, lost contact)
• 15-month (n=5, refused; n=2, lost • 15-month (n=6, refused; n=2, lost contact)
contact)

Analyzed Analyzed
ANALYSIS

(N=53 [within-tx], N=49 [post-tx]) (N=43 [within-tx], N=41 [post-tx])

Excluded (N=2, 6) Excluded (N=1, 3)
Reason: not followed Reason: not followed

Figure 1. CONSORT figure of study participant flow.

Treatment Engagement with an average of 3.68 on a scale of 1– 4 and no differences

There were no differences between treatment conditions on
treatment engagement indices, including the number of sessions
completed, therapeutic alliance, client satisfaction, and percentage
of assigned homework completed. Overall, participants were
highly engaged and satisfied with both treatments. Women at-
tended an average of 8.87 of the 12 sessions, with 58% of women
attending all 12 sessions and no significant differences between
conditions (GN-CBT, M ⫽ 8.85, FS-CBT, M ⫽ 8.91), t(93.11) ⫽
⫺0.06, p ⫽ .95. Participants completed 67.6% of assigned home-
work with no differences between conditions (GN-CBT, M ⫽
66.36%, FS-CBT, M ⫽ 69.06%), t(96.02) ⫽ ⫺0.47, p ⫽ .64.
Women reported high therapeutic alliance with no differences
between conditions (GN-CBT, M ⫽ 77.00, FS-CBT, M ⫽ 75.26)
t(56.11) ⫽ 0.74, p ⫽ .46, and high satisfaction with the therapy
between conditions (GN-CBT, M ⫽ 3.64, FS-CBT, M ⫽ 3.73),
t(83.67) ⫽ ⫺0.99, p ⫽ .32.
Hypothesis 1: Drinking Outcomes
Growth curve model results of changes in drinking are pre-
sented in Table 3. PDD and PDH decreased significantly for
both conditions over the first eight weeks of treatment at an
average of 2.50 percentage points of PDD each week and 1.38
percentage points of PDH per week (i.e., 20.02 and 11.06 total
percentage points reduction in PDD and PDH, respectively,
over the first 8-week period). PDD and PDH did not change
significantly during treatment Weeks 9 –16 or in the posttreat-
ment period. There were no main effects of treatment condition
or Treatment Condition ⫻ Time Interactions, indicating that the
 8 EPSTEIN ET AL.

Table 2
Participant Baseline Demographic, Drinking, and Other Psychopathology

Full Sample GN-CBT FS-CBT

(N ⫽ 99) (N ⫽ 55) (N ⫽ 44)
Variables M or N (SD) or % M or N (SD) or % M or N (SD) or %

Age 47.88 (8.85) 48.24 (8.45) 47.43 (9.41)

Education 15.09 (2.50) 15.22 (2.57) 14.93 (2.42)
Household income (thousands) 96.62 (60.44) 97.97 (46.16) 94.91 (75.20)
Employed (N) 53 53.54% 28 50.91% 25 56.82%
White non-Hispanic (N) 91 91.92% 52 94.55% 39 88.64%
Mean N of alcohol abuse and dependence symptoms (of 11) 8.48 (1.82) 8.60 (1.78) 8.34 (1.88)
Percent drinking days 70.99 (28.77) 69.51 (29.61) 72.83 (27.91)
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Percent heavy drinking daysa 53.18 (34.14) 55.07 (33.47) 50.81 (35.20)
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Drinks per drinking day 6.83 (3.94) 7.07 (4.28) 6.52 (3.49)
Beck depression inventoryⴱ 20.67 (10.44) 23.04 (11.32) 17.70 (8.44)
Coping behaviors inventoryⴱⴱ 37.61 (17.55) 41.27 (18.33) 33.04 (15.53)
Situational confidenceⴱⴱ 48.85 (23.95) 53.51 (21.63) 42.89 (25.65)
SOCRATES taking stepsⴱⴱ 28.83 (7.73) 30.18 (7.71) 27.14 (7.50)
⬎1 Current Axis-I disorder (N) 45 45.45% 29 52.73% 16 36.36%
Current MDDb or dysthymia (N) 34 34.34% 24 43.64% 10 22.73%
ⱖ1 current anxiety disorder (N) 33 33.33% 19 34.54% 14 31.82%
⬎1 Lifetime Axis-I disorder (N) 76 76.77% 42 76.36% 34 77.27%
PDQc 28.11 (14.43) 29.95 (14.16) 25.82 (14.60)
Note. GN-CBT ⫽ Gender-neutral cognitive-behavioral therapy; FS-CBT ⫽ Female-specific cognitive-behavioral therapy. No significant differences were
noted between FS-CBT and CBT on independent samples t test or Fisher’s exact test.
a
Defined for women as ⱖ4 standard drinks within a day. b MDD ⫽ Major Depressive Disorder. c Personality Disorder Questionnaire ⬎25 indicates
likely personality disorder.
ⴱ
p ⬍ .01. ⴱⴱ p ⬍ .05.

