ReviewEcological momentary assessment in the investigation of craving and substance use in daily life: A systematic review
Introduction
Individuals with substance use disorders present high rates of lapses and relapses when attempting to quit (Auriacombe, 1997, Auriacombe and Franques, 1994, Hughes et al., 2008, Hunt et al., 1971). Maintaining abstinence is, therefore, a major target in addiction treatment, and the precipitants of relapse have been widely examined (Fatseas et al., 2011, Marlatt and Gordon, 1985). Craving, often defined as the subjective, unwanted desire or urge to use a substance while attempting to abstain, is frequently experienced during abstinence periods and has been proposed as a major contributor to relapse (Weiss, 2005). However, the link between craving and relapse remains controversial (Rohsenow and Monti, 1999, Tiffany, 1990, Wray et al., 2013). The inconsistencies found in the literature concerning this association may be due to limitations of craving assessment in the laboratory such as restrained number of triggers that are far from real-life contexts, or to retrospective recall biases that are increased by assessments in clinical setting (Bradburn et al., 1987, Epstein and Preston, 2003, Hammersley, 1994). For these reasons, failure to find an association of craving with subsequent substance use could be partly explained in that craving is a momentary phenomenon that is difficult to reliably investigate using traditional methods.
Mobile technologies have permitted the collection of real-time data in natural environments and therefore offer a solution to many methodological barriers that characterize clinical research. In particular, ecological momentary assessment (EMA; Stone and Shiffman, 1994) constitutes a rapidly-expanding approach, notably in clinical psychology, behavioral neurosciences and addiction research (aan het Rot et al., 2012, Lukasiewicz et al., 2007, Shiffman et al., 2008, van den Bos et al., 2013). EMA also includes the experience sampling method (ESM; Csikszentmihalyi and Larson, 1987) but uses a broader range of sampling options. In EMA, there are three types of recording: signal-contingent (made in response to a signal); interval-contingent (made after a fixed period of time); event-contingent (made when a specific event occurs). Of these different methods, only signal-contingent recording (repeated over a short time frame) constitutes an experience sampling procedure. Nonetheless, the feasibility and validity of EMA has been demonstrated in individuals with many types of addiction (Freedman et al., 2006, Johnson et al., 2009, Serre et al., 2012), and it is particularly well-suited to explore craving as a function of a wide range of environmental determinants. Repeated within-day assessments capture the rapid fluctuations of these variables (Shiyko and Ram, 2011) and inform researchers about prospective (and potentially causal) relationships.
Craving expression is also influenced by a large range of factors, including both stable and fluctuating moderators that characterize the individual or the environment. Diverse paradigms have been developed in laboratory settings to explore craving (Sinha and O’Malley, 1999), of which the most widely used involves exposure to cues that were previously paired with substance use in dependent individuals (e.g., drug paraphernalia, smell of alcohol; Carter and Tiffany, 1999, Childress et al., 1993, Franques et al., 1999). Craving has also been evoked experimentally through manipulation of mood, stress, or negative affect (Childress et al., 1994, Cooney et al., 1997, Litt et al., 1990, Sinha et al., 1999, Sinha et al., 2000) and, to a lesser extent, after “priming” involving acute exposure to the substance (de Wit, 1996, Mahoney et al., 2007). While some studies have found an association between craving or cue-reactivity assessed in the laboratory and relapse (al’Absi et al., 2005, Back et al., 2010, Fatseas et al., 2011, Sinha et al., 2006, Waters et al., 2004), other studies have failed to reveal such a link (Abrams et al., 1988, Rohsenow et al., 1994, Shadel et al., 1998). Moreover, a minority of participants failed to respond to cues or reported little craving in some studies (Avants et al., 1995, Childress et al., 1986). These discrepancies highlight the importance of individual variability in craving and relapse, as well as the necessity of interpreting findings as a function of these sources of variance. For example, men seem to experience lower craving than women in some studies (Elman et al., 2001, Robbins et al., 1999, Saladin et al., 2012, Yu et al., 2007), while individuals with higher impulsivity or novelty seeking traits tend to have higher craving levels (Franken, 2002, Papachristou et al., 2012, Powell et al., 1992, Zilberman et al., 2003). Craving ratings are also influenced by substance use characteristics such as levels of use, duration of abstinence, or dependence severity (Herd and Borland, 2009, Hughes, 1992, Watson et al., 2010), and by period of assessments relative to the target quit date (Hughes, 1992, McCarthy et al., 2006). Given the growing role of EMA in clinical research, but also important variation in findings, the objective of this systematic review was to: (1) assess the link between craving and substance use in all published studies using EMA; and (2) identify the relevant moderators of this relationship among substance users.
Section snippets
Methods
This review was based on a systematic examination of the literature based on the cochrane collaboration method (Chalmers and Altman, 1995, Denis et al., 2004) and Preferred Reporting items for systematic review and meta-analyses guidance for systematic reviews (Moher et al., 2009). Reviewers trained in systematic searches screened study references identified by the search strategy.
Search results (Fig. 1)
The online PubMed and PsycInfo search resulted in 117 publications. The hand search resulted in 36 additional articles. Based on titles and abstracts, 123 articles were considered of interest and obtained. Based on a review of full text versions of these articles, 91 met the inclusion criteria. These studies were published between 1996 and 2013.
Relevant studies are described separately and presented in Table 1 (craving moderators) and Table 2 (association between craving and substance use).
Discussion
The objective of this systematic review was to both assess the link between craving and substance use in EMA studies, and to identify relevant moderators of craving among substance users. A focus on EMA investigations is justified because this method of mobile assessment is most suitable to study high dynamic phenomena in their natural environment, and it permits clearer descriptions of the temporal associations of craving and substance use. Ninety-one studies were included in the present
Role of funding source
Funding for this study was provided by a Research Grant PHRC (2006–2014) from the French Ministry of Health, Research Grant AAP-Recherche-CRA (20091301018) from the Aquitaine Regional Council and French Government Addiction Agency MILDT grant 2010 to M. Auriacombe, French National Research Agency PRA-CNRS-CHU-Bordeaux award (2008–2010) to M. Fatseas, and from a CNRS ATIP award to J. Swendsen. The funding sponsors had no role in the design and conduct of the study and interpretation of the data,
Contributors
Marc Auriacombe was the overall principal investigator of the study and obtained funding. Melina Fatseas, Fuschia Serre, Joël Swendsen and Marc Auriacombe designed the review. Fuschia Serre managed the literature searches and summaries. Fuschia Serre and Melina Fatseas led the drafting of the manuscript. All authors undertook the critical revision of the manuscript for important intellectual content. All authors have reviewed and approved the current version of the manuscript.
Conflict of interest
No conflict declared.
Acknowledgments
We thank Hortense Hurmic, Charlotte Kervran, Cécile Trouba and Magali Alix for their help in screening and critical reading of studies selected in this review. Finally the authors are very thankful to Paolo, son of Fuschia Serre born on day of final acceptance of manuscript.
References (154)
- et al.
Mood disorders in everyday life: a systematic review of experience sampling and ecological momentary assessment studies
Clin. Psychol. Rev.
(2012) - et al.
Reactivity to smoking cues and relapse: two studies of discriminant validity
Behav. Res. Ther.
(1988) - et al.
Differences between responders and nonresponders to cocaine cues in the laboratory
Addict. Behav.
(1995) - et al.
Reactivity to laboratory stress provocation predicts relapse to cocaine
Drug Alcohol Depend.
(2010) - et al.
Distinct coping strategies differentially predict urge levels and lapses in a smoking cessation attempt
Addict. Behav.
(2013) - et al.
Immediate antecedents of marijuana use: an analysis from ecological momentary assessment
J. Behav. Ther. Exp. Psychiatry
(2012) - et al.
Social anxiety and cannabis use: an analysis from ecological momentary assessment
J. Anxiety Disord.
(2012) - et al.
Within-day temporal patterns of smoking, withdrawal symptoms, and craving
Drug Alcohol Depend.
(2011) - et al.
Can induced moods trigger drug-related responses in opiate abuse patients?
J. Subst. Abuse Treat.
(1994) - et al.
Conditioned responses in a methadone population. A comparison of laboratory, clinic, and natural settings
J. Subst. Abuse Treat.
(1986)