Development of the Caffeine Withdrawal Symptom Questionnaire: Caffeine withdrawal symptoms cluster into 7 factors☆
Introduction
Caffeine is the most widely used mood altering drug in the world. In the United States, 87% of children and adults regularly consume caffeine (Frary et al., 2005), with adult caffeine consumers ingesting approximately 280 mg per day on average (Barone and Roberts, 1996). The habitual use of caffeine produces physical dependence, which manifests as caffeine withdrawal symptoms upon acute abstinence. Caffeine withdrawal has been documented in animals and humans (Dingle et al., 2008, Juliano and Griffiths, 2004, Sigmon et al., 2009), with the biological basis believed to be increased functional sensitivity to endogenous adenosine via the upregulation of adenosine receptors (Ferre, 2008). Caffeine withdrawal symptoms typically emerge after 12–24 h of caffeine abstinence, peak in the first day or two, and last from 2 to 9 days (Juliano and Griffiths, 2004). Caffeine withdrawal can impair normal functioning (e.g., incapacitating headache, missing work; Juliano and Griffiths, 2004, Strain et al., 1994), and it has been suggested that fatigue and performance decrements resulting from acute caffeine abstinence may have important implications for safety (e.g., sleepiness while driving) and academic achievement (Heatherley et al., 2006, James and Keane, 2007, James et al., 2011).
In 1994 caffeine withdrawal was added to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) as a proposed diagnosis (Hughes, 1994). Tentative criteria were put forth consisting of headache plus one or more of the following symptoms: 1. Marked fatigue or drowsiness, 2. Marked anxiety or depression, and 3. Nausea or vomiting (American Psychiatric Association, 1994). In 2004 Juliano and Griffiths conducted a comprehensive review of caffeine withdrawal research and found highly convincing empirical support for the following symptoms: headache, fatigue, decreased energy/activeness, decreased alertness, drowsiness, decreased contentedness, depressed mood, difficulty concentrating, irritability, foggy/not clearheaded, flu-like symptoms, nausea/vomiting, and muscle pain/stiffness. Additional symptoms for which there was good evidence included decreased desire to socialize, unmotivated for work, decreased self-confidence, and heavy feelings in arms and legs (Juliano and Griffiths, 2004). Using a conceptual approach and findings from a prior cluster analysis (Evans and Griffiths, 1999), Juliano and Griffiths (2004) proposed a revision of caffeine withdrawal diagnostic criteria comprised of five symptom clusters. They suggested that future research with sufficient sample sizes use statistical techniques to examine the co-variation of caffeine withdrawal symptoms. Subsequently, Ozsungur et al. (2009) had caffeine consumers (N = 495) retrospectively rate the degree to which they experienced 13 symptoms validated in the Juliano and Griffiths (2004) review plus anxiety/nervousness (a symptom proposed in the DSM-IV-TR) “up to 48 h after ceasing to consume caffeine-containing beverages”. Principal components analysis (PCA) of the 14 items yielded three factors: fatigue/headache, dysphoric mood, and flu-like somatic symptoms, with difficulty concentrating loading on two of the factors. However, one important limitation of this study is its reliance on the retrospective recall, which can be subject to memory errors and systematic bias (Shiffman, 1994). Moreover, it is not clear if participants’ ratings were based on one or multiple incidents of caffeine abstinence, which is important to a meaningful analysis of clustering of symptoms during a discrete episode of caffeine withdrawal.
Prospective studies involving controlled periods of caffeine abstinence, and consistent measurement of caffeine withdrawal symptoms along with data reduction analytic strategies are needed to further our knowledge of caffeine withdrawal symptoms and their co-variation, and guide the development of DSM diagnostic criteria. In addition, a validated measure of caffeine withdrawal is needed to facilitate these objectives. Various measures have been used in prior research to assess caffeine withdrawal but have important limitations in terms of content validity and/or unknown or weak psychometric properties. For example, the Profile of Mood States (McNair et al., 1992) has been widely used in caffeine research but does not assess key features of caffeine withdrawal (e.g., headache, flu-like symptoms). In many prior studies, dozens of adjectives describing caffeine withdrawal have been individually analyzed thus increasing unreliability and the risk of Type 1 error.