trajectories of PDD and PDH during and after treatment did not
differ significantly by treatment condition. Raw values of PDD
and PDH are presented for each treatment condition in Figure 2.
Effect size estimates of drinking differences between conditions
aggregated across the early-, late-, and posttreatment periods
were all nonsignificant and small for PDD (d ⫽ 0.08, 0.13, and
0.09, respectively) and nonsignificant and small-to-medium for
PDH (d ⫽ 0.33, 0.28, and 0.19, respectively). Additional con-
fidence interval statistics for each drinking outcome measure
during each period are provided in online supplementary ma-
terials. Treatment conditions also were compared using chi-
square tests (df ⫽ 1) on the percentage of the sample abstinent
and the percentage of the sample with no heavy drinking days
aggregated within each week of treatment and month of follow-

Table 3
Drinking Outcome Growth Curve Models

Percent drinking days (PDD) Percent days heavy drinking (PDH)

Within-treatment Estimate (SE) p Estimate (SE) p

(Intercept) 27.10 (3.39) ⬍.001 13.54 (2.75) ⬍.001

Treatment condition 1.02 (7.29) .89 11.06 (6.00) .07
Linear time (Weeks 1–8) ⫺2.51 (.48) ⬍.001 ⫺1.38 (.37) ⬍.001
Linear time (Weeks 9–16) .22 (.42) .60 .57 (.37) .12
Propensity score 11.72 (13.97) .40 ⫺.81 (12.14) .95
Baseline PDD or PDH .38 (.09) ⬍.001 .26 (.07) ⬍.001
Treatment Condition ⫻ Time (Weeks 1–8) ⫺.87 (.96) .37 .20 (.75) .79
Treatment Condition ⫻ Time (Weeks 9–16) .88 (.85) .31 ⫺.31 (.75) .68
Posttreatment (Weeks 17–63)
(Intercept) 31.62 (3.48) ⬍.001 15.74 (3.18) ⬍.001
Treatment condition 3.23 (7.58) .68 2.14 (6.77) .75
Linear time .15 (.09) .09 .04 (.09) .69
Propensity score 8.38 (16.53) .62 ⫺4.04 (12.27) .74
Baseline PDD or PDH .36 (.11) .002 .26 (.07) ⬍.001
Treatment Condition ⫻ Time ⫺.03 (.17) .85 .16 (.18) .36
Note. GN-CBT ⫽ Gender-neutral cognitive-behavioral therapy; FS ⫽ Female-specific. PDD/PDH values ranged from 0 (no drinking/heavy-drinking
days) to 100 (drinking/heavy-drinking every day). Within-treatment, N ⫽ 96 (53 and 43 for GN-CBT and FS); Posttreatment, N ⫽ 85 (47 and 38 for
GN-CBT and FS).
 FEMALE-SPECIFIC CBT FOR ALCOHOL 9
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 2. Percent drinking days, heavy drinking days, and of sample with no drinking days per week during
treatment and per month during follow-up. No treatment condition differences were found.

up, as well as across the full treatment and follow-up periods.
There were no significant differences between conditions in
categorical drinking outcomes (see Figure 2). In summary,
drinking outcomes improved equally across both FS-CBT and
GN-CBT, with no differences between conditions.
Hypothesis 2: Coping, Self-Efficacy, and Motivation
for Abstinence
Means and standard deviations of nondrinking baseline and
follow-up variables are presented in supplemental materials. All
within-subject comparisons of change at the end of treatment
and during the follow-up period, relative to baseline and con-
trolling for treatment condition, were significant and indicated
small to medium increases in coping, medium to large increases
in self-efficacy, and medium to large changes in motivation that
indicated reduced ambivalence and problem recognition and
increased taking steps. All effect size estimates of differences
between treatment conditions were small and nonsignificant (all
Cohen’s | d | ⬍ 0.20, all p ⬎ .60). However, none of the 95% CI
intervals met the strict criterion of being contained entirely
within the range of d ⫽ ⫾0.5, preventing us from making
stronger conclusions about the equivalence of the two treat-
ments (Rogers et al., 1993).
Growth-curve models (see supplementary materials) tested pat-
terns of change in these outcomes over the 12-month posttreatment
period, anchored to the scores at the start of follow-up. Use of
coping behaviors decreased over the follow-up period (p ⫽ .02)
 10 EPSTEIN ET AL.

