- Email: [email protected]
All cravings are not created equal. Correlates of menstrual versus non-cyclic chocolate craving
Appetite 57 (2011) 1–5
Contents lists available at ScienceDirect
Appetite journal homepage: www.elsevier.com/locate/appet
Research report
All cravings are not created equal. Correlates of menstrual versus non-cyclic chocolate craving§ Julia M. Hormes a,*, C. Alix Timko b a b
Department of Psychology, University of Pennsylvania, United States Department of Psychology, Towson University, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 June 2010 Received in revised form 23 February 2011 Accepted 20 March 2011 Available online 2 April 2011
Chocolate cravings occur regularly in about 45% of American women. Approximately half of these women link their craving temporally to the menstrual cycle, with a significant proportion of cravings reportedly peaking around ovulation or the onset of menstruation. This study aimed to elucidate the correlates of menstrual craving versus non-cyclic craving. Questionnaires assessing the relationships between craving, eating behaviors, attitudes towards weight and shape, and general pathology were completed by 97 women. Menstrual craving was reported by 28.9% (n = 28) and was associated with potentially maladaptive weight- and eating-related behaviors and attitudes, including higher body mass index, elevated dietary restraint, less flexible control over intake, and more guilt associated with the consumption of chocolate. Findings point to potential mechanisms involved in the etiology of menstrual cravings, such as the view of craving as a response to abstinence from high-calorie foods in an attempt to manage cyclically occurring weight fluctuations. ß 2011 Elsevier Ltd. All rights reserved.
Keywords: Craving Chocolate Menstrual Perimenstrual Premenstrual
A food craving is a desire ‘‘so strong that it will cause a person to go far out of his or her way to satisfy [it]’’ (Rozin, Levine, & Stoess, 1991) and leads to the subjective experience of a compulsive urge to eat. Food cravings are common, with up to 100% of women and 75% of men reporting food or drink cravings (Pelchat, 1997; Rodin, Mancuso, Granger, & Nelbach, 1991; Zellner, Garriga-Trillo, Rohm, Centeno, & Parker, 1999). They tend to be specific, and if a certain food is craved usually no other substance can satisfy the craving (Rozin et al., 1991). Chocolate is the most commonly and intensely craved food among North American women (Rozin et al., 1991), with 91% of American female undergraduates reporting ever having experienced a craving for chocolate (Osman & Sobal, 2006), and 45% reporting regular chocolate cravings (Zellner et al., 1999). Craving for chocolate has received special attention among researchers for three reasons. First, while chocolate is so commonly and intensely craved among young adults in North American countries (Weingarten & Elston, 1991), it does not stand out in other countries as a food that is craved more than any other (Osman & Sobal, 2006; Zellner et al., 1999). The term ‘‘craving’’ does not
§ The authors thank Alexia Butcher, Mariya Fishbayn, Lauren Hurwitz, and Alissa Kur for their help in conducting this study. Both authors declare no conflicts of interest associated with the submission of this manuscript. No competing financial interests exist. * Corresponding author. E-mail address: [email protected] (J.M. Hormes).
0195-6663/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2011.03.008
lexicalize in a majority of languages other than English (Hormes & Rozin, 2010), pointing to important cultural differences in the phenomenology of craving, which are not yet fully understood. Second, there are striking gender differences in the prevalence of chocolate craving in the United States (U.S.), with women consistently reporting significantly higher rates of craving than men (Osman & Sobal, 2006; Rozin et al., 1991; Zellner et al., 1999). Third, as has been established repeatedly via retrospective and longitudinal respondent self-report, in about half of female chocolate cravers in the U.S., chocolate craving is linked specifically to the menstrual cycle (Hormes & Rozin, 2009). In fact, the gender differences in chocolate craving prevalence in the U.S. appear to be due primarily to the added presence of menstrual cravings in women, with about half of cravings reported by women occurring at specific times of the menstrual cycle. A majority of menstrual cravings occur primarily perimenstrually, with a pronounced increase in craving beginning several days before the onset of and lasting several days into menstruation (Bruinsma & Taren, 1999; Hill & Heaton-Brown, 1994; Mercer & Holder, 1997; Rozin et al., 1991; Zellner, Garriga-Trillo, Centeno, & Wadsworth, 2004). In spite of a clear temporal association between chocolate craving and the menstrual cycle there appears to be no obvious causal link such that hormonal changes associated specifically with menstruation somehow directly elicit cravings. There is no evidence for significant correlations between levels of estrogen and progesterone and craving frequency and intensity throughout the cycle (Rodin et al., 1991), and exogenous administration of progesterone during the premenstrual period
2
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5
does not prevent or weaken perimenstrual chocolate cravings (Michener, Rozin, Freeman, & Gale, 1999). Furthermore, craving prevalence does not decrease with the absence of regular menstrual periods following menopause to the extent that would be expected if hormonal fluctuations were primary in bringing about cravings (Hormes & Rozin, 2009). There is also no evidence to suggest that chocolate contains pharmacologically active ingredients in large enough quantities to exert any noticeable effects that may be involved in perimenstrual or other cravings (Mumford et al., 1994; Pelchat & Schaeffer, 2000; Smit, Gaffan, & Rogers, 2004). In the absence of clearly identifiable physiological mechanisms, a number of alternative hypothesis have been formulated to account for the etiology of cyclically occurring chocolate and other cravings. These include the view of craving as a result of conflicting or ambivalent attitudes towards chocolate (Cartwright & Stritzke, 2008). However, the exact mechanisms involved in the etiology of these cravings remain to be determined (Hormes & Rozin, 2009; Michener et al., 1999; Rodin et al., 1991). The present study was designed to identify correlates of menstrual versus non-cyclic chocolate cravings as a way to gain a better understanding of potential mechanisms involved in the etiology of these cravings. In the absence of evidence to suggest a causal role of hormonal fluctuations or other physiological processes it was hypothesized that certain weight- and eatingrelated attitudes and behaviors may be involved in the emergence of cravings at the time of the perimenstrum. While chocolate craving and other food cravings are for the most part harmless (compared to cravings for cigarettes, alcohol, and drugs of abuse) they have been shown to lead to binge eating episodes in bulimic patients (Kales, 1990) and obese women (Bjoervell, Roennberg, & Roessner, 1985), and to increased food consumption in restrained eaters (Fedoroff, Polivy, & Herman, 1997). Even in non-pathological populations chocolate craving can elicit powerful feelings of ambivalence and guilt (Macdiarmid & Hetherington, 1995). A better understanding of the mechanisms underlying chocolate craving and the potential relationship between craving and problematic weight- and eating-related attitudes and behaviors thus has immediate clinical relevance.
