The American Journal on Addictions, 19: 510–514, 2010
C American Academy of Addiction Psychiatry
Copyright ISSN: 1055-0496 print / 1521-0391 online
DOI: 10.1111/j.1521-0391.2010.00082.x
An Examination of Drug Craving Over Time in Abstinent
Methamphetamine Users
Gantt P. Galloway, PharmD,1 Edward G. Singleton, PhD,2 Raymond Buscemi, PsyD,1
Matthew J. Baggott, PhD,1 René M. Dickerhoof, PhD,1 John E. Mendelson, MD,1
for The Methamphetamine Treatment Project Corporate Authors
1
2
Addiction & Pharmacology Research Laboratory, California Pacific Medical Center Research Institute, San Francisco, California
Department of Psychology, Stevenson University, Stevenson, Maryland
Craving for addictive drugs may predict relapse in abstinent addicts. To assess relationships between craving and
use, we examined changes in craving for methamphetamine
(MA) in a sample of 865 outpatients in a multisite 16-week
MA-treatment study. Craving was assessed on a 0–100 scale,
and MA use was assessed by self-report and confirmed by
urinalysis. We hypothesized that the magnitude of craving
would decline (decay) with increased time of abstinence, and
that decay would be greater for more frequent MA users,
and greater for intravenous (IV) users and smokers as compared to those who used MA intranasally. Craving declined
significantly as the number of weeks of consecutive abstinence increased. Rate of decay was greater for IV users and
smokers as compared to both intranasal users and oral users,
but not for more frequent users of MA. Rate of decay was
independent of age, gender, and race/ethnicity. The trajectory to 0 (no) craving was 1 week shorter for females than
males because females had significantly lower pretreatment
craving scores compared to males. This study confirms that
the sooner MA-dependent people are able to quit using and
the longer that they are able to stay abstinent, the more likely
it is that their craving for MA will decrease over time. (Am
J Addict 2010;19:510–514)
INTRODUCTION
Methamphetamine (MA) abuse and dependence continue to challenge both law enforcement and public-health
officials across the United States. The National Drug
Threat Survey of 2007 indicates that MA has been identified throughout the entire United States by state and local
Received August 27, 2009; revised October 28, 2009; accepted
January 7, 2010.
A complete list of the members of the Methamphetamine
Treatment Project is given at the end of this paper. Address correspondence to Dr. Galloway, Addiction & Pharmacology Research
Laboratory, St. Luke’s Hospital, 7th floor, 3555 Cesar Chavez
Street, San Francisco, CA 94110. E-mail: [email protected].
510
agencies as their second-greatest drug threat, after cocaine.
In the western and central United States, MA was by far
the greatest drug threat identified by these agencies.1 Lessfrequent pretreatment MA use is strongly associated with
treatment completion and less use of MA during treatment.2–4 In-treatment MA use is greatest in intravenous
(IV) users, slightly less in smokers, and lower still in intranasal users.5 Women appear to have more severe dependence than men and they begin using at an earlier age and
stay in treatment longer and respond better to treatment as
compared to men.2
An association between craving and in-treatment MA
use has been established.6,7 When craving and MA use
were assessed weekly in outpatients, craving was a significant predictor of subsequent MA use (hereafter referred to
as relapse).8 This relationship held even when controlling
for recent MA use. The predictive validity of craving scores
was greater when use was assessed at the earliest opportunity after craving was assessed (1 week later) than when use
was assessed at more distal timepoints (2–7 weeks after the
craving assessment). The level of craving over time among
inpatients undergoing treatment for MA dependence has
been shown to decline significantly by the end of the first
week of abstinence and to remain stable thereafter.9 The
time course of craving in abstinent outpatients, who may
be exposed to MA-associated cues, has not been reported.
In addition to its importance in predicting relapse, craving also arises as a clinical issue insofar as patients report it to be aversive and wish to know how long it will
persist.
We hypothesized that the magnitude of craving would
decay linearly with increased time of abstinence and that
the rate of MA-craving decay would be greater for more
frequent MA users and greater for IV users and smokers
as compared to intranasal users. We predicted that the rate
of decay would be independent of gender, age, race, and
ethnicity.