with no differences between conditions. Situational confidence and
SOCRATES recognition scores did not change significantly over
follow-up in either condition. SOCRATES ambivalence and taking
steps scores decreased significantly over follow-up for the full
sample (p ⫽ .004 and p ⬍ .001, respectively) with no difference
between conditions (see Figure 3).
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Hypothesis 3: Inter- and Intrapersonal Functioning
and Social Support
Within-subject comparisons in both conditions showed that de-
pression and anxiety were significantly lower at each follow-up
assessment relative to baseline, with large effect sizes. Sociotropy

Figure 3. Change over time, within treatment and posttreatment, for outcome variables targeted in both
Gender-neutral cognitive-behavioral therapy (GN-CBT) and Female-specific cognitive-behavioral therapy (FS-
CBT) and hypothesized to not differ by treatment condition.
 FEMALE-SPECIFIC CBT FOR ALCOHOL
was significantly lower in all follow-up periods relative to baseline
with small effect sizes, and autonomy was significantly higher at
Month 3 with a small effect size but was not different from
baseline at Month 9 or 15. Network encouragement of abstinence
and network abstaining/recovery status were not significantly
higher at posttreatment relative to baseline.
Between-subjects comparisons showed that none of these vari-
ables differed significantly by condition during treatment or in the
follow-up period. Depression and anxiety scores were nonsignifi-
cantly lower in FS-CBT compared with GN-CBT, with between-
groups effect sizes ranging from d ⫽ ⫺0.09 to ⫺0.34 (p ⱖ .22),
and relationship functioning was nonsignificantly higher in FS-
CBT with effect sizes ranging from d ⫽ .30 to .37 (p ⱖ .21).
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Between-groups effect size magnitudes for sociotropy, autonomy,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
social network support for abstinence, and social network absti-
nence/recovery status were always smaller in absolute value than
d ⫽ 0.31.
Growth curve models of changes in these nondrinking variables
during the posttreatment period are shown in detail in online
supplemental materials. In summary, analyses indicated that so-
ciotropy and autonomy declined significantly over the posttreat-
ment period with no difference in rates of change between condi-
tions. There was a significant difference in rate of change of social
network abstainer/recovery status over time during follow-up by
treatment condition (p ⫽ .01). Subgroup analyses indicated there
was no change during the follow-up period in network abstainer/
recovery status in the GN-CBT condition (rate ⫽ ⫺0.03 per
month, SE ⫽ 0.18, p ⫽ .85), but there was an increase in the
percentage of abstaining network members in the FS-CBT condi-
tion (rate ⫽ 0.69 per month, SE ⫽ 0.21, p ⫽ .002); however, there
were no significant differences between treatment conditions at
any time point (all p ⬎ .21; see Figure 4).
Discussion
The purpose of the current study was to develop and test a
motivational and coping skills-based outpatient FS-CBT for
women with AUD, hypothesized to be superior to gender-neutral
CBT in drinking and FS-CBT targeted outcomes. We modified an
efficacious, individual, gender-neutral treatment manual for AUD
(Epstein & McCrady, 2009) based on the extant literature on the
unique clinical presentation and treatment needs of women with
AUD. The new FS-CBT protocol was compared to GN-CBT in an
RCT, using a rigorous “pure comparison” design (Sobell et al.,
2009) to isolate hypothesized female specific treatment compo-
nents in CBT for AUD from a control condition of equivalent core
GN-CBT interventions.
Hypothesis 1, that FS-CBT would yield superior drinking out-
comes compared with GN-CBT, was not supported. With no
differences by treatment condition, women in both FS-CBT and
GN-CBT were highly engaged, satisfied, and compliant with treat-
ment, reported high therapeutic alliance, and significantly reduced
their drinking frequency and intensity, usually in the first 8 weeks
of treatment, and then maintained positive drinking outcomes
throughout the rest of treatment and the following year. Hypothesis
2 was supported; as predicted, no treatment condition differences
were noted in outcomes of secondary variables targeted in both
GN-CBT and FS-CBT—women during both treatments reported
better coping strategies, increased abstinence self-efficacy, more
11
steps taken to stop drinking, and reduced ambivalence about
change. Across both conditions, significant reductions were noted
in the use of alcohol-specific coping behaviors, ambivalence about
change, and taking steps to change over 12 months following
treatment; however, we do not know if these reductions after
treatment are secondary to increases in abstinence thus requiring
less coping with drinking-related situations, reduced need for steps
to change drinking, and less ambivalence about whether they had
an alcohol problem.
Hypothesis 3, that women in the FS-CBT condition would
report better intra- and interpersonal functioning and better social
support, was only partially supported. Women in both treatments
reported significant reductions in overconcern with evaluation by
others and dysregulated emotional reactivity to others, and im-
provements in autonomy (i.e., self-confidence, self-esteem, and
independence), depression, and anxiety from baseline to posttreat-
ment, with no differential improvement in the FS-CBT condition.
Relationship satisfaction did not increase across the treatment
period for either condition, but did improve significantly in both
conditions by the 9 months postbaseline assessment. Social sup-
port for abstinence also did not increase during treatment in either
condition, however, over the following year women in the FS-CBT
but not the GN-CBT condition reported a significant increase in
the number of abstainers in their social network.
Thus, women had positive outcomes with either FS-CBT or GN-
CBT; the lack of added value of the FS material in all areas except
social support for abstinence after treatment was unexpected. The
substantial body of research supporting gender differences in most
aspects of AUD and the efficacy of female-specific content in female-
segregated AUD treatment suggested that comparing the relative
efficacy of FS-CBT with GN-CBT was warranted. The wider litera-
ture has suggested a likely additive benefit from female-centric AUD
treatment compared to a gender-neutral AUD treatment (see Epstein
& Menges, 2013); however, no study to date had evaluated this using
a “pure comparison” (Sobell, Sobell, & Agrawal, 2009) design that
isolated female-specific compared with gender-neutral programming,
as in the current study.
We considered several possible reasons for the specific findings.
First, lack of a manipulation effect (i.e., no actual difference in content
of FS-CBT and GN-CBT delivered sessions despite different manu-
als) was considered as a possible reason for lack of differences in
outcome, but deemed unlikely. Given that both treatments were
individual modality and thus tailored to each client’s clinical presen-
tation, it might have been possible that therapists delivered equivalent
FS content in GN-CBT if relevant FS clinical material arose in
session. Relatedly, it is possible that the current design comparing two
individual therapies did not optimally isolate female specific adapta-
tions; advantages of FS-CBT compared to GN-CBT may be more
apparent in a comparison of group treatment modalities for women, in
which the GN-CBT group would less likely be tailored to FS aspects
of individual clinical presentations of each participant. However,
detailed and thorough treatment integrity ratings showed that the
FS-CBT protocol was distinctly identifiable from the gender-neutral
CBT protocol with significant differences in quantity of FS-CBT
interventions and female-specific themes compared with the GN-CBT
condition.
Second, the use of a “pure comparison” study design may have
attenuated treatment condition differences. The control condition was
an efficacious treatment that may have produced ceiling effects that
 12 EPSTEIN ET AL.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 4. Change over time, during and post treatment, in nondrinking variables explicitly targeted in
Female-specific cognitive-behavioral therapy (FS-CBT) and hypothesized to differ in outcome compared to
Gender-neutral cognitive-behavioral therapy (GN-CBT). The only case where this was true was for percentage
of network abstainers/recovering.