Methods All methods were approved by the Institutional Review Board of the University of Pennsylvania. Participants Participants were 115 women recruited from a department subject pool who participated in the study in exchange for research participation credit. Of the 115 participants, 18 were excluded either because they did not report any chocolate craving (n = 6), or because they indicated body mass indices BMI above 25.0, putting them in the ‘‘overweight’’ and ‘‘obese’’ range (n = 12). The remaining 97 participants were included in the analyses reported here. Respondents were on average 20 years old (M = 19.57, SD = 2.97) and self-identified as Caucasian (50.5%, n = 49), Asian (35.1%, n = 34), Hispanic (5.0%, n = 5), African-American (1.0%, n = 1), and ‘‘other’’ (8.2%, n = 8). Even though only respondents with a BMIs in the normal range were included, participants on average indicated a seven pound discrepancy between their current weight and perceived ideal weight (M = 7.36, SD = 6.13). About one third of respondents (27.8%, n = 27) self-identified as being on a diet to lose weight at the time of the study, hoping to lose an average of 9.88 pounds (SD = 5.53).
Materials and procedures Participants completed the battery of questionnaires described below confidentially online via a secure server. A number of measures were included to paint a nuanced picture of respondents’ weight- and eating-related attitudes and behaviors, and to identify ways in which response patterns may differ in menstrual versus non-cyclic cravers. Specifically, the relationships between craving, depression, anxiety and stress, food cravings in general, attitudes to chocolate, dietary restraint, disordered eating, and other eatingrelated pathology were assessed. Measures Participants provided demographic information, including age and ethnicity, their weight and height (to determine BMI), ideal weight, self-identified dieting status (‘‘Are you currently on a diet to lose weight?’’), other dieting habits, and nature and patterns of chocolate and other food cravings (‘‘Do you have any cravings for food or drink? If you do crave something, what is your strongest craving for?’’). They were asked to indicate the frequency (on a scale of ‘‘0 = never crave’’ to ‘‘5 = crave chocolate more than once a day’’) and intensity (on a scale of ‘‘0 = no cravings’’ to ‘‘4 = extreme cravings’’) of their chocolate cravings. Participants were asked if they believed there was a link between their chocolate cravings and menstrual cycle. If respondents thought such a link existed they were asked to indicate the times at which cravings tend to peak by checking off boxes corresponding to 28 days of the menstrual cycle, ranging from day 1 ‘‘onset of menstruation’’ to day 28 ‘‘day before menstruation.’’ This manner of assessing patterns of craving across the menstrual cycle has been used in a previous study (Hormes & Rozin, 2009). Respondents then completed the following well-established, validated, and widely used measures: The Power of Food Scale (PFS) (Lowe et al., 2009): an 18-item measure of the susceptibility to the psychological influence of the mere presence or availability of food.1 It has been demonstrated previously that external food-related cues reliably trigger cravings (Sobik, Hutchison, & Craighead, 2005); it was hypothesized that the two groups of menstrual and non-cyclic cravers may differ in their sensitivity to these external cues. The Restraint Scale (RS) (Herman & Polivy, 1980): a ten-item measure assessing restrained eating that contains two subscales, measuring ‘‘Weight Fluctuation’’ and ‘‘Concern for Dieting.’’ Though dietary restraint has repeatedly been demonstrated to be unrelated to the occurrence of food cravings (Hill, Weaver, & Blundell, 1991; Rodin et al., 1991), potential differences in restraint between menstrual and non-cyclic cravers have not yet been assessed. The Attitudes to Chocolate Questionnaire (ACQ) (Benton, Greenfield, & Morgan, 1998): a 24-item, three-factor questionnaire assessing attitudes to chocolate. The first factor (‘‘Craving’’) measures craving for chocolate and the tendency to seek comfort from chocolate under emotionally stressful conditions, the second factor (‘‘Guilt’’) assesses negative feelings associated with eating chocolate and dissatisfaction with weight and body image, and the third factor (‘‘Functional’’) reflects a functional approach, such as using chocolate to gain energy when exercising or if missing a meal. The ACQ was included to test for potential differences in attitudes to and use of chocolate between menstrual and noncyclic cravers.
1 This study used an early version of the Power of Food scale that contains 18 items. It correlates highly with the current 15-item and alternate 21-item versions of the scale. (M.R. Lowe, personal communication, August 19, 2008).
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5
The Eating Attitude Test (EAT-26) (Garner & Garfinkel, 1979): a 26-item measure of symptoms and concerns characteristic of eating disorders. It contains three factors assessing ‘‘Dieting,’’ ‘‘Oral Control,’’ and ‘‘Bulimia and Food Preoccupation.’’ The EAT-26 was administered to gauge the extent to which menstrual and noncyclic cravers exhibit symptoms of eating disorders pathology. The Depression Anxiety Stress Scales-21 (DASS-21) (Lovibond & Lovibond, 1995): a 21-item self-report instrument measuring the three related negative emotional states of depression, anxiety, and tension/stress. This measure of general psychopathology was included based on existing reports of links between mood states and food cravings (Weingarten & Elston, 1990). The Eating Inventory (EI) with Rigid (RC) and Flexible Control (FC) (Stunkard & Messick, 1985; Westenhoefer, Stunkard, & Pudel, 1999): the original 51-item measure assessing ‘‘Cognitive Restraint of Eating,’’ ‘‘Disinhibiton,’’ and ‘‘Hunger’’ with an added 14 items to create the expanded versions of the RC And FC. Rigid and Flexible Control represent different aspects of dietary restraint (Westenhoefer et al., 1999) and the EI was administered to capture subtle aspects of the types of eating behaviors exhibited by menstrual versus non-cyclic cravers. The Food Craving Questionnaires Trait and State (FCQ-T, FCQS) (Cepeda-Benito, Gleaves, Williams, & Erath, 2000): two selfreport measures of the typical manifestation of craving in the individual (FCQT), and of craving as a psychological state in response to specific situations, such as stress (FCQ-S). Constructs assessed include FCQ-T factors 1: ‘‘Intentions and plans to consume food’’, 2: ‘‘Anticipation of positive reinforcement that may result from eating’’, 3: ‘‘Anticipation of relief from negative states and feelings as a result of eating,’’4: ‘‘Lack of control over eating,’’ 5: ‘‘Thoughts or preoccupation with food,’’ 6: ‘‘Craving as a physiological state,’’ 7: ‘‘Emotions that may be experienced before or during food cravings or eating,’’ 8: ‘‘Cues that may trigger food cravings,’’ and 9 ‘‘Guilt from cravings and/or giving in to them,’’ as well as FCQ-S factors 1: ‘‘Intense desire to eat,’’ 2: ‘‘Anticipation of positive reinforcement that may result from eating,’’ 3: ‘‘Anticipation of relief from negative states and feelings as a result of eating,’’ 4: ‘‘Lack of control over eating,’’ and 5: ‘‘Craving as a physiological state.’’ The FCQ-T and FCQ-S were administered to assess potential differences in the types (i.e., trait versus state) and characteristics of craving exhibited by menstrual versus non-cyclic cravers. Statistical analyses Menstrual craving was determined based on an individual’s answer (‘‘yes/no’’) to the question of whether cravings tend to peak at a specific time each month. In order to explore the impact of the presence of menstrual versus non-cyclic craving on measures of dietary restraint, disordered eating, depression, anxiety, and stress, general food cravings, attitudes to chocolate, and other eatingrelated pathology, a series of multivariate analysis of co-variance (MANCOVAs, for the FCQ-T, FCQ-S, ACQ, EAT-26, DASS-21 and EI, given that these measures contain multiple factors) and analyses of co-variance (ANCOVAs, for the remainder of the variables) was conducted. There were significant differences in body mass index between menstrual (M = 22.04, SD = 2.07) and non-cyclic cravers [M = 20.94, SD = 1.91; t(94) = 2.50, p = .01, Cohen’s d = .55], and BMI was initially included as a co-variate in all analyses. If found to be non-significant, it was subsequently excluded. Results Chocolate craving About one quarter of respondents thought that their chocolate cravings were linked to their menstrual cycle (28.9%, n = 28). Of