METHODS
RESULTS
Design
This study was part of the Center for Substance Abuse
Treatment Methamphetamine Treatment Project (MTP).10
The MTP was conducted in eight community outpatient
settings: the coordinating center was at the University of
California, Los Angeles (UCLA), while the other seven
investigative teams conducted the study at eight sites in
California, Hawaii, and Montana. Subjects were randomly
assigned to receive either the 16-week manualized Matrix
Model or the treatment-as-usual (TAU) at each site.10–12
Inclusion and exclusion criteria are described in other publications.2,10,13 Subjects gave written informed consent according to guidelines for the protection of human research
volunteers of the U.S. Department of Health and Human
Services. Each local Institutional Review Board approved
the study.
Participant Characteristics
Data were based on the entire sample of 865 subjects
with 6,748 total observations and 7.8 ± 5.0 observations
per subject. A total 4,641 observations were MA abstinent
(68.8%). Most of the subjects were white (62%) and female (53%). Mean age was 32.0 ± 8.4 years, frequency of
episodes of MA use (number of times used per day) was
5.5 ± 8.0, duration of MA involvement (number of years
heavy use) was 5.3 ± 5.0, number of treatment visits was
7.8 ± 5.0, and number of weeks of consecutive abstinence
was 2.7 ± 4.3. The majority (80%) of subjects used one or
more substances in addition to MA, primarily marijuana
(60%) and alcohol (63%). The usual route of administration
of MA for the majority of subjects (61%) was smoking, with
more women (64%) than men (58%) indicating that this was
their primary route of administration. Intranasal use was
the second most common usual route (23%) for men (26%)
and for women (21%). IV use the third most common route
reported by 13% of all subjects (15% of men and 11% of
women). Oral ingestion was the primary route of administration for 2% of all subjects (3% of women and for 1% of
men). The majority of subjects used MA less than five times
per day (69%, 80%, 94%, and 84% for smokers, intranasal
users, IV users, and oral users, respectively).
Measures
Craving was assessed weekly and was defined as “an urgent desire, longing, yearning, not just a passing thought.”
It was measured on a 0–100 scale, with endpoint anchors of
“no craving” and “most craving ever experienced.” Subjects
reported the highest level of craving experienced on the previous day. Subjects were asked to report the number of days
of MA use since the previous assessment, and urine samples
were tested weekly for MA by gas chromatography/mass
spectroscopy. A subject was considered to have had an abstinent week when MA was not detected in the urine and
the subject provided a self-report of no use. A subject was
considered to have used MA when urine MA was greater
than or equal to 1,000 ng/mL or the subject provided a
self-report of use. Craving and MA use were assessed once
weekly for up to 16 weeks. Variables measured included the
number of weeks of consecutive abstinence (WA; 0–16),
gender, age (coded younger-older; cutoff > mean = 32,
SD = 8), race, and self-reported frequency of episodes of
MA use per day (coded less frequent–more frequent use;
cutoff > mean = 5, SD = 8).
Data Analyses
First, a fully unconditional Linear Growth Model
(LGM)14 was estimated to test the hypothesis that MA
craving would decay significantly as the number of weeks
of consecutive abstinence increased. Next, univariate conditional LGM models were estimated for frequent MA users,
route of administration, sex, age, and ethnicity separately
to test the hypotheses that the rate of decay is independent
of gender, age, and ethnicity and race but are greater for
more frequent MA users versus less frequent MA users,
and for IV users and smokers as compared to those who
used intranasal MA. All LGM were estimated using SAS
PROC MIXED (SAS Institute Inc., Cary, North Carolina,
USA) and tested at the .05 level of statistical significance.
Galloway et al.
Craving Decays with Increased Time of Abstinence
As hypothesized, craving declined significantly with increased time of abstinence (p < .001). The average participant began with a craving score of 23.8. Craving scores
decreased linearly at a rate of 2.1 points per week. The
trajectory of the latent-growth curve indicated that craving
no longer differed significantly from zero (p > .37) after
10 weeks of consecutive abstinence.
The Rate of Decay of Craving Was Greater for
Intravenous Users and Smokers than for Intranasal
Users or Oral Users
Regarding route of administration, the rate of decay of
craving was greater for IV users and smokers as compared
to both intranasal users (p < .02) and oral users (p < .02).