precluded incrementally better outcomes for a comparison treatment,
particularly one including many of the same efficacious treatment
components. It is possible that GN-CBT is as potent as any outpatient
treatment for AUD can be, so that it would not be fruitful to adapt it
for special populations, at least not women. However, there is docu-
mented appeal of female-segregated treatments that confer advantage
only when female-specific programming is provided. Thus, one in-
terpretation of the current results is that FS-CBT may be a viable
alternative to GN-CBT programming in outpatient individual therapy
for AUD, or to integrate into existing female-segregated AUD treat-
ments that currently use gender-neutral programming. It is possible
that some women may be more likely to enter female-specific gender-
neutral treatment, and this potential can be tested in subsequent
research testing moderators of treatment access and outcomes.
Third, treatment differences may have been further diminished
because clinicians delivering both treatments in the context of a
rigorously executed RCT were highly educated, extensively trained,
and closely supervised, and delivery of both treatments was excel-
lent. In a community AUD setting where CBT might not be
applied as rigorously, the female-specific components of FS-CBT
 FEMALE-SPECIFIC CBT FOR ALCOHOL
may have had more impact against a backdrop of overall less
positive treatment outcomes with gender-neutral CBT-style treat-
ment as usual.
Fourth, it is possible that the female specific interventions
chosen for this FS-CBT protocol were not sufficiently relevant or
potent to provide added value above that of the GN-CBT. How-
ever, high client satisfaction scores belie this interpretation, as
does our anecdotal experience providing the therapy and listening
to session tapes. Women in the FS-CBT condition often sponta-
neously reported that the vignettes, handout information, and FS
topics were highly relevant, helpful, and resonated with the wom-
en’s personal experiences in their struggles with AUD.
There are limitations of the current study. First, the present
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
study was potentially underpowered to detect differences between
This document is copyrighted by the American Psychological Association or one of its allied publishers.
the treatment conditions. Although the sample provided adequate
power to detect group differences with medium effect sizes,
achieving those effects was likely limited by using an active
comparison individual treatment (GN-CBT). Second, to maintain
consistency in inclusion criteria across two arms of the larger
study, the women in the individual arm of the study all were
married or in a committed relationship, so we must use caution in
generalizing findings to the population of all women with AUD.
Third, the sample was primarily Caucasian, despite efforts to
recruit a more diverse sample; however, there was heterogeneity in
level of psychosocial functioning, and in prevalence of Axis I and
II pathology. Fourth, as mentioned, comparison of two individual
modalities in which the control was a sophisticated, efficacious
gender-neutral CBT approach tailored to specific cases and may
have mitigated treatment condition effects.
Strengths of the study include a rigorous design that isolated the
female-specific components of a CBT manual for AUD; compared
two manual-guided, well-specified and empirically informed treat-
ments each delivered with high integrity; used random assignment
to treatment conditions; used validated measures; had good
follow-up rates; and used a sophisticated data analytic approach.
Female-specific adaptions to the gender-neutral manual were cho-
sen carefully based on extant literature and measurement of con-
structs was linked to a conceptual framework of the unique treat-
ment needs of women with AUD. Therapists were cross-trained in
both treatment conditions to avoid therapist bias, well-trained,
carefully supervised to adhere to the manuals and to administer
them skillfully, and there was a distinct manipulation effect be-
tween treatment conditions.
In summary, results from the current study demonstrate that
both female-specific CBT and gender-neutral CBT were highly
palatable and associated with improvements in alcohol use and
other outcomes (depression, anxiety, coping, motivation, self-
efficacy, empowerment) for a sample of women who reported
heavy and regular use of alcohol and severe alcohol-related prob-
lems at baseline. Improvements in drinking and most of the other
variables were sustained over a 12-month follow-up period.
Women in FS-CBT may garner additional social support for ab-
stinence in the 12 months after treatment. Thus, though The
FS-CBT protocol did not yield superior outcomes compared with
GN-CBT, FS-CBT appears to be as efficacious a treatment option
for women with AUD and related problems as GN-CBT, and
might be a viable evidence-based, stand-alone outpatient treatment
or might complement existing gender-neutral programming in
AUD treatment facilities. Subsequent research is warranted to test
13
if there is added value of the FS-CBT protocol (a) in superior rates
of treatment utilization compared to GN-CBT or GN treatment-
as-usual in community settings; (b) in a group FS-CBT compared
with group GN-CBT design; (c) in single versus mixed gender
groups; and (c) in conjunction with specific moderators of change
to examine whether FS-CBT and GN-CBT is differentially effec-
tive for certain patients.
References
Abulseoud, O. A., Karpyak, V. M., Schneekloth, T., Hall-Flavin, D. K.,