3
these, a majority (67.9%, n = 19) reported perimenstrual cravings (i.e. during the 4 days prior to the onset and 4 days lasting into the menstrual period); 32.1% (n = 9) indicated non-perimenstrual cravings. Most respondents had regular monthly periods (89.6%, n = 86), with no difference between menstrual and non-menstrual cravers in the occurrence of menstrual irregularities. Since prior research suggests that hormonal factors play little or no role in the etiology of menstrual craving (Hormes & Rozin, 2009; Michener et al., 1999; Rodin et al., 1991), participants were included regardless of regularity of their periods or factors such as use of hormonal birth control. Twenty-two respondents (22.7%) spontaneously (i.e., prior to any other questions about chocolate craving) indicated chocolate as their most craved food. Respondents on average reported craving ‘‘a few times a month’’ (M = 2.34, SD = 1.15 on a scale of 0 = ’’never crave’’ to 5 = ’’crave more than once a day’’), and described their cravings as ‘‘mild’’ to ‘‘moderate’’ in intensity (M = 1.97, SD = .91 on a scale of 0 = ’’no cravings’’ to 4 = ’’extreme cravings’’). There were no significant differences in craving frequency or intensity between menstrual and non-cyclic cravers (all p > .05). Phenomenology of menstrual versus non-cyclic craving Attitudes towards chocolate There was a significant multivariate main effect for craving cyclicity [F(3,92) = 3.62, p = .02, Wilk’s l = .89, hp2 = .11], and a significant (using a Bonferroni adjusted alpha level of .02) univariate effect for cyclicity on the ‘‘Guilt’’ factor of the ACQ [F(1,94) = 8.83, p = .004, hp2 = 09]. Menstrual cravers reported significantly more guilt than non-cyclic cravers (Table 1). There were no significant differences in scores on the ACQ ‘‘Craving’’ or ‘‘Functional’’ factors. Trait and state craving There were no multivariate main effects for craving cyclicity on trait (FCQ-T) [F(9,72) = .92, p = .51, Wilk’s l = 90, hp2 = .10] or state craving (FCQ-S) [F(5,90) = .12, p = .99, Wilk’s l = 99, hp2 = .01], suggesting no significant differences between menstrual and noncyclic cravers on the constructs assessed by the two instruments. Eating pathology Menstrual cravers reported significantly more weight dissatisfaction than non-cyclic cravers, as assessed using the difference Table 1 Means and standard deviations on dependent measures with significant differences between menstrual and non-cyclic chocolate cravers. Menstrual cravers (n = 28)
Non-cyclic cravers (n = 69)
M (SD)
M (SD)
Attitudes to chocolate questionnaire Factor 1: Craving 3.26 Factor 2: Guilt 3.36 Factor 3: Functional 2.77 Weight status Body mass index 22.03 Current minus ideal weight 9.64 Restraint scale Total score 17.36 Weight fluctuation 7.61 Concern for dieting 9.75 Eating inventory Disinhibition 20.32 Hunger 18.14 Cognitive restraint 22.54 Rigid control 19.29 Flexible control 13.57
(1.13) (1.14) (1.22)
2.89 (0.97) 2.69 (0.90) 2.38 (1.05)
(2.07) (7.05)
20.94 (1.91) 6.31 (5.44)
(5.62) (3.00) (3.81)
11.66 (5.40) 4.29 (2.81) 7.37 (3.46)
(2.14) (1.98) (2.78) (2.05) (1.87)
19.99 18.04 21.12 19.93 14.91
(2.38) (1.77) (3.07) (1.90) (1.69)
4
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5
between self-reported actual and ideal weights; however, this difference was no longer statistically significant when controlling for BMI [F(1,93) = .93, p = .34] (Table 1). There were no significant differences in total scores on the PFS, regardless of BMI, and no significant differences in eating disorder symptomatology, as measured by the EAT-26 (all p > .05). There was a statistically significant main effect of menstrual craving on scores on the Restraint Scale [F(1,93) = 15.78, p .001, hp2 = 15], with BMI as a significant co-variate [F(1,93) = 7.23, p = .01, hp2 = 07]. Menstrual cravers reported significantly higher levels of dietary restraint than non-cyclic cravers (Table 1). Looking at the two subscales of the Restraint Scales, significant differences emerged on both the Weight Fluctuation [F(1,93) = 19.81, p .001, hp2 = 18] and Concern for Dieting subscales [F(1,93) = 6.15, p = .02, hp2 = .06], with menstrual cravers scoring significantly higher on both, compared to non-cyclic cravers (Table 1). There was a significant main effect of craving cyclicity on scores on the EI [F(5,89) = 3.46, p = .01, Wilk’s l = 84, hp2 = .16], with BMI as a significant co-variate [F(5,89) = 2.90, p = .02, Wilk’s l = .86, hp2 = .14] and significant differences between menstrual and noncyclic cravers on the ‘‘Flexible Control’’ subscale [F(1,93) = 12.93, p = .001, hp2 = .12], such that menstrual cravers reported less flexible control over their intake than non-cyclic cravers (Table 1). General pathology There were no significant differences between menstrual and non-cyclic cravers on measures of overall Depression, Stress, or Anxiety (all p > .05), suggesting that differences between menstrual and non-menstrual cravers are specific to weight and eatingrelated measures. Discussion Findings presented here represent a first step in characterizing the weight- and eating-related correlates of menstrual versus noncyclic chocolate craving. Women who link their chocolate craving temporally to the menstrual cycle reported more maladaptive weight- and eating-related attitudes and behaviors, including higher BMI, elevated dietary restraint, less flexible control over their food intake, and more guilt associated with the consumption of chocolate. Non-cyclic cravers overall did not report attitudes or behaviors indicative of pathology. Based on current findings it is not possible to determine the exact mechanisms underlying menstrual craving or the direction of causality of the association between craving and the menstrual cycle. Findings do, however, point towards a role of certain attitudes and behaviors in the etiology of cyclically occurring cravings and can guide the formulation of hypotheses regarding the way in which these cravings may come about. It has previously been speculated that menstrual craving is the result of learned associations between the craving experience and times of the menstrual cycle characterized by unpleasant physical ‘‘symptoms,’’ such as mid-cycle and perimenstrual cramps, increased hunger, bloating, and weight gain (Altabe & Thompson, 1990; O’Herlihy, Robinson, & de Crespigny, 1980; Stubbs & Costos, 2004; Yonkers, O’Brien, & Eriksson, 2008). According to this hypothesis women use chocolate to cope with the emotional and physical symptoms of the perimenstrum and thus ‘‘learn’’ to crave it at that time (Rogers & Jas, 1994; Rogers & Smit, 2000; Zellner & Edwards, 2001; Zellner et al., 2004). While in some ways compelling as a theory, a learned associations view of menstrual craving does not explain why menstrual cravers would exhibit the negative eating- and weight-related attitudes they demonstrate in the present study. Furthermore, the fact that menstrual and noncyclic cravers did not differ their perception of the functionality of craving as reflected in similar scores on the ‘‘Functional’’ and