However, IV users had a significantly (all p < .03) higher
initial mean craving score (30.2) than smokers (24.2), intranasal users (21.0), and oral users (13.2). Figure 1 shows
how pretreatment craving levels affected the trajectory of
the latent-growth curve, such that MA craving decayed to
zero after 9 weeks of consecutive abstinence in oral and IV
users, after 10 weeks of consecutive abstinence in smokers,
and after 8 weeks of consecutive abstinence in intranasal
users.
The Decay of Craving Is Independent of Frequency
of Use, Age, Race, and Gender
Frequency of use was not a significant predictor of
level of craving (p > .48) and there was no significant
November–December 2010
511
a. By route of administration
35
Smoke
VAS craving score
30
Intranasal
25
Oral
Injection
20
15
10
5
0
0
1
2
3
4
5
6
7
8
9
10
11
b. By gender
30
Female
VAS craving score
25
Male
20
15
10
5
0
0
1
2
3
4
5
6
7
8
9
10
11
c. Unconditional and adjusted models
25
Unconditional model
Adjusted model
VAS craving score
20
15
10
5
0
0
1
2
3
4
5
6
7
8
9
10
11
Weeks consecutive abstinence
FIGURE 1. Craving decay over time. (a) By route of administration. (b) By gender. (c) Unconditional and adjusted models.
difference in rate of decay between less frequent and more
frequent users (p > .10). Neither age nor race/ethnicity
were significant predictors of pretreatment level of craving
(both p > .47), and there was no significant interaction for
512
Drug Craving Over Time
either age or race/ethnicity and rate of decay (both p > .20).
There also was no significant interaction between gender
and rate of decay (p > .49). Irrespective of gender, craving scores decreased 2.2 points per week of consecutive
November–December 2010
abstinence. However, women had significantly lower mean
craving scores (p < .046) than did men (22.0 and 25.5, respectively). As a result of this difference, the trajectory to
zero craving is 10 weeks for women compared with 11 weeks
of consecutive abstinence for men).
Correlates of Intital Craving and Rate of Decay
of Craving Remained Significant in Multivariate
Models
We estimated a multivariate conditional LGM controlling for all variables in the study that significantly effected
the value of pretreatment craving (gender and route of administration) and the rate of change in the value of craving over time (weeks of consecutive abstinence and route
of administration). Gender and route of administration
remained significant predictors of the pretreatment level
of craving (all p < .04) and there remained a significant
interaction for weeks of consecutive abstinence with intranasal and oral routes of administration (all p < .02) (see
Fig. 1a and 1b). Nevertheless, weeks of consecutive abstinence remained a significant predictor of the rate of craving
decay (p < .0001). The direction of the effects also remained
unchanged. The variance for the intercept (level of pretreatment craving) changed from 23.8 to 32.2 and the variance in
linear slope (rate at which craving decayed) changed from
−2.1 to −2.9. Figure 1c compares the covariate adjusted
(conditional) model with the unadjusted (unconditional)
model of the relationship between weeks of consecutive abstinence and craving for MA. The overall trajectory of the
models was largely unaffected by the presence of covariates.
DISCUSSION
This study provides the first evaluation of the rate of
decay of craving among abstinent outpatient addicts. In
addition to determining whether there is decay of craving
over time, we wanted to determine if there were differences
among subsets of subjects in the rate of decay of craving.
We hypothesized that craving would decline with increased
time of abstinence and the rate of decay would be independent of age, race, and gender, and would be greater for
more frequent users, and greater for IV users and smokers as compared to those who used intranasally. Relapse is
common in MA-dependent outpatients and most data sets
do not have appreciable numbers of subjects who have sustained abstinence. We analyzed data from a large clinical
trial, which yielded an adequate number of subjects with
prolonged abstinence to assess the course of craving after
cessation of MA use.
We estimated a two-level unconditional LGM and found
that craving decreased with each week of consecutive abstinence. The rate of decay was independent of gender, age,
race, and frequency of use prior to abstinence. We found
that the rate of decay of craving was greater for IV MA
users compared to intranasal users, but found no differGalloway et al.
ence in comparison to smokers. Findings support previous
studies that found IV users are the most impaired MA
users, with smokers nearly as impaired as IV users, and intranasal users showing the least psychological and medical
impairment of the three.5
We previously found that craving was a significant
predictor of subsequent use during outpatient MAdependence treatment.6,8 In a study of the natural history
of MA withdrawal among in-patients, symptom severity
declined in a linear fashion for the duration of the study,
3 weeks.9 We have expanded on this previous work by assessing craving among outpatient MA-dependent subjects
beyond 3 weeks and, similarly, found that craving declines
with increased time of abstinence and that it no longer differs significantly from zero after 10 weeks of abstinence.
We hypothesized that the rate of decay of craving would
be greater in more frequent users but found that the frequency of use was not a significant predictor of level of
craving (p > .48) and there was no significant interaction
between frequency of use and rate of decay (p > .10). This
finding suggests that frequency of use may not be as important a factor as it is believed to be. Our analysis also
confirms prior findings regarding impairment and treatment outcomes of IV users, who show greater levels of
impairment than intranasal users and smokers.5
Our study confirms previous findings regarding the effect of route of administration on treatment outcomes and
suggests that interventions tailored by route of administration warrant additional study. The development and implementation of therapies tailored to users may increase
positive outcomes in MA treatment in-treatment performance variables such as engagement, participant retention,
and abstinence.2
Our study found that women reported lower levels of
baseline craving than men. Past studies that have looked
at women MA users have focused on co-morbid diagnoses such as post-traumatic stress disorder and depression, and personality factors such as impulsivity.15,16 The
difference in reported levels of craving between men and
women in our study suggests two additional lines for future
inquiry. Women may be reporting lower levels of craving
as a result of physiological differences that mitigate the
effects of MA. Women have been found to show diminished amphetamine-stimulated dopamine responses and a
decreased degree of toxicity, as indicated by a lower incidence of emergency department-related deaths involving
MA.17 Alternately, women may be reporting lower levels
of craving as a result of having different subjective reference points for what constitutes an “an urgent desire,
longing, yearning.” Regardless of the cause of these differences, these findings indicate the importance of considering
gender in subsequent craving research.
Potentially important predictors were not included in
the analyses. For example, participants also were users of
alcohol, marijuana, and other substances in addition to
MA. Another reasonable hypothesis is that the duration
November–December 2010
513
of MA involvement was associated with outcome, rather
than frequency of use. Post hoc analyses revealed that the
rate of decay of craving was greater for participants who
used one or more substances in addition to MA compared
to those who used only MA, while the rate of decay of
craving also increased as the duration of MA involvement
increased. However, weeks abstinent retained its significant
relationship with craving (p < .0001), and the magnitude
and direction of the effect (−1.9 points per week) remained
largely unaffected (vs. −2.1) after controlling for these covariates.
Finally, these data enable clinicians to answer one of the
pressing questions that people who enter treatment have:
when will the cravings go away? Patients who inquire about
how long craving for MA will persist can be told that, if
they remain abstinent, it will subside steadily and end in 2
to 3 months.
The research presented in this paper was supported by
grant numbers TI 11440-01, TI 11427-01, TI 11425-01,
TI 11443-01, TI 11484-01, TI 11441-01, TI 11410-01,
and TI11411-01, provided by the Center for Substance
Abuse Treatment (CSAT), Substance Abuse and Mental
Health Services Administration (SAMHSA), US Department of Heath and Human Services; and also by grant P50
DA018179 from the National Institutes of Health, Bethesda,
MD (Dr. Mendelson).
The Methamphetamine Treatment Project Corporate Authors: M. Douglas Anglin, PhD, Richard A. Rawson, PhD,
Patricia Marinelli-Casey, PhD, Joseph Balabis, BA, Richard
Bradway, Alison Hamilton Brown, PhD, Cynthia Burke,
PhD, Darrell Christian, PhD, Judith Cohen, PhD, MPH,
Florentina Cosmineanu, MS, Alice Dickow, BA, Melissa
Donaldson, Yvonne Frazier, Thomas E. Freese, PhD, Cheryl
Gallagher, MA, Gantt P. Galloway, PharmD, Vikas Gulati,
BS, James Herrell, PhD, MPH, Kathryn Horner, BA, Alice
Huber, PhD, Martin Y. Iguchi, PhD, Russell H. Lord, EdD,
Michael J. McCann, MA, Sam Minsky, MFT, Pat Morrisey,
MA, MFT, Jeanne Obert, MFT, MSM, Susan Pennell, MA,
Chris Reiber, PhD, MPH, Norman Rodrigues, Jr., Janice
Stalcup, MSN, DrPH, S. Alex Stalcup, MD, Ewa S. Stamper, PhD, Janice Stimson, PsyD, Sarah Turcotte Manser,
MA, Denna Vandersloot, MEd, Ahndrea Weiner, MS, MFT,
Kathryn Woodward, BA, Joan Zweben, PhD.
Declaration of Interest
The authors report no conflicts of interest. The authors
alone are responsible for the content and writing of this
paper.
514
Drug Craving Over Time
REFERENCES
1. US Department of Justice, USDEA. National Drug Threat
Assessment 2008. Washington, DC: USDEA; 2008.
2. Hillhouse MP, Marinelli-Casey P, Gonzales R, et al. Predicting in-treatment performance and post-treatment outcomes
in methamphetamine users. Addiction. 2007;102(Suppl 1):84–
95.
3. Brecht ML, Greenwell L, von Mayrhauser C, et al. Two-year outcomes of treatment for methamphetamine use. J Psychoactive Drugs.
2006;(Suppl 3):415–426.
4. Maglione M, Chao B, Anglin MD. Correlates of outpatient drug
treatment drop-out among methamphetamine users. J Psychoactive
Drugs. 2000;32:221–228.
5. Rawson RA, Gonzales R, Marinelli-Casey P, et al. Methamphetamine dependence: A closer look at treatment response
and clinical characteristics associated with route of administration in outpatient treatment. Am J Addict. 2007;16:291–
299.
6. Hartz DT, Frederick-Osborne SL, Galloway GP. Craving predicts
use during treatment for methamphetamine dependence: A prospective, repeated-measures, within-subject analysis. Drug Alcohol Depend. 2001;63:269–276.
7. Hammersley R. A digest of memory phenomena for addiction research. Addiction. 1994;89:283–293.
8. Galloway G, Singleton E, The Methamphetamine Treatment Project.
How long does craving predict use of methamphetamine? Assessment
of use one to seven weeks after the assessment of craving. Subst Abuse.
2008;1:63–79.
9. McGregor C, Srisurapanont M, Jittiwutikarn J, et al. The nature,
time course and severity of methamphetamine withdrawal. Addiction.
2005;100:1320–1329.
10. Rawson RA, Marinelli-Casey P, Anglin MD, et al. A multisite comparison of psychosocial approaches for the treatment of methamphetamine dependence. Addiction. 2004;99:708–
717.
11. Galloway GP, Marinelli-Casey P, Stalcup J, et al. Treatment-as-usual
in the methamphetamine treatment project. J Psychoactive Drugs.
2000;32:165–175.
12. Huber A, Lord RH, Gulati V, et al. The CSAT methamphetamine treatment program: Research design accommodations
for “real world” application. J Psychoactive Drugs. 2000;32:149–
156.
13. Cohen JB, Dickow A, Horner K, et al. Abuse and violence history of
men and women in treatment for methamphetamine dependence. Am
J Addict. 2003;12:377–385.
14. Potthoff RF, Roy SN. A generalized multivariate analysis of variance model useful especially for growth curve problems. Biometrika.
1964;51:313–326.
15. Zilberman ML, Tavares H, el-Guebaly N. Relationship between craving and personality in treatment-seeking women with substancerelated disorders. BMC Psychiatry. 2003;3:1.
16. Dluzen DE, Liu B. Gender differences in methamphetamine use and
responses: A review. Gend Med. 2008;5:24–35.
17. Dluzen DE, McDermott JL. Sex differences in dopamine- and
vesicular monoamine-transporter functions. Ann N Y Acad Sci.
2008;1139:140–150.
November–December 2010
Copyright of American Journal on Addictions is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.