Loukianova, L. L., Geske, J. R., . . . Frye, M. A. (2013). A retrospective
study of gender differences in depressive symptoms and risk of relapse
in patients with alcohol dependence. The American Journal on Addic-
tions, 22, 437– 442. http://dx.doi.org/10.1111/j.1521-0391.2013.12021.x
American Psychiatric Association. (2000). Diagnostic and statistical man-
ual of mental disorders (4th ed., text revision). Washington, DC: Author.
Ashley, O. S., Marsden, M. E., & Brady, T. M. (2003). Effectiveness of
substance abuse treatment programming for women: A review. The
American Journal of Drug and Alcohol Abuse, 29, 19 –53. http://dx.doi
.org/10.1081/ADA-120018838
Attkisson, C. C., & Zwick, R. (1982). The client satisfaction questionnaire.
Psychometric properties and correlations with service utilization and
psychotherapy outcome. Evaluation and Program Planning, 5, 233–237.
http://dx.doi.org/10.1016/0149-7189(82)90074-X
Austin, P. C. (2011). An introduction to propensity score methods for
reducing the effects of confounding in observational studies. Multivar-
iate Behavioral Research, 46, 399 – 424. http://dx.doi.org/10.1080/
00273171.2011.568786
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2014). lme4: Linear
mixed-effects models using Eigen and S4 (R package version 1.1– 6).
Retrieved from http://CRAN.R-project.org/package⫽lme4
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory
for measuring clinical anxiety: Psychometric properties. Journal of
Consulting and Clinical Psychology, 56, 893– 897. http://dx.doi.org/10
.1037/0022-006X.56.6.893
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck Depression
Inventory-II (BDI-II) manual. San Antonio, TX: Pearson.
Bieling, P. J., Beck, A. T., & Brown, G. K. (2000). The Sociotropy-
Autonomy Scale: Structure and implications. Cognitive Therapy and
Research, 24, 763–780. http://dx.doi.org/10.1023/A:1005599714224
Breslin, F. C., Sobell, L. C., Sobell, M. B., & Agrawal, S. (2000). A
comparison of a brief and long version of the Situational Confidence
Questionnaire. Behaviour Research and Therapy, 38, 1211–1220. http://
dx.doi.org/10.1016/S0005-7967(99)00152-7
Breslow, R. A., Castle, I. P., Chen, C. M., & Graubard, B. I. (2017). Trends
in alcohol consumption among older Americans: National health inter-
view surveys, 1997 to 2014. Alcoholism: Clinical and Experimental
Research, 41, 976 –986.
Busseri, M. A., & Tyler, J. D. (2003). Interchangeability of the working
alliance inventory and working alliance inventory, short form. Psycho-
logical Assessment, 15, 193–197. http://dx.doi.org/10.1037/1040-3590
.15.2.193
Connors, G. J., & Walitzer, K. S. (2001). Reducing alcohol consumption
among heavily drinking women: Evaluating the contributions of life-
skills training and booster sessions. Journal of Consulting and Clinical
Psychology, 69, 447– 456. http://dx.doi.org/10.1037/0022-006X.69.3
.447
Covington, S. S. (2002). Helping women recover: Creating gender-
responsive treatment. In S. L. A. Straussner & S. Brown (Eds.), The
handbook of addiction treatment for women: Theory and practice (pp.
52–72). San Francisco, CA: Jossey-Bass.
Cucciare, M. A., Simpson, T., Hoggatt, K. J., Gifford, E., & Timko, C.
(2013). Substance use among women veterans: Epidemiology to
 14 EPSTEIN ET AL.
evidence-based treatment. Journal of Addictive Diseases, 32, 119 –139.
http://dx.doi.org/10.1080/10550887.2013.795465
Dahlgren, L., & Willander, A. (1989). Are special treatment facilities for
female alcoholics needed? A controlled 2-year follow-up study from a
specialized female unit (EWA) versus a mixed male/female treatment
facility. Alcoholism, Clinical and Experimental Research, 13, 499 –504.
http://dx.doi.org/10.1111/j.1530-0277.1989.tb00366.x
Dawson, D. A. (1996). Gender differences in the probability of alcohol
treatment. Journal of Substance Abuse, 8, 211–225. http://dx.doi.org/10
.1016/S0899-3289(96)90260-6
Epstein, E. E., & McCrady, B. S. (2009). Overcoming alcohol use prob-
lems: A cognitive-behavioral treatment program therapist guide. New
York, NY: Oxford University Press.
Epstein, E. E., & Menges, D. (2013). Women and addiction. In B. S.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
McCrady & E. E. Epstein (Eds.), Addictions: A comprehensive guide-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
book (2nd ed., pp. 788 – 818). New York, NY: Oxford University Press.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. (2002).
Structured clinical interview for DSM–IV–TR axis I disorders. New
York, NY: Biometrics Research, NY State Psychiatric Institute.
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental
state.” A practical method for grading the cognitive state of patients for
the clinician. Journal of Psychiatric Research, 12, 189 –198. http://dx
.doi.org/10.1016/0022-3956(75)90026-6
Gamble, S. A., Talbot, N. L., Cashman-Brown, S. M., He, H., Poleshuck,
E. L., Connors, G. J., & Conner, K. R. (2013). A pilot study of
interpersonal psychotherapy for alcohol-dependent women with co-
occurring major depression. Substance Abuse, 34, 233–241. http://dx.doi
.org/10.1080/08897077.2012.746950
Gjestad, R., Franck, J., Lindberg, S., & Haver, B. (2011). Early treatment
for women with alcohol addiction (EWA) reduces mortality: A random-
ized controlled trial with long-term register follow-up. Alcohol and
Alcoholism, 46, 170 –176. http://dx.doi.org/10.1093/alcalc/agq097
Greenfield, S. F., Brooks, A. J., Gordon, S. M., Green, C. A., Kropp, F.,
McHugh, R. K., . . . Miele, G. M. (2007). Substance abuse treatment
entry, retention, and outcome in women: A review of the literature. Drug
and Alcohol Dependence, 86, 1–21. http://dx.doi.org/10.1016/j