‘‘Craving’’ factors of the ACQ speaks against their use of chocolate to cope or gain comfort. With hunger and food intake increasing in the days following ovulation and preceding the onset of menstruation (Dalvit, 1981), women who are especially susceptible to weight gain shortly before menstruation may restrict their intake of high-fat, highcalorie foods (such as chocolate) in an effort to counteract anticipated weight gain. It has long been known that food restriction can lead to preoccupation with and craving for the avoided food (Channon & Hayward, 1990); women who restrict prior to menstruation may be setting themselves up to crave exactly those foods they are avoiding. Thus, an alternative hypothesis suggests that menstrual cravings are the result of unsuccessful attempts to manage cyclic weight fluctuations by restricting high-calorie foods. Findings presented here are consistent with this ‘‘restriction’’ hypothesis in that they point to more pronounced weight fluctuations, greater dietary restraint, and more guilt associated with eating chocolate in menstrual cravers. The relatively less flexible control over intake that characterizes menstrual cravers assessed in this study is also consistent with this account as it has previously been shown to be associated with greater susceptibility to overeating, greater self-reported energy intake, more guilt after consuming ‘‘forbidden’’ foods, and higher BMI (Westenhoefer et al., 1999). The possibility that menstrual craving is a response to restriction of specific foods at certain times of the cycle should be assessed further. There are some limitations inherent in the present study. Prior studies consistently found menstrual craving in approximately half of women interviewed; the present sample included a much lower number of self-identified menstrual cravers. This could be due to the fact that participants included here were all collegeaged women when menstrual craving may be more prevalent in older women. The exact reasons for these differences in prevalence of menstrual craving merit further attention. Furthermore, findings reported here are based exclusively on self-report of recalled past behaviors, provided via online questionnaires. Future studies might attempt to track chocolate craving and the correlates identified here longitudinally, for example via daily diaries. Problems with self-report via the internet have been documented in the past (Nosek, Banaji, & Greenwald, 2002); however, more recent evidence suggests that data collected via the internet is quite comparable to paper-and-pencil administration of the same measures (Hedman et al., 2010; Meyerson & Tryon, 2003; Zlomke, 2009). It can be concluded that chocolate craving that occurs cyclically at specific times of the menstrual cycle is associated with certain negative attitudes and behaviors related to weight and eating, an observation which may warrant attention, especially in clinical settings. Based on the evidence presented here it is recommended that future research on craving differentiate between menstrual and non-menstrual cravers. References Altabe, M., & Thompson, J. K. (1990). Menstrual cycle, body image, and eating disturbance. International Journal of Eating Disorders, 9(4), 395–401doi:10.1002/1098108X(199007)9:4. Benton, D., Greenfield, K., & Morgan, M. (1998). The development of the attitudes to chocolate questionnaire. Personality and Individual Differences, 24(4), 513– 520doi:10.1016/S0191-8869(97)00215-8. Bjoervell, H., Roennberg, S., & Roessner, S. (1985). Eating patterns described by a group of treatment seeking overweight women and normal weight women. Scandinavian Journal of Behavior Therapy, 14, 147–156. Bruinsma, K., & Taren, D. L. (1999). Chocolate. Food or drug? Journal of the American Dietetic Association, 99, 1249–1256doi:10.1016/S0002-8223(99)00307-7. Cartwright, F., & Stritzke, W. G. K. (2008). A multidimensional ambivalence model of chocolate craving. Construct validity and associations with chocolate consumption and disordered eating. Eating Behaviors, 9(1), 1–12.
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5 Cepeda-Benito, A., Gleaves, D. H., Williams, T. L., & Erath, S. A. (2000). The development and validation of the state and trait food-cravings questionnaires. Behavior Therapy, 31, 151–173doi:10.1016/S0005-7894(00)80009-X. Channon, S., & Hayward, A. (1990). The effect of short-term fasting on processing of food cues in normal subjects. International Journal of Eating Disorders, 9(4), 447–452. Dalvit, S. P. (1981). The effect of the menstrual cycle on patterns of food intake. American Journal of Clinical Nutrition, 34, 1811–1815. Fedoroff, I. C., Polivy, J., & Herman, C. P. (1997). The effect of preexposure to food cues on the eating behavior or restrained and unrestrained eaters. Appetite, 28, 33– 47doi:10.1006/appe.1996.0057. Garner, D. M., & Garfinkel, P. E. (1979). The eating attitudes test. An index of the symptoms of anorexia nervosa. Psychological Medicine, 9, 273–279. Hedman, E., Ljotsson, B., Ruck, C., Furmark, T., Carlbring, P., Lindefors, N., et al. (2010). Internet administration of self-report measures commonly used in research on social anxiety. A psychometric evaluation. Computers in Human Behavior, 26(4), 736–740. Herman, C. P., & Polivy, J. (1980). Restrained eating. In A. J. Stunkard (Ed.), Obesity. Philadelphia: Saunders. Hill, A. J., & Heaton-Brown, L. (1994). The experience of food craving. A prospective investigation in healthy women. Journal of Psychosomatic Research, 38, 801– 814doi:10.1016/0022-3999(94)90068-X. Hill, A. J., Weaver, C. F., & Blundell, J. E. (1991). Food craving, dietary restraint, and mood. Appetite, 17(3), 187–197. Hormes, J. M., & Rozin, P. (2009). Perimenstrual chocolate craving. What happens after menopause? Appetite, 53, 256–259doi:10.1016/j.appet.2009.07.003. Hormes, J. M., & Rozin, P. (2010). Does ‘‘craving’’ carve nature at the joints? Absence of a synonym for craving in most languages. Addictive Behaviors, 35, 459–463. Kales, E. F. (1990). Macronutrient analysis of binge eating in bulimia. Physiology & Behavior, 48, 837–840doi:10.1016/0031-9384(90)90236-W. Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the depression anxiety stress scales (2nd ed.). Sydney: Psychology Foundation. Lowe, M. R., Butryn, M. L., Didie, E. R., Annunziato, R. A., Thomas, G. J., & Crerand, C. E. (2009). The power of food scale. A new measure of the psychological influence of the food environment. Appetite, 53, 114–118doi:10.1016/j.appet.2009.05.016. Macdiarmid, J. I., & Hetherington, M. M. (1995). Mood modulation by food. An exploration of affect and cravings in ‘chocolate addicts’. British Journal of Clinical Psychology, 34, 129–138. Mercer, M. E., & Holder, M. D. (1997). Food cravings, endogenous opioid peptides, and food intake. A review. Appetite, 29, 325–352doi:10.1006/appe.1997.0100. Meyerson, P., & Tryon, W. W. (2003). Validating Internet research. A test of the psychometric equivalence of Internet and in-person samples. Behavior Research Methods, 35(4), 614–620. Michener, W., Rozin, P., Freeman, E., & Gale, L. (1999). The role of low progesterone and tension as triggers of perimenstrual chocolate and sweets craving. Some negative experimental evidence. Physiology & Behavior, 67(3), 417–420. Mumford, G. K., Evans, S. M., Kaminski, B. J., Preston, K. L., Sannerud, C. A., Silverman, K., et al. (1994). Discriminative stimulus and subjective effects of theobromine and caffeine in humans. Psychopharmacology, 115, 1–8doi:10.1007/BF02244744. Nosek, B. A., Banaji, M. R., & Greenwald, A. G. (2002). E-research. Ethics, security, design, and control in psychological research on the internet. Journal of Social Issues, 58(1), 161–176.