To address these important gaps in research on caffeine withdrawal, we developed and administered a 23-item Caffeine Withdrawal Symptom Questionnaire (CWSQ)1 to regular caffeine consumers (N = 213) after 16 h of caffeine abstinence. PCA was conducted to reduce the 23-items to a smaller set of composite variables. To test the sensitivity of the CWSQ, ratings during caffeine abstinence were compared to ratings after double-blind caffeine administration and ratings during a typical day of caffeine consumption, but not immediately after consuming caffeine (i.e., baseline). Ratings obtained from daily and non-daily coffee consumers were also compared.
Section snippets
Participants
Participants were 213 adults (65% female, 81% Caucasian) with a mean age of 24 (SD 7.8, range 18–64 yrs) recruited from American University and the Washington, DC metropolitan area via flyers and web-based advertisements seeking coffee drinkers to participate in paid research. This sample size was chosen based on the recommendation of 10 participants per item ratio suggested for instrument analysis (DeVellis, 2003, Nunnally and Bernstein, 1994). To be eligible, participants had to consume
Principal components analysis (PCA)
Item means, standard deviations and inter-item correlations were computed and examined. Descriptive data were calculated for each item, including frequencies for each response level on the 5-point scale (0-not at all to 4-extremely). The items had inter-total correlations between .3 and .8, which are acceptable (Pett et al., 2003) indicating that items may cluster together in a subsequent PCA. Data from the 213 participants were entered into a PCA analysis. A combination of methods was used to
Discussion
A 23-item Caffeine Withdrawal Symptom Questionnaire (CWSQ) with high content validity was developed based on a rigorous review of caffeine withdrawal symptomatology (Juliano and Griffiths, 2004). Using PCA a 7-factor solution was derived that accounted for 68% of the total variance. The factors are described as follows: 1. Fatigue/drowsiness; 2. Low alertness/difficulty concentrating; 3. Mood disturbances; 4. Low sociability/motivation to work; 5. Nausea/upset stomach; 6. Flu-like feelings; and
Role of funding source
Funding for this study was provided by the College of Arts and Sciences at American University. The College had no further role in the study.
Contributors
Authors Juliano, Harrell, and Westerman contributed to the study design, subject recruitment, and running of the subjects. Authors Juliano and Huntley conducted the statistical analyses. All authors contributed to the writing of the manuscript.
Conflict of interest
No conflict declared.
Acknowledgements
We thank Sarah E. Moore and Peter G. Kardel for their assistance with this project.
References (39)
- et al.
Caffeine consumption
Food. Chem. Toxicol.
(1996) - et al.
Caffeine dependence in rats: effects of exposure duration and concentration
Physiol. Behav.
(2008) - et al.
Food sources and intakes of caffeine in the diets of persons in the United States
J. Am. Diet. Assoc.
(2005) - et al.
Endorsement of DSM-IV dependence criteria among caffeine users
Drug Alcohol Depend.
(1998) - et al.
Adolescent substance use, sleep, and academic achievement: evidence of harm due to caffeine
J. Adolesc.
(2011) - et al.
Caffeinated energy drinks—a growing problem
Drug Alcohol Depend.
(2009) - et al.
Conditions under which a bogus pipeline procedure enhances the validity of self-reported cigarette smoking: a meta-analytic review
J. Appl. Soc. Psychol.
(1993) Diagnostic and Statistical Manual of Mental Disorders
(1994)The scree test for the number of factors
Multivar. Behav. Res.
(1966)- et al.
Caffeine metabolism: disposition in liver disease and hepatic-function testing
Scale Development: Theory and Applications
Theories of drug craving, ancient and modern
Addiction
Caffeine withdrawal: a parametric analysis of caffeine dosing conditions
J. Pharmacol. Exp. Ther.
Evaluating the use of exploratory factor analysis in psychological research
Psychol. Methods
An update on the mechanisms of the psychostimulant effects of caffeine
J. Neurochem.
Low-dose caffeine physical dependence in humans
J. Pharmacol. Exp. Ther.
Caffeine expectancies influence the subjective and behavioral effects of caffeine
Psychopharmacology (Berl.)
Factor retention decisions in exploratory factor analysis: a tutorial on parallel analysis
ORM
Psychostimulant and other effects of caffeine in 9- to 11-year-old children
J. Child Psychiatry Psychol.
Cited by (0)
- ☆
The Caffeine Withdrawal Symptom Questionnaire appears as Supplementary Material and can be found by accessing the online version of this paper at http://dx.doi.org.