.drugalcdep.2006.05.012
Greenfield, S. F., & Grella, C. E. (2009). What is “women-focused”
treatment for substance use disorders? Psychiatric Services, 60, 880 –
882. http://dx.doi.org/10.1176/ps.2009.60.7.880
Greenfield, S. F., Pettinati, H. M., O’Malley, S., Randall, P. K., & Randall,
C. L. (2010). Gender differences in alcohol treatment: An analysis of
outcome from the COMBINE study. Alcoholism, Clinical and Experi-
mental Research, 34, 1803–1812. http://dx.doi.org/10.1111/j.1530-0277
.2010.01267.x
Greenfield, S. F., Sugarman, D. E., Freid, C. M., Bailey, G. L., Crisafulli,
M. A., Kaufman, J. S., . . . Fitzmaurice, G. M. (2014). Group therapy for
women with substance use disorders: Results from the Women’s Re-
covery Group Study. Drug and Alcohol Dependence, 142, 245–253.
http://dx.doi.org/10.1016/j.drugalcdep.2014.06.035
Greenfield, S. F., Trucco, E. M., McHugh, R. K., Lincoln, M., & Gallop,
R. J. (2007). The Women’s Recovery Group Study: A Stage I trial of
women-focused group therapy for substance use disorders versus mixed-
gender group drug counseling. Drug and Alcohol Dependence, 90,
39 – 47. http://dx.doi.org/10.1016/j.drugalcdep.2007.02.009
Hallgren, K. A., Atkins, D. C., & Witkiewitz, K. (2016). Aggregating and
analyzing daily drinking data in clinical trials: A comparison of type-I
errors, power, and bias. Journal of Studies on Alcohol and Drugs, 77,
986 –991. http://dx.doi.org/10.15288/jsad.2016.77.986
Hallgren, K. A., & Witkiewitz, K. (2013). Missing data in alcohol clinical
trials: A comparison of methods. Alcoholism, Clinical and Experimental
Research, 37, 2152–2160. http://dx.doi.org/10.1111/acer.12205
Haver, B., & Gjestad, R. (2005). Phobic anxiety and depression as predic-
tor variables for treatment outcome. A LISREL analysis on treated
female alcoholics. Nordic Journal of Psychiatry, 59, 25–30. http://dx
.doi.org/10.1080/08039480510018797
Heslin, K. C., Gable, A., & Dobalian, A. (2015). Special services for
women in substance use disorders treatment: How does the Department
of Veterans Affairs compare with other providers? Women’s Health
Issues, 25, 666 – 672. http://dx.doi.org/10.1016/j.whi.2015.07.005
Hien, D. A., Litt, L., Cohen, L. C., Miele, G. M., & Campbell, A. N.
(2009). Integrating trauma services for women in addictions treatment.
New York, NY: APA.
Hunsley, J., Best, M., Lefebvre, M., & Vito, D. (2001). The seven-item
short form of the Dyadic Adjustment Scale: Further evidence for con-
struct validity. The American Journal of Family Therapy, 29, 325–335.
http://dx.doi.org/10.1080/01926180126501
Hyler, S. E. (1994). Personality Questionnaire, PDQ-41. New York, NY:
New York State Psychiatric Institute.
Leonard, K. E., & Homish, G. G. (2008). Predictors of heavy drinking and
drinking problems over the first 4 years of marriage. Psychology of
Addictive Behaviors, 22, 25–35. http://dx.doi.org/10.1037/0893-164X
.22.1.25
Lewis, E. T., Jamison, A. L., Ghaus, S., Durazo, E. M., Frayne, S. M.,
Hoggatt, K. J., . . . Cucciare, M. A. (2016). Receptivity to alcohol-related
care among U.S. women Veterans with alcohol misuse. Journal of
Addictive Diseases, 35, 226 –237. http://dx.doi.org/10.1080/10550887
.2016.1171670
Litman, G. K., Stapleton, J., Oppenheim, A. N., & Peleg, M. (1983). An
instrument for measuring coping behaviours in hospitalized alcoholics:
Implications for relapse prevention treatment. British Journal of Addic-
tion, 78, 269 –276. http://dx.doi.org/10.1111/j.1360-0443.1983
.tb02511.x
Litt, M. D., Kadden, R. M., & Tennen, H. (2015). Network Support
treatment for alcohol dependence: Gender differences in treatment
mechanisms and outcomes. Addictive Behaviors, 45, 87–92. http://dx
.doi.org/10.1016/j.addbeh.2015.01.005
Longabaugh, R., Wirtz, P. W., Zweben, A., & Stout, R. L. (1998). Network
support for drinking, alcoholics anonymous and long-term matching
effects. Addiction, 93, 1313–1333. http://dx.doi.org/10.1046/j.1360-
0443.1998.93913133.x
McCrady, B. S. (2004). To have but one true friend: Implications for
practice of research on alcohol use disorders and social network. Psy-
chology of Addictive Behaviors, 18, 113–121. http://dx.doi.org/10.1037/
0893-164X.18.2.113
McCrady, B. S., Epstein, E. E., Cook, S., Jensen, N., & Hildebrandt, T.
(2009). A randomized trial of individual and couple behavioral alcohol
treatment for women. Journal of Consulting and Clinical Psychology,
77, 243–256. http://dx.doi.org/10.1037/a0014686
McCrady, B. S., Epstein, E. E., Cook, S., Jensen, N. K., & Ladd, B. O.
(2011). What do women want? Alcohol treatment choices, treatment
entry and retention. Psychology of Addictive Behaviors, 25, 521–529.
http://dx.doi.org/10.1037/a0024037
McCrady, B. S., Epstein, E. E., Hallgren, K. A., Cook, S., & Jensen, N. K.
(2016). Women with alcohol dependence: A randomized trial of couple
versus individual plus couple therapy. Psychology of Addictive Behav-
iors, 30, 287–299. http://dx.doi.org/10.1037/adb0000158
McCrady, B. S., Epstein, E. E., & Hirsch, L. S. (1999). Maintaining change
after conjoint behavioral alcohol treatment for men: Outcomes at 6
months. Addiction, 94, 1381–1396. http://dx.doi.org/10.1046/j.1360-
0443.1999.949138110.x
Miller, W. R., & Tonigan, J. S. (1996). Assessing drinkers’ motivations for
change: The SOCRATES. Psychology of Addictive Behaviors, 10, 81–
89. http://dx.doi.org/10.1037/0893-164X.10.2.81
Rogers, J. L., Howard, K. I., & Vessey, J. T. (1993). Using significance
tests to evaluate equivalence between two experimental groups. Psycho-
logical Bulletin, 113, 553–565. http://dx.doi.org/10.1037/0033-2909.113
.3.553
 FEMALE-SPECIFIC CBT FOR ALCOHOL 15

Rosenthal, R. N. (2013). Treatment of persons with substance use disorder
and co-occurring other mental disorders. In B. S. McCrady & E. E.
Epstein (Eds.), Addictions: A comprehensive guidebook (pp. 659 –707).
New York, NY: Oxford University Press.
Sobell, L. C., & Sobell, M. B. (2003). Alcohol consumption measures. In
J. P. Allen & V. Wilson (Eds.), Assessing alcohol problems (2nd ed., pp.
75–99). Rockville, MD: National Institute on Alcohol Abuse and Alco-
holism.
Sobell, L. C., Sobell, M. B., & Agrawal, S. (2009). Randomized controlled
trial of a cognitive-behavioral motivational intervention in a group
versus individual format for substance use disorders. Psychology of
Addictive Behaviors, 23, 672– 683. http://dx.doi.org/10.1037/a0016636
Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for
assessing the quality of marriage and similar dyads. Journal of Marriage
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
holism, Clinical and Experimental Research, 29, 612– 621. http://dx.doi
.org/10.1097/01.ALC.0000158832.07705.22
Velasquez, M. M., & Stotts, A. L. A. (2003). Substance abuse and depen-
dence disorders in women. In M. Kopala & M. Keitel (Eds.), Handbook
of counseling women (pp. 482–502). Thousand Oaks, CA: Sage. http://
dx.doi.org/10.4135/9781452229546.n29
Walitzer, K. S., & Dearing, R. L. (2006). Gender differences in alcohol and
substance use relapse. Clinical Psychology Review, 26, 128 –148. http://
dx.doi.org/10.1016/j.cpr.2005.11.003
Williams, J. B., Gibbon, M., First, M. B., Spitzer, R. L., Davies, M., &
Borus, J. (1992). The Structured Clinical Interview for DSM–III–R
(SCID) II. Multisite test-retest reliability. Archives of General Psychia-
try, 49, 630 – 636. http://dx.doi.org/10.1001/archpsyc.1992.018200
80038006