5
O’Herlihy, C., Robinson, H. P., & de Crespigny, L. J. (1980). Mittelschmerz is a preovulatory symptom. British Medical Journal, 280(6219), 986. Osman, J. L., & Sobal, J. (2006). Chocolate cravings in American and Spanish individuals. Biological and cultural influences. Appetite, 47, 290–301doi:10.1016/ j.appet.2006.04.008. Pelchat, M. L. (1997). Food cravings in young and elderly adults. Appetite, 28, 103– 113doi:10.1006/appe.1996.0063. Pelchat, M. L., & Schaeffer, S. (2000). Dietary monotony and food cravings in young and elderly adults. Physiology & Behavior, 68, 353–359doi:10.1016/S00319384(99)00190-0. Rodin, J., Mancuso, J., Granger, J., & Nelbach, E. (1991). Food cravings in relation to body mass index, restraint and estradiol levels. A repeated measures study in healthy women. Appetite, 17, 177–185doi:10.1016/0195-6663(91)90020-S. Rogers, P. J., & Jas, P. (1994). Menstrual cycle effects on mood, eating and food choice. Appetite, 23, 289. Rogers, P. J., & Smit, H. J. (2000). Food craving and food addiction. A critical review of the evidence from a biopsychosocial perspective. Pharmacology Biochemistry and Behavior, 66, 3–14. Rozin, P., Levine, E., & Stoess, C. (1991). Chocolate craving and liking. Appetite, 17, 199– 212doi:10.1016/0195-6663(91)90022-K. Smit, H. J., Gaffan, E. A., & Rogers, P. J. (2004). Methylxanthines are the pharmacologically active constituents of chocolate. Psychopharmacology, 176, 412– 419doi:10.1007/s00213-004-1898-3. Sobik, L., Hutchison, K., & Craighead, L. (2005). Cue-elicited craving for food. A fresh approach to the study of binge eating. Appetite, 44(3), 253–263. Stubbs, M. L., & Costos, D. (2004). Negative attitudes toward menstruation. Implications for disconnection within girls and between women. Women and Therapy, 27(3/4), 37–54doi:10.1300/J015v27n03_04. Stunkard, A. J., & Messick, S. (1985). The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. Journal of Psychosomatic Research, 29(1), 71–83doi:10.1016/0022-3999(85)90010-8. Weingarten, H., & Elston, D. (1990). The phenomenology of food cravings. Appetite, 15, 231–246. Weingarten, H., & Elston, D. (1991). Food cravings in a college population. Appetite, 17, 167–175. Westenhoefer, J., Stunkard, A. J., & Pudel, V. (1999). Validation of the flexible and rigid control dimensions of dietary restraint. International Journal of Eating Disorders, 26(1), 53–64. Yonkers, K. A., O’Brien, P. M. S., & Eriksson, E. (2008). Premenstrual syndrome. Lancet, 371, 1200–1210doi:10.1016/S0140-6736(08)60527-9. Zellner, D. A., & Edwards, E. M. (2001). Conditioning models of food craving. In M. M. Hetherington (Ed.), Food cravings and addiction (pp. 365–390). Surrey, UK: Leatherhead. Zellner, D. A., Garriga-Trillo, A., Centeno, S., & Wadsworth, E. (2004). Chocolate craving and the menstrual cycle. Appetite, 42, 119–121doi:10.1016/j.appet.2003.11.004. Zellner, D. A., Garriga-Trillo, A., Rohm, E., Centeno, S., & Parker, S. (1999). Food liking and craving. A cross-cultural approach. Appetite, 33, 61–70doi:10.1006/ appe.1999.0234. Zlomke, K. R. (2009). Psychometric properties of Internet administered versions of Penn State Worry Questionnaire (PSWQ) and Depression, Anxiety, and Stress Scale (DASS). Computers in Human Behavior, 25(4), 841–843.
Contents lists available at ScienceDirect
Appetite journal homepage: www.elsevier.com/locate/appet
Research report
All cravings are not created equal. Correlates of menstrual versus non-cyclic chocolate craving§ Julia M. Hormes a,*, C. Alix Timko b a b
Department of Psychology, University of Pennsylvania, United States Department of Psychology, Towson University, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 June 2010 Received in revised form 23 February 2011 Accepted 20 March 2011 Available online 2 April 2011
Chocolate cravings occur regularly in about 45% of American women. Approximately half of these women link their craving temporally to the menstrual cycle, with a significant proportion of cravings reportedly peaking around ovulation or the onset of menstruation. This study aimed to elucidate the correlates of menstrual craving versus non-cyclic craving. Questionnaires assessing the relationships between craving, eating behaviors, attitudes towards weight and shape, and general pathology were completed by 97 women. Menstrual craving was reported by 28.9% (n = 28) and was associated with potentially maladaptive weight- and eating-related behaviors and attitudes, including higher body mass index, elevated dietary restraint, less flexible control over intake, and more guilt associated with the consumption of chocolate. Findings point to potential mechanisms involved in the etiology of menstrual cravings, such as the view of craving as a response to abstinence from high-calorie foods in an attempt to manage cyclically occurring weight fluctuations. ß 2011 Elsevier Ltd. All rights reserved.
