Psychology of Addictive Behaviors © 2013 American Psychological Association

2014, Vol. 28, No. 2, 619 – 624 0893-164X/14/$12.00 DOI: 10.1037/a0034127

BRIEF REPORT

Ecological Momentary Assessment of the Effects of Craving and Affect on

Risk for Relapse During Substance Abuse Treatment
Todd M. Moore, Amanda Seavey, Kathrin Ritter, James K. McNulty, Kristina C. Gordon,
and Gregory L. Stuart
University of Tennessee
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

The primary goals of this study were to use ecological momentary assessment (EMA) to examine the
This document is copyrighted by the American Psychological Association or one of its allied publishers.

extent to which proximal factors (cravings and affect) were temporally associated with relapse, and to
assess the role of distal factors (e.g., coping styles) in moderating these associations. We also examined
whether using EMA procedures impacted relapse rates. A sample of 100 male (n ⫽ 66) and female
(n ⫽ 34) patients entering outpatient treatment for substance abuse completed a baseline assessment of
substance dependence, stress, social support, coping styles, family history of substance abuse, and
self-efficacy. Half of the participants were randomly assigned to receive training in using a handheld
computer to respond to 3 random prompts each day for 4 months regarding cravings, affect, and
substance use. All participants completed 2- and 4-month follow-up assessments of substance use.
Results showed that using EMA procedures did not influence relapse rates. Hierarchical linear modeling
showed that individuals who reported an increase in cravings on a given prompt were 14 times more
likely to report relapse on the subsequent prompt than individuals who did not experience an increase in
cravings. In addition, women, older individuals, and individuals who use distraction and disengagement
as coping styles were at increased risk for relapse when experiencing an increase in cravings. Individuals
who use acceptance as a coping style were at decreased risk for relapse when experiencing an increase
in cravings. The study highlights the importance of tailoring treatments to address the needs of particular
individuals and risk factors.

Keywords: relapse, craving, coping, electronic diary, ecological momentary assessment

Although treatments designed to reduce substance use show
promise, up to 50% of patients relapse within 6 months of treat-
ment (e.g., McKay & Weiss, 2001), and up to 40% cycle through
periods of relapse, treatment reentry, and recovery (e.g., Scott,
Foss & Dennis, 2005). Though operational definitions of relapse
vary widely, this study used the commonly recognized definition
of relapse of a return to any use of a substance (see Chung &
Maisto, 2006; Witkiewitz & Marlatt, 2007, for reviews), and this
definition has been used in prior ecological momentary assessment
(EMA) studies of relapse (e.g., Marhe, Waters, van de Wetering, &
Franken, 2013). Given the high rate and chronicity of relapse, it is
This article was published Online First October 14, 2013.
Todd M. Moore, Amanda Seavey, Kathrin Ritter, James K. McNulty,
Kristina C. Gordon, and Gregory L. Stuart, Department of Psychology,
University of Tennessee.
James K. McNulty is now at Department of Psychology, Florida State
University.
This research was supported by Grant R03DA025781 from the National
Institute on Drug Abuse awarded to Todd M. Moore. We also want to
express our tremendous gratitude to all the staff and patients at Cornerstone
of Recovery for supporting this research.
Correspondence concerning this article should be addressed to Todd M.
Moore, University of Tennessee, Department of Psychology, Austin Peay
Building (416F), Knoxville, TN 37996. E-mail: tmoore24@utk.edu
imperative to better understand the complex interplay of risk
factors in the relapse process to improve treatment outcome.
Witkiewitz and Marlatt’s (2004, 2007) Systems Model views relapse
as a complex process involving links between distal and proximal
processes occurring during high-risk situations that increase the
risk for relapse. Distal risk factors are stable or enduring charac-
teristics of the individual or his or her environment (e.g., coping
skills, chronic stress) theorized to increase risk for relapse (Mc-
Kay, Franklin, Patapis, & Lynch, 2006; Witkiewitz & Marlatt,
2007). Studies examining the influence of these factors on relapse
have yielded mixed results. Theoretically, these mixed results are
not surprising, given that distal risk factors may not exert main
effects on relapse, but rather, they may influence relapse by
interacting with proximal risk factors. Proximal risk factors are the
temporally immediate precursors to relapse, with cravings and
affect being among the most robust (Witkiewitz & Marlatt, 2004).
Although most research examining proximal factors used retro-
spective questionnaires, recent advances in technology have af-
forded the opportunity to more precisely measure within-day vari-
ations of factors that may increase immediate risk for relapse.
In one of the first studies to use EMA to examine relapse to
alcohol, Krahn, Bohn, Henk, Grossman, and Gosnell (2005) re-
cruited men in alcohol treatment who used handheld computers to
answer random prompts about cravings and affect. Results of
cluster analyses indicated that clusters differed in terms of levels of

619
 620 MOORE, SEAVEY, RITTER, MCNULTY, GORDON, AND STUART
cravings and negative affect but not relapse. Cooney et al. (2007)
recruited patients from outpatient substance abuse programs who
used handheld computers to respond to random prompts assessing
alcohol cravings, mood states, and alcohol abstinence self-
efficacy. Relapse was predicted by lower ratings of self-efficacy to
resist drinking, and a trend was found for increases in cravings to
predict relapse. Positive and negative affect did not predict relapse.
Holt, Litt, and Cooney (2012) did not find an association between
cravings or negative affect and risk for relapse among a sample of
men and women in concurrent alcohol and tobacco treatment. In
regard to illicit substances, studies using EMA consistently dem-
onstrate that increases in cravings are associated with increased
temptations to use and actual relapse to heroin and cocaine (Marhe
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
et al., 2013; Waters, Marhe, & Franken, 2012), ecstasy (Hopper et
This document is copyrighted by the American Psychological Association or one of its allied publishers.
al., 2006), and any substance (Johnson, Barrault, Nadeau, &
Swendsen, 2009). In the only known study to examine both crav-
ings and negative affect, Epstein et al. (2009) found that negative
affect and cravings (“tempted to use”) were related to increased
risk for relapse to cocaine use, and feeling worried was related to
relapse to heroin use. Moreover, related research on smoking
lapses demonstrates that negative affect is associated with in-
creased risk for smoking lapses (e.g., Shiffman et al., 2007),
especially rapid increases in negative affect (e.g., Shiffman &
Waters, 2004).
Taken together, studies examining distal factors yield inconsis-
tent results, whereas studies assessing proximal factors suggest a
strong association between negative affect, cravings, and risk for
relapse. The inconsistency across distal and proximal factors sug-
gests the need for a more nuanced approach to examining the
relapse process. To the extent that proximal factors are operating
in episodes of relapse, they should be linked to distal risk factors
according to Witkiewitz and Marlatt (2004, 2007). For example,
individuals with poor coping skills (distal factor) may have limited
resources to manage increases in cravings or negative affect (prox-
imal factors), which may increase risk of relapse. However, there
is a paucity of data examining the confluence of distal and prox-
imal factors on relapse. The previously described EMA studies
demonstrate that EMA technology can be invaluable for assessing
the process of relapse, but a number of limitations greatly reduce
their generalizability. Namely, these studies did not assess distal
risk factors, did not include a non-EMA control group to rule out
the possibility of reactivity to assessment, and assessed risk for
relapse that occurred usually within the first week to 1 month of
treatment. The literature clearly needs research examining the
influence of proximal and distal risk factors that may lead to
relapse.
To elucidate the relapse process, this study examined the extent
to which EMA measures of cravings and positive and negative
affect predicted relapse over a 4-month period among men and
women entering intensive outpatient substance abuse treatment. It
was hypothesized that increases in cravings and affect would
predict onset of relapse. Furthermore, it was hypothesized that
individuals who report poor coping skills, low self-efficacy and
social support, and report greater stress and history of family
substance abuse are at elevated risk for relapse when experiencing
increases in cravings and affect. Because little is known about the
potential treatment effects of using EMA measures, we random-
ized participants to two groups— one who used a handheld com-
puter to complete electronic diary entries and one who just com-
pleted study measures. This approach allowed us to rule out the
possibility that using EMA procedures significantly influenced the
relapse process.
Method
Participants
This study was approved by the authors’ institutional review
board. Participants were 101 male (n ⫽ 66) and female (n ⫽ 35)
patients entering intensive outpatient treatment for substance de-
pendence at Cornerstone of Recovery in Louisville, Tennessee. All
patients completed a 4-week inpatient treatment prior to beginning
outpatient treatment and were drug-free at the beginning of out-
patient treatment based on urine testing conducted by the facility.
Standard treatment lasts 8 weeks and emphasizes the 12-step
recovery model. Inclusion criteria were (a) at least 18 years old; (b)
met criteria for dependence on alcohol or an illicit substance; and
(c) ability to read and speak English. Participants were randomized
to the EMA (n ⫽ 51) and Non-EMA (n ⫽ 50) groups. One person
(female) dropped out before completing the handheld computer
training, leaving a final sample of 50 patients in each group for
data analysis.
Baseline and Follow-Up Measures
The substance use module of the Structured Clinical Interview
for DSM–IV (SCID-P; First, Spitzer, Gibbon, & Williams, 1995)
assessed diagnostic criteria for Axis I substance use disorders.
Stress was assessed with the Life Experiences Survey (Sarason,
Johnson, & Siegel, 1978), which assesses the stressfulness of life
events occurring in the past 6 months. Social support was assessed
using the Social Support Questionnaire (Sarason, Levine, Basham, &
Sarason, 1983), which measures availability of, and satisfaction
with, social supports. A modified version of the Family Tree
Questionnaire (Mann, Sobell, Sobell, & Pavan, 1985) was used to
assess family history of alcohol and drug abuse. General self-
efficacy was assessed with the Self-Efficacy Scale (Sherer et al.,
1982). Fourteen coping styles were assessed with the Brief COPE
(Carver, 1997). All measures evidenced acceptable psychometric
properties.
Two and 4 months following baseline, participants completed
measures of alcohol (Alcohol Use Disorders Identification Test;
AUDIT; Saunders, Aasland, Babor, De La Fuente, & Grant, 1993)
and drug use (Drug Use Disorders Identification Test; DUDIT;
Stuart, Moore, Kahler, & Ramsey, 2003) to assess the occurrence
of relapse in the non-EMA group and to compare relapse rates
across groups. Specifically, this allowed us to examine whether
any reactivity to assessment that may occur from using the hand-
held computers impacted relapse rates. The AUDIT and DUDIT
include items that assessed use of any substance since the prior
assessment.
Ecological Momentary Assessment
The electronic diary assessments were obtained using the HP
iPAQ 110 handheld computer and SnapSurvey Professional
Edition® software package. Participants in the EMA group were
prompted three times daily on a quasi-random basis (once during
 EMA ASSESSMENT OF RELAPSE 621

every 4-hr block between 9:00 a.m. and 9:00 p.m.) to respond to Table 1
questions about cravings and affect (independent variables) and Baseline Participant Characteristics (N ⫽ 100)
substance use (dependent variable). For each prompt, participants
were asked to rate, from 1 ⫽ none to 5 ⫽ very much, their current Characteristic
craving level for each of the major classes of drugs (alcohol, Age (M, SD) 33.16 (9.94)
marijuana, cocaine, etc.), and their level of positive affect (happy, Sex (%)
excited, determined, energetic, and enthusiastic) and negative af- Males 66.00
fect (angry, irritable, jittery, afraid, nervous, scared, and hostile), Females 34.00
Race (%)
adapted from the PANAS-X (Watson & Clark, 1994). Though not Caucasian 93.00
a PANAS-X item, the term “stressed” was also included in the African American 3.00
measure of negative affect, given past research documenting the Asian American 1.00
association between stress and relapse (e.g., Sinha, 2001). Partic- Other 3.00
Education (%)
ipants also indicated whether they used any substances (yes/no)
Grade/high school 24.00
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

since the last prompt (presented on the screen as a list of common Attended college 42.00
This document is copyrighted by the American Psychological Association or one of its allied publishers.

substances). Responses were sent via wireless encrypted commu-
nication from the handheld computers to a stand-alone secure
computer.
Procedures
Patients were informed about the study by treatment staff during
standard intake sessions and were invited to visit a member of the
research team. Following informed consent, participants were ad-
ministered the SCID-P substance use module and asked to com-
plete baseline questionnaires. They returned at a designated time
with the completed packet and were randomized to the EMA and
non-EMA groups. Participants in the EMA group were trained to
use a handheld computer to complete diary entries three times each
day for 4 months. All participants were asked to return 2 and 4
months following baseline to complete follow-up questionnaires
and address any questions about study participation. Participants in
the non-EMA group did not receive a handheld computer but
completed all measures. All participants earned $25 for completing
the baseline and follow-up assessments ($75 total). Participants in
the EMA group received $1.00 for each on-time (completed within
1 hr of prompting) diary entry and an additional $7 for each week
of on-time responding to at least 75% of the prompts.
Results
Descriptive Statistics
For the overall sample, the average age was 33.2 years (SD ⫽
9.9; range ⫽ 18 to 58), with the majority being Caucasian (93%;
see Table 1 for sample characteristics). Although relatively fewer
females participated in the study, they were fairly equally repre-
sented across the EMA (n ⫽ 19) and non-EMA (n ⫽ 15) groups.
Among the sample, 47% received an alcohol dependence diagnosis
and 83% received a drug dependence diagnosis (cannabis, 24%;
cocaine, 28%; other stimulants, 16%; opioids, 41%; sedatives,
26%; hallucinogens, 5%; other, 8%; and polydrug dependence
without alcohol, 20%). A total of 7,618 of 18,000 random prompts
were answered (M ⫽ 152.36 [42.3%], SD ⫽ 96.35, Mdn ⫽ 139,
range ⫽ 12 to 333 out of 360 possible). Compliance rates for
completing the 2- and 4-month follow-ups were 83% and 67%,
respectively.
EMA Methods and Relapse
Defining relapse as any use of substances, 35% of the partici-
pants relapsed during the study. The rate of relapse for the EMA
College degree 20.00
Graduate degree 14.00
Employment (%)
Unemployed/disability 57.00
Employed part time 8.00
Employed full time 35.00
Annual income (%)
⬍$25,000 38.00
$25,001–$50,000 24.00
$50,001–$100,000 28.00
⬎$100,000 10.00
group (38%) was not significantly different than the non-EMA
group (32%). Of the 19 patients who relapsed in the EMA group,
12 reported relapse via the handheld computers, and these reports
were used in all subsequent analyses. Analyses including and
excluding the remaining seven patients who reported relapse only
on the AUDIT or DUDIT at follow-up were virtually identical, so
these participants were retained in all analyses.
Proximal Predictors of Relapse
To test our first hypothesis, that increases in cravings and affect
would predict onset of relapse, we used a lagged two-level model
analysis using HLM 6 (Raudenbush, Bryk, & Congdon, 2004).
Analyses examined the extent to which cravings and affect re-
ported on a given diary entry predicted relapse on the subsequent
diary entry in both separate and simultaneous analyses (length of
time between reports of relapse via diary entries and the preceding
diary entry was no more than 12 hr). The association between
cravings, negative affect, and substance use following first use of
a substance are not included in these models. Cravings and affect
were person-centered at Level 1 and the individual means were
entered on the intercept at Level 2, which was allowed to vary
randomly across people.
Results showed that, controlling for average levels of cravings,
individuals who reported a sudden increase in cravings were nearly
14 times more likely (OR ⫽ 13.81, 95% CI [1.09, 175.54]) to
relapse than individuals who did not experience an increase in
cravings, t(49) ⫽ 2.07, p ⬍ .05 (effect size ⫽ 1.45). Results
showed that after controlling for average levels of negative
affect, individuals who reported a sudden increase in negative
affect were somewhat more likely (OR ⫽ 1.13, 95% CI [.99, 1.28])
to relapse than individuals who did not experience an increase in
negative affect, t(49) ⫽ 1.99, p ⫽ .051. No effects were found for
positive affect. However, after controlling for affect, individuals
 622 MOORE, SEAVEY, RITTER, MCNULTY, GORDON, AND STUART

who reported a sudden increase in cravings were over 3 times more
likely (OR ⫽ 3.49, 95% CI [1.84, 6.64]) to relapse than individuals
who did not experience an increase in cravings, t(49) ⫽ 3.91, p ⬍
.001 (effect size ⫽ .69). Results further showed that individuals
who reported lower average cravings were more likely to relapse
following a sudden increase in cravings than were individuals who
reported higher average cravings, t(48) ⫽ ⫺2.12, p ⬍ .05. Nega-
tive and positive affect were not associated with relapse after
controlling for cravings and subsequent analyses excluded affect.
Effects of Distal Factors on Moderating the
Craving/Relapse Association
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
than controls. Age moderated the association between cravings and
relapse, wherein older individuals were significantly more likely to
relapse when reporting a sudden increase in cravings than controls.
Finally, gender moderated the association between cravings and
relapse, wherein women were 3.36 times more likely to relapse
when reporting a sudden increase in cravings compared with men.
Discussion
The findings of this study significantly extend the relapse liter-
ature by being the first to demonstrate that cravings emerged as the
only unique temporal predictor of relapse, and that people with
lower average levels of cravings are at increased risk for relapse

Analyses using HLM examined whether distal factors (entered when they experience a sudden spike in cravings. The lack of an
This document is copyrighted by the American Psychological Association or one of its allied publishers.

as grand-centered continuous variables, except for gender, which
was dummy-coded, such that 0 ⫽ male and 1 ⫽ female) exerted an
influence on relapse. None of the distal factors (including age and
gender as exploratory factors) predicted relapse. However, a num-
ber of significant cross-level interactions indicated that distal fac-
tors moderated the association between cravings and relapse (see
Table 2). Three coping styles interacted with cravings to increase
risk for relapse. Specifically, individuals who reported greater use
of coping via distraction and behavioral disengagement (e.g., giv-
ing up attempt to cope), and less use of coping via acceptance,
were significantly more likely (ORs ⫽ 2.3 to 6.5) to relapse when
reporting a sudden increase in cravings than controls. Further,
family history of drug abuse evidenced a trend effect, wherein
individuals who reported a greater history of family drug abuse
were somewhat more likely to relapse (OR ⫽ 2.60; p ⫽ .053;
effect size ⫽ .52) when reporting a sudden increase in cravings
association between cravings and relapse among individuals with
high average levels of cravings may be due to a ceiling effect. This
study is also the first to find that cravings lead to relapse for
different people in different ways, with greatest risk when those
experiencing an increase in cravings are women, older individuals,
individuals with a family history of drug abuse, and individuals
who used distraction and disengagement or rarely used acceptance
as coping strategies, and those who tended to experience relatively
lower levels of cravings on average. This study highlights the
importance of considering the benefits of tailoring treatment to
address the needs of individuals who may be at greatest risk of
relapse during periods of heightened cravings. It also suggests that
patients may benefit from incorporating greater emphasis on help-
ing patients improve coping skills. This is consistent with the
ambivalence model of craving in which affect and craving are
critical factors in exacerbating or attenuating approach inclinations

Table 2
Moderating Effect of Distal Factors on Associations Between Cravings and Relapse

Relapse
Distal Factors ⫻ Cravings t Odds ratio CI Effect size (d)
ⴱ
Age 2.08 1.24 1.01, 1.53 .12
Gender 4.58ⴱⴱⴱ 3.36 1.98, 5.71 .67
Stress ⫺0.51 0.96 0.81, 1.13 .02
Family history
Alcohol 0.58 1.29 0.53, 3.17 .14
Drugs 1.99† 2.57 0.99, 6.72 .52
Self-efficacy 0.14 1.02 0.81, 1.27 .01
Social support 0.70 3.68 0.09, 155.89 .72
Coping styles
Acceptance ⫺3.25ⴱⴱ 3.32 1.58, 6.94 .66
Active coping ⫺1.34 0.59 0.27, 1.30 .37
Behavioral disengagement 2.54ⴱ 2.35 1.20, 4.60 .47
Denial 0.76 1.46 0.53, 4.01 .21
Humor 0.07 1.04 0.31, 3.56 .02
Planning 0.05 1.03 0.33, 3.19 .02
Positive reframing ⫺0.39 0.88 0.47, 1.67 .07
Religion 0.20 1.09 0.47, 2.50 .05
Self-blame 0.97 2.12 0.45, 10.09 .42
Self-distraction 2.12ⴱ 6.54 1.11, 38.71 1.04
Substance use 1.02 1.40 0.72, 2.70 .19
Use of emotional support 0.61 1.50 0.40, 5.60 .22
Use of instrumental support ⫺0.18 0.81 0.07, 9.26 .11
Venting ⫺0.90 0.39 0.05, 3.22 .41
Note. CI ⫽ confidence interval.
†
p ⬍ .10. ⴱ p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ
p ⬍ .001.
 EMA ASSESSMENT OF RELAPSE
to substance use cues (Breiner, Stritzke, & Lang, 1999; Schlauch,
Gwynn-Shapiro, Stasiewicz, Molnar, & Lang, 2013). Thus,
mindfulness-based additions to treatments may reduce relapse by
helping patients better manage affect and cravings via use of
acceptance-based coping (e.g., Witkiewitz, Marlatt, & Walker,
2005).
A major strength of this study was the use of handheld com-
puters to assess proximal risk factors. The use of state-of-the-art
technology afforded the opportunity to address limitations of ret-
rospective studies (i.e., retrospective recall bias) and to more
precisely examine factors that immediately precede relapse with
minimal concern potentially reactivity to assessment would in-
crease the relapse rate, especially given that we found no differ-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ence in relapse rates between groups. By demonstrating that pa-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
tients in treatment will provide frequent responses about sensitive
information, this study also suggests that handheld computers and
similar devices may be of great utility for primary or adjunctive
treatment delivery, especially for patients who might otherwise not
seek treatment or have limited capabilities to attend treatment.
Although this study possessed a number of strengths, one lim-
itation is that some EMA patients reported relapse on follow-up
self-report measures but not on the handheld computers. This may
have stemmed from fears about negative consequences associated
with reporting (e.g., treatment termination), leading them to report
relapse only after completing treatment. Future studies should use
a calendar-based interview, frequent biochemical verification, and
collateral reports to assess any non-EMA reported relapses to more
accurately calculate the role of proximal variables on the risk for
relapse. Another limitation is the relatively low EMA compliance
rate. Studies of similar length and design obtained compliance
rates at or above 65% (e.g., Epstein et al., 2009; Holt et al., 2012).
These studies incorporated counselor awareness of compliance
rates, study discharge for low compliance, and/or repeated prompts
when initial prompts were not answered. We considered these
approaches during the design phase but opted to use monetary
incentives to enhance compliance rather than methods that may
reduce confidentiality, add a potential punishment for noncompli-
ance, or risk burnout and reduced compliance from repeated
prompts, respectively. However, use of these procedures may
explain the higher compliance rate in these studies, and future
researchers should carefully consider the strengths and weaknesses
of utilizing specific methods to improve compliance.
Another limitation, given the small sample, is that we lacked
sufficient power to detect what may be a significant association
between affect and relapse, and we lacked sufficient power to
examine whether risk factors lead to relapse in the same way for
different substances. Future research with larger samples could
ascertain the extent to which these findings hold specifically for
alcohol and each drug class, and whether negative affect plays a
larger role in predicting relapse as has been found in prior research
(e.g., Shiffman & Waters, 2004). However, it should be noted that
effect sizes confirmed the robustness of the results even within a
relatively small sample. Despite these limitations, results from this
study support a growing body of research showing that cravings
are strongly related to relapse. Results also show that this relation-
ship is not always a direct one, and that distal factors play an
important role. The treatments that may be most effective in
reducing relapse may need to be tailored and handheld devices
may be useful tools for delivering these interventions.
623
References
Breiner, M. J., Stritzke, W. G. K., & Lang, A. R. (1999). Approaching
avoidance: A step essential to the understanding of craving. Alcohol
Research & Health, 23, 197–206.
Carver, C. S. (1997). You want to measure coping but your protocol’s too
long: Consider the Brief COPE. International Journal of Behavioral
Medicine, 4, 92–100. doi:10.1207/s15327558ijbm0401_6
Chung, T., & Maisto, S. A. (2006). Relapse to alcohol and other drugs use
in treated adolescents: Review and reconsideration of relapse as a
change point in clinical course. Clinical Psychology Review, 26, 149 –
161. doi:10.1016/j.cpr.2005.11.004
Cooney, N. L., Litt, M. D., Cooney, J. L., Pilkey, D. T., Steinberg, H. R., &
Oncken, C. A. (2007). Alcohol and tobacco cessation in alcohol-
dependent smokers: Analysis of real-time reports. Psychology of Addic-
tive Behaviors, 21, 277–286. doi:10.1037/0893-164X.21.3.277
Epstein, D. H., Willner-Reid, J., Vahabzadeh, M., Mezghanni, M., Lin,
J. L., & Preston, K. L. (2009). Real-time electronic diary reports of cue
exposure and mood in the hours before cocaine and heroin craving and
use. Archives of General Psychiatry, 66, 88 –94. doi:10.1001/
archgenpsychiatry.2008.509
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995).
Structured clinical interview for DSM–IV Axis I disorders – Patient ed.
(SCID-I/P, Version 2.0). New York, NY: New York State Psychiatric
Institute.
Holt, L. J., Litt, M. D., & Cooney, N. L. (2012). Prospective analysis of
early lapse to drinking and smoking among individuals in concurrent
alcohol and tobacco treatment. Psychology of Addictive Behaviors, 26,
561–572. doi:10.1037/a0026039
Hopper, J. W., Su, Z., Looby, A. R., Ryan, E. T., Penetar, D. M., Palmer,
C. M., & Lukas, S. E. (2006). Incidence and patterns of polydrug use and
craving for ecstasy in regular ecstasy users: An ecological momentary
assessment study. Drug and Alcohol Dependence, 85, 221–235. doi:
10.1016/j.drugalcdep.2006.04.012
Johnson, E. I., Barrault, M., Nadeau, L., & Swendsen, J. (2009). Feasibility
and validity of computerized ambulatory monitoring in drug-dependent
women. Drug and Alcohol Dependence, 99, 322–326. doi:10.1016/j
.drugalcdep.2008.06.010
Krahn, D. D., Bohn, M. J., Henk, H. J., Grossman, J. L., & Gosnell, B.
(2005). Patterns of urges during early abstinence in alcohol-dependent
subjects. The American Journal on Addictions, 14, 248 –255. doi:
10.1080/10550490590949424
Mann, R. E., Sobell, L. C., Sobell, M. B., & Pavan, D. (1985). Reliability
of a family tree questionnaire for assessing family history of alcohol
problems. Drug and Alcohol Dependence, 15, 61– 67. doi:10.1016/0376-
8716(85)90030-4
Marhe, R., Waters, A. J., van de Wetering, B. J. M., & Franken, I. H. A.
(2013). Implicit and explicit drug-related cognitions during detoxifica-
tion treatment are associated with drug relapse: An ecological momen-
tary assessment study. Journal of Consulting and Clinical Psychology,
81, 1–12. doi:10.1037/a0030754
McKay, J. R., Franklin, T. R., Patapis, N., & Lynch, K. G. (2006).
Conceptual, methodological, and analytical issues in the study of re-
lapse. Clinical Psychology Review, 26, 109 –127. doi:10.1016/j.cpr.2005
.11.002
McKay, J. R., & Weiss, R. V. (2001). A review of temporal effects and
outcome predictors in substance abuse treatment studies with long-term
follow-ups: Preliminary results and methodological issues. Evaluation
Review, 25, 113–161. doi:10.1177/0193841X0102500202
Raudenbush, S. W., Bryk, A. S., & Congdon, R. T. (2004). HLM: Hier-
archical linear modeling with the HLM/2L and HLM/3L programs.
Chicago, IL: Scientific Software International.
Sarason, I. G., Johnson, J. H., & Siegel, J. M. (1978). Assessing the impact
of life changes: Development of the Life Experiences Survey. Journal of
 624 MOORE, SEAVEY, RITTER, MCNULTY, GORDON, AND STUART
Consulting and Clinical Psychology, 46, 932–946. doi:10.1037/0022-
006X.46.5.932
Sarason, I. G., Levine, H. M., Basham, R. B., & Sarason, B. R. (1983).
Assessing social support: The Social Support Questionnaire. Journal of
Personality and Social Psychology, 44, 127–139. doi:10.1037/0022-
3514.44.1.127
Saunders, J. B., Aasland, O. G., Babor, T. F., De La Fuente, J. R., & Grant,
M. (1993). Development of the Alcohol Use Disorders Identification
Test (AUDIT): WHO Collaborative Project on Early Detection of Per-
sons with Harmful Alcohol Consumption–II. Addiction, 88, 791– 804.
doi:10.1111/j.1360-0443.1993.tb02093.x
Schlauch, R. C., Gwynn-Shapiro, D., Stasiewicz, P. R., Molnar, D. S., &
Lang, A. R. (2013). Affect and craving: Positive and negative affect are
differentially associated with approach and avoidance inclinations. Ad-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
dictive Behaviors, 38, 1970 –1979. doi:10.1016/j.addbeh.2012.12.003
Scott, C. K., Foss, M. A., & Dennis, M. L. (2005). Pathways in the
This document is copyrighted by the American Psychological Association or one of its allied publishers.
relapse-treatment-recovery cycle over 3 years. Journal of Substance
Abuse Treatment, 28, S63–S72. doi:10.1016/j.jsat.2004.09.006
Sherer, M., Maddux, J., Mercadante, B., Prentice-Dunn, S., Jacobs, B., &
Rogers, R. (1982). Self-Efficacy Scale: Construction and validation.
Psychological Reports, 51, 663– 671. doi:10.2466/pr0.1982.51.2.663
Shiffman, S., Balabanis, M. H., Gwaltney, C. J., Paty, J. A., Gnys, M.,
Kassel, J. D., . . . Paton, S. M. (2007). Prediction of lapse from
associations between smoking and situational antecedents assessed by
ecological momentary assessment. Drug and Alcohol Dependence, 91,
159 –168. doi:10.1016/j.drugalcdep.2007.05.017
Shiffman, S., & Waters, A. J. (2004). Negative affect and smoking lapses:
A prospective analysis. Journal of Consulting and Clinical Psychology,
72, 192–201. doi:10.1037/0022-006X.72.2.192
Sinha, R. (2001). How does stress increase risk of drug abuse and relapse?
Psychopharmacology, 158, 343–359. doi:10.1007/s002130100917
Stuart, G. L., Moore, T. M., Kahler, C. W., & Ramsey, S. E. (2003).
Substance abuse and relationship violence among men court-referred to
batterers’ intervention programs. Substance Abuse, 24, 107–122. doi:
10.1080/08897070309511539
Waters, A. J., Marhe, R., & Franken, I. H. A. (2012). Attentional bias to
drug cues is elevated before and during temptations to use heroin and
cocaine. Psychopharmacology, 219, 909 –921. doi:10.1007/s00213-011-
2424-z
Watson, D. L., & Clark, L. A. (1994). Manual for the Positive and
Negative Affect Schedule (Expanded form). Unpublished manuscript,
University of Iowa, IA.
Witkiewitz, K., & Marlatt, G. A. (2004). Relapse prevention for alcohol
and drug problems: That was zen, this is tao. American Psychologist, 59,
224 –235. doi:10.1037/0003-066X.59.4.224
Witkiewitz, K., & Marlatt, G. A. (2007). Modeling the complexity of
post-treatment drinking: It’s a rocky road to relapse. Clinical Psychology
Review, 27, 724 –738. doi:10.1016/j.cpr.2007.01.002
Witkiewitz, K., & Marlatt, G. A., & Walker, D. (2005). Mindfulness-
based relapse prevention for alcohol and substance use disorders.
Journal of Cognitive Psychotherapy, 19, 211–228. doi:10.1891/jcop
.2005.19.3.211
Received March 19, 2013
Revision received July 2, 2013
Accepted July 8, 2013 䡲