Witkiewitz, K., Finney, J. W., Harris, A. H. S., Kivlahan, D. R., &

and the Family, 38, 15–28. http://dx.doi.org/10.2307/350547
Kranzler, H. R. (2015). Recommendations for the design and analysis of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Stout, R. L., Wirtz, P. W., Carbonari, J. P., & Del Boca, F. K. (1994).
Ensuring balanced distribution of prognostic factors in treatment out-
come research. Journal of Studies on Alcohol, 12, 70 –75. http://dx.doi
.org/10.15288/jsas.1994.s12.70
Sword, W., Jack, S., Niccols, A., Milligan, K., Henderson, J., & Thabane,
L. (2009). Integrated programs for women with substance use issues and
their children: A qualitative meta-synthesis of processes and outcomes.
Harm Reduction Journal, 6, 32. http://dx.doi.org/10.1186/1477-7517-
6-32
Timko, C., Finney, J. W., & Moos, R. H. (2005). The 8-year course of
alcohol abuse: Gender differences in social context and coping. Alco-
treatment trials for alcohol use disorders. Alcoholism: Clinical and
Experimental Research, 39, 1557–1570. http://dx.doi.org/10.1111/acer
.12800
Zweig, R. D., McCrady, B. S., & Epstein, E. E. (2009). Investigation of the
psychometric properties of the Drinking Patterns Questionnaire. Addic-
tive Disorders & Their Treatment, 8, 39 –51. http://dx.doi.org/10.1097/
ADT.0b013e3181690c8e
Received June 20, 2017
Revision received September 28, 2017
Accepted September 28, 2017 䡲

E-Mail Notification of Your Latest Issue Online!

Would you like to know when the next issue of your favorite APA journal will be available
online? This service is now available to you. Sign up at https://my.apa.org/portal/alerts/ and you will
be notified by e-mail when issues of interest to you become available!