Keywords: Craving Chocolate Menstrual Perimenstrual Premenstrual
A food craving is a desire ‘‘so strong that it will cause a person to go far out of his or her way to satisfy [it]’’ (Rozin, Levine, & Stoess, 1991) and leads to the subjective experience of a compulsive urge to eat. Food cravings are common, with up to 100% of women and 75% of men reporting food or drink cravings (Pelchat, 1997; Rodin, Mancuso, Granger, & Nelbach, 1991; Zellner, Garriga-Trillo, Rohm, Centeno, & Parker, 1999). They tend to be specific, and if a certain food is craved usually no other substance can satisfy the craving (Rozin et al., 1991). Chocolate is the most commonly and intensely craved food among North American women (Rozin et al., 1991), with 91% of American female undergraduates reporting ever having experienced a craving for chocolate (Osman & Sobal, 2006), and 45% reporting regular chocolate cravings (Zellner et al., 1999). Craving for chocolate has received special attention among researchers for three reasons. First, while chocolate is so commonly and intensely craved among young adults in North American countries (Weingarten & Elston, 1991), it does not stand out in other countries as a food that is craved more than any other (Osman & Sobal, 2006; Zellner et al., 1999). The term ‘‘craving’’ does not
§ The authors thank Alexia Butcher, Mariya Fishbayn, Lauren Hurwitz, and Alissa Kur for their help in conducting this study. Both authors declare no conflicts of interest associated with the submission of this manuscript. No competing financial interests exist. * Corresponding author. E-mail address: [email protected] (J.M. Hormes).
0195-6663/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2011.03.008
lexicalize in a majority of languages other than English (Hormes & Rozin, 2010), pointing to important cultural differences in the phenomenology of craving, which are not yet fully understood. Second, there are striking gender differences in the prevalence of chocolate craving in the United States (U.S.), with women consistently reporting significantly higher rates of craving than men (Osman & Sobal, 2006; Rozin et al., 1991; Zellner et al., 1999). Third, as has been established repeatedly via retrospective and longitudinal respondent self-report, in about half of female chocolate cravers in the U.S., chocolate craving is linked specifically to the menstrual cycle (Hormes & Rozin, 2009). In fact, the gender differences in chocolate craving prevalence in the U.S. appear to be due primarily to the added presence of menstrual cravings in women, with about half of cravings reported by women occurring at specific times of the menstrual cycle. A majority of menstrual cravings occur primarily perimenstrually, with a pronounced increase in craving beginning several days before the onset of and lasting several days into menstruation (Bruinsma & Taren, 1999; Hill & Heaton-Brown, 1994; Mercer & Holder, 1997; Rozin et al., 1991; Zellner, Garriga-Trillo, Centeno, & Wadsworth, 2004). In spite of a clear temporal association between chocolate craving and the menstrual cycle there appears to be no obvious causal link such that hormonal changes associated specifically with menstruation somehow directly elicit cravings. There is no evidence for significant correlations between levels of estrogen and progesterone and craving frequency and intensity throughout the cycle (Rodin et al., 1991), and exogenous administration of progesterone during the premenstrual period
2
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5
does not prevent or weaken perimenstrual chocolate cravings (Michener, Rozin, Freeman, & Gale, 1999). Furthermore, craving prevalence does not decrease with the absence of regular menstrual periods following menopause to the extent that would be expected if hormonal fluctuations were primary in bringing about cravings (Hormes & Rozin, 2009). There is also no evidence to suggest that chocolate contains pharmacologically active ingredients in large enough quantities to exert any noticeable effects that may be involved in perimenstrual or other cravings (Mumford et al., 1994; Pelchat & Schaeffer, 2000; Smit, Gaffan, & Rogers, 2004). In the absence of clearly identifiable physiological mechanisms, a number of alternative hypothesis have been formulated to account for the etiology of cyclically occurring chocolate and other cravings. These include the view of craving as a result of conflicting or ambivalent attitudes towards chocolate (Cartwright & Stritzke, 2008). However, the exact mechanisms involved in the etiology of these cravings remain to be determined (Hormes & Rozin, 2009; Michener et al., 1999; Rodin et al., 1991). The present study was designed to identify correlates of menstrual versus non-cyclic chocolate cravings as a way to gain a better understanding of potential mechanisms involved in the etiology of these cravings. In the absence of evidence to suggest a causal role of hormonal fluctuations or other physiological processes it was hypothesized that certain weight- and eatingrelated attitudes and behaviors may be involved in the emergence of cravings at the time of the perimenstrum. While chocolate craving and other food cravings are for the most part harmless (compared to cravings for cigarettes, alcohol, and drugs of abuse) they have been shown to lead to binge eating episodes in bulimic patients (Kales, 1990) and obese women (Bjoervell, Roennberg, & Roessner, 1985), and to increased food consumption in restrained eaters (Fedoroff, Polivy, & Herman, 1997). Even in non-pathological populations chocolate craving can elicit powerful feelings of ambivalence and guilt (Macdiarmid & Hetherington, 1995). A better understanding of the mechanisms underlying chocolate craving and the potential relationship between craving and problematic weight- and eating-related attitudes and behaviors thus has immediate clinical relevance.
Methods All methods were approved by the Institutional Review Board of the University of Pennsylvania. Participants Participants were 115 women recruited from a department subject pool who participated in the study in exchange for research participation credit. Of the 115 participants, 18 were excluded either because they did not report any chocolate craving (n = 6), or because they indicated body mass indices BMI above 25.0, putting them in the ‘‘overweight’’ and ‘‘obese’’ range (n = 12). The remaining 97 participants were included in the analyses reported here. Respondents were on average 20 years old (M = 19.57, SD = 2.97) and self-identified as Caucasian (50.5%, n = 49), Asian (35.1%, n = 34), Hispanic (5.0%, n = 5), African-American (1.0%, n = 1), and ‘‘other’’ (8.2%, n = 8). Even though only respondents with a BMIs in the normal range were included, participants on average indicated a seven pound discrepancy between their current weight and perceived ideal weight (M = 7.36, SD = 6.13). About one third of respondents (27.8%, n = 27) self-identified as being on a diet to lose weight at the time of the study, hoping to lose an average of 9.88 pounds (SD = 5.53).
Materials and procedures Participants completed the battery of questionnaires described below confidentially online via a secure server. A number of measures were included to paint a nuanced picture of respondents’ weight- and eating-related attitudes and behaviors, and to identify ways in which response patterns may differ in menstrual versus non-cyclic cravers. Specifically, the relationships between craving, depression, anxiety and stress, food cravings in general, attitudes to chocolate, dietary restraint, disordered eating, and other eatingrelated pathology were assessed. Measures Participants provided demographic information, including age and ethnicity, their weight and height (to determine BMI), ideal weight, self-identified dieting status (‘‘Are you currently on a diet to lose weight?’’), other dieting habits, and nature and patterns of chocolate and other food cravings (‘‘Do you have any cravings for food or drink? If you do crave something, what is your strongest craving for?’’). They were asked to indicate the frequency (on a scale of ‘‘0 = never crave’’ to ‘‘5 = crave chocolate more than once a day’’) and intensity (on a scale of ‘‘0 = no cravings’’ to ‘‘4 = extreme cravings’’) of their chocolate cravings. Participants were asked if they believed there was a link between their chocolate cravings and menstrual cycle. If respondents thought such a link existed they were asked to indicate the times at which cravings tend to peak by checking off boxes corresponding to 28 days of the menstrual cycle, ranging from day 1 ‘‘onset of menstruation’’ to day 28 ‘‘day before menstruation.’’ This manner of assessing patterns of craving across the menstrual cycle has been used in a previous study (Hormes & Rozin, 2009). Respondents then completed the following well-established, validated, and widely used measures: The Power of Food Scale (PFS) (Lowe et al., 2009): an 18-item measure of the susceptibility to the psychological influence of the mere presence or availability of food.1 It has been demonstrated previously that external food-related cues reliably trigger cravings (Sobik, Hutchison, & Craighead, 2005); it was hypothesized that the two groups of menstrual and non-cyclic cravers may differ in their sensitivity to these external cues. The Restraint Scale (RS) (Herman & Polivy, 1980): a ten-item measure assessing restrained eating that contains two subscales, measuring ‘‘Weight Fluctuation’’ and ‘‘Concern for Dieting.’’ Though dietary restraint has repeatedly been demonstrated to be unrelated to the occurrence of food cravings (Hill, Weaver, & Blundell, 1991; Rodin et al., 1991), potential differences in restraint between menstrual and non-cyclic cravers have not yet been assessed. The Attitudes to Chocolate Questionnaire (ACQ) (Benton, Greenfield, & Morgan, 1998): a 24-item, three-factor questionnaire assessing attitudes to chocolate. The first factor (‘‘Craving’’) measures craving for chocolate and the tendency to seek comfort from chocolate under emotionally stressful conditions, the second factor (‘‘Guilt’’) assesses negative feelings associated with eating chocolate and dissatisfaction with weight and body image, and the third factor (‘‘Functional’’) reflects a functional approach, such as using chocolate to gain energy when exercising or if missing a meal. The ACQ was included to test for potential differences in attitudes to and use of chocolate between menstrual and noncyclic cravers.
1 This study used an early version of the Power of Food scale that contains 18 items. It correlates highly with the current 15-item and alternate 21-item versions of the scale. (M.R. Lowe, personal communication, August 19, 2008).
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5
The Eating Attitude Test (EAT-26) (Garner & Garfinkel, 1979): a 26-item measure of symptoms and concerns characteristic of eating disorders. It contains three factors assessing ‘‘Dieting,’’ ‘‘Oral Control,’’ and ‘‘Bulimia and Food Preoccupation.’’ The EAT-26 was administered to gauge the extent to which menstrual and noncyclic cravers exhibit symptoms of eating disorders pathology. The Depression Anxiety Stress Scales-21 (DASS-21) (Lovibond & Lovibond, 1995): a 21-item self-report instrument measuring the three related negative emotional states of depression, anxiety, and tension/stress. This measure of general psychopathology was included based on existing reports of links between mood states and food cravings (Weingarten & Elston, 1990). The Eating Inventory (EI) with Rigid (RC) and Flexible Control (FC) (Stunkard & Messick, 1985; Westenhoefer, Stunkard, & Pudel, 1999): the original 51-item measure assessing ‘‘Cognitive Restraint of Eating,’’ ‘‘Disinhibiton,’’ and ‘‘Hunger’’ with an added 14 items to create the expanded versions of the RC And FC. Rigid and Flexible Control represent different aspects of dietary restraint (Westenhoefer et al., 1999) and the EI was administered to capture subtle aspects of the types of eating behaviors exhibited by menstrual versus non-cyclic cravers. The Food Craving Questionnaires Trait and State (FCQ-T, FCQS) (Cepeda-Benito, Gleaves, Williams, & Erath, 2000): two selfreport measures of the typical manifestation of craving in the individual (FCQT), and of craving as a psychological state in response to specific situations, such as stress (FCQ-S). Constructs assessed include FCQ-T factors 1: ‘‘Intentions and plans to consume food’’, 2: ‘‘Anticipation of positive reinforcement that may result from eating’’, 3: ‘‘Anticipation of relief from negative states and feelings as a result of eating,’’4: ‘‘Lack of control over eating,’’ 5: ‘‘Thoughts or preoccupation with food,’’ 6: ‘‘Craving as a physiological state,’’ 7: ‘‘Emotions that may be experienced before or during food cravings or eating,’’ 8: ‘‘Cues that may trigger food cravings,’’ and 9 ‘‘Guilt from cravings and/or giving in to them,’’ as well as FCQ-S factors 1: ‘‘Intense desire to eat,’’ 2: ‘‘Anticipation of positive reinforcement that may result from eating,’’ 3: ‘‘Anticipation of relief from negative states and feelings as a result of eating,’’ 4: ‘‘Lack of control over eating,’’ and 5: ‘‘Craving as a physiological state.’’ The FCQ-T and FCQ-S were administered to assess potential differences in the types (i.e., trait versus state) and characteristics of craving exhibited by menstrual versus non-cyclic cravers. Statistical analyses Menstrual craving was determined based on an individual’s answer (‘‘yes/no’’) to the question of whether cravings tend to peak at a specific time each month. In order to explore the impact of the presence of menstrual versus non-cyclic craving on measures of dietary restraint, disordered eating, depression, anxiety, and stress, general food cravings, attitudes to chocolate, and other eatingrelated pathology, a series of multivariate analysis of co-variance (MANCOVAs, for the FCQ-T, FCQ-S, ACQ, EAT-26, DASS-21 and EI, given that these measures contain multiple factors) and analyses of co-variance (ANCOVAs, for the remainder of the variables) was conducted. There were significant differences in body mass index between menstrual (M = 22.04, SD = 2.07) and non-cyclic cravers [M = 20.94, SD = 1.91; t(94) = 2.50, p = .01, Cohen’s d = .55], and BMI was initially included as a co-variate in all analyses. If found to be non-significant, it was subsequently excluded. Results Chocolate craving About one quarter of respondents thought that their chocolate cravings were linked to their menstrual cycle (28.9%, n = 28). Of
3
these, a majority (67.9%, n = 19) reported perimenstrual cravings (i.e. during the 4 days prior to the onset and 4 days lasting into the menstrual period); 32.1% (n = 9) indicated non-perimenstrual cravings. Most respondents had regular monthly periods (89.6%, n = 86), with no difference between menstrual and non-menstrual cravers in the occurrence of menstrual irregularities. Since prior research suggests that hormonal factors play little or no role in the etiology of menstrual craving (Hormes & Rozin, 2009; Michener et al., 1999; Rodin et al., 1991), participants were included regardless of regularity of their periods or factors such as use of hormonal birth control. Twenty-two respondents (22.7%) spontaneously (i.e., prior to any other questions about chocolate craving) indicated chocolate as their most craved food. Respondents on average reported craving ‘‘a few times a month’’ (M = 2.34, SD = 1.15 on a scale of 0 = ’’never crave’’ to 5 = ’’crave more than once a day’’), and described their cravings as ‘‘mild’’ to ‘‘moderate’’ in intensity (M = 1.97, SD = .91 on a scale of 0 = ’’no cravings’’ to 4 = ’’extreme cravings’’). There were no significant differences in craving frequency or intensity between menstrual and non-cyclic cravers (all p > .05). Phenomenology of menstrual versus non-cyclic craving Attitudes towards chocolate There was a significant multivariate main effect for craving cyclicity [F(3,92) = 3.62, p = .02, Wilk’s l = .89, hp2 = .11], and a significant (using a Bonferroni adjusted alpha level of .02) univariate effect for cyclicity on the ‘‘Guilt’’ factor of the ACQ [F(1,94) = 8.83, p = .004, hp2 = 09]. Menstrual cravers reported significantly more guilt than non-cyclic cravers (Table 1). There were no significant differences in scores on the ACQ ‘‘Craving’’ or ‘‘Functional’’ factors. Trait and state craving There were no multivariate main effects for craving cyclicity on trait (FCQ-T) [F(9,72) = .92, p = .51, Wilk’s l = 90, hp2 = .10] or state craving (FCQ-S) [F(5,90) = .12, p = .99, Wilk’s l = 99, hp2 = .01], suggesting no significant differences between menstrual and noncyclic cravers on the constructs assessed by the two instruments. Eating pathology Menstrual cravers reported significantly more weight dissatisfaction than non-cyclic cravers, as assessed using the difference Table 1 Means and standard deviations on dependent measures with significant differences between menstrual and non-cyclic chocolate cravers. Menstrual cravers (n = 28)
Non-cyclic cravers (n = 69)
M (SD)
M (SD)
Attitudes to chocolate questionnaire Factor 1: Craving 3.26 Factor 2: Guilt 3.36 Factor 3: Functional 2.77 Weight status Body mass index 22.03 Current minus ideal weight 9.64 Restraint scale Total score 17.36 Weight fluctuation 7.61 Concern for dieting 9.75 Eating inventory Disinhibition 20.32 Hunger 18.14 Cognitive restraint 22.54 Rigid control 19.29 Flexible control 13.57
(1.13) (1.14) (1.22)
2.89 (0.97) 2.69 (0.90) 2.38 (1.05)
(2.07) (7.05)
20.94 (1.91) 6.31 (5.44)
(5.62) (3.00) (3.81)
11.66 (5.40) 4.29 (2.81) 7.37 (3.46)
(2.14) (1.98) (2.78) (2.05) (1.87)
19.99 18.04 21.12 19.93 14.91
(2.38) (1.77) (3.07) (1.90) (1.69)
4
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5
between self-reported actual and ideal weights; however, this difference was no longer statistically significant when controlling for BMI [F(1,93) = .93, p = .34] (Table 1). There were no significant differences in total scores on the PFS, regardless of BMI, and no significant differences in eating disorder symptomatology, as measured by the EAT-26 (all p > .05). There was a statistically significant main effect of menstrual craving on scores on the Restraint Scale [F(1,93) = 15.78, p .001, hp2 = 15], with BMI as a significant co-variate [F(1,93) = 7.23, p = .01, hp2 = 07]. Menstrual cravers reported significantly higher levels of dietary restraint than non-cyclic cravers (Table 1). Looking at the two subscales of the Restraint Scales, significant differences emerged on both the Weight Fluctuation [F(1,93) = 19.81, p .001, hp2 = 18] and Concern for Dieting subscales [F(1,93) = 6.15, p = .02, hp2 = .06], with menstrual cravers scoring significantly higher on both, compared to non-cyclic cravers (Table 1). There was a significant main effect of craving cyclicity on scores on the EI [F(5,89) = 3.46, p = .01, Wilk’s l = 84, hp2 = .16], with BMI as a significant co-variate [F(5,89) = 2.90, p = .02, Wilk’s l = .86, hp2 = .14] and significant differences between menstrual and noncyclic cravers on the ‘‘Flexible Control’’ subscale [F(1,93) = 12.93, p = .001, hp2 = .12], such that menstrual cravers reported less flexible control over their intake than non-cyclic cravers (Table 1). General pathology There were no significant differences between menstrual and non-cyclic cravers on measures of overall Depression, Stress, or Anxiety (all p > .05), suggesting that differences between menstrual and non-menstrual cravers are specific to weight and eatingrelated measures. Discussion Findings presented here represent a first step in characterizing the weight- and eating-related correlates of menstrual versus noncyclic chocolate craving. Women who link their chocolate craving temporally to the menstrual cycle reported more maladaptive weight- and eating-related attitudes and behaviors, including higher BMI, elevated dietary restraint, less flexible control over their food intake, and more guilt associated with the consumption of chocolate. Non-cyclic cravers overall did not report attitudes or behaviors indicative of pathology. Based on current findings it is not possible to determine the exact mechanisms underlying menstrual craving or the direction of causality of the association between craving and the menstrual cycle. Findings do, however, point towards a role of certain attitudes and behaviors in the etiology of cyclically occurring cravings and can guide the formulation of hypotheses regarding the way in which these cravings may come about. It has previously been speculated that menstrual craving is the result of learned associations between the craving experience and times of the menstrual cycle characterized by unpleasant physical ‘‘symptoms,’’ such as mid-cycle and perimenstrual cramps, increased hunger, bloating, and weight gain (Altabe & Thompson, 1990; O’Herlihy, Robinson, & de Crespigny, 1980; Stubbs & Costos, 2004; Yonkers, O’Brien, & Eriksson, 2008). According to this hypothesis women use chocolate to cope with the emotional and physical symptoms of the perimenstrum and thus ‘‘learn’’ to crave it at that time (Rogers & Jas, 1994; Rogers & Smit, 2000; Zellner & Edwards, 2001; Zellner et al., 2004). While in some ways compelling as a theory, a learned associations view of menstrual craving does not explain why menstrual cravers would exhibit the negative eating- and weight-related attitudes they demonstrate in the present study. Furthermore, the fact that menstrual and noncyclic cravers did not differ their perception of the functionality of craving as reflected in similar scores on the ‘‘Functional’’ and
‘‘Craving’’ factors of the ACQ speaks against their use of chocolate to cope or gain comfort. With hunger and food intake increasing in the days following ovulation and preceding the onset of menstruation (Dalvit, 1981), women who are especially susceptible to weight gain shortly before menstruation may restrict their intake of high-fat, highcalorie foods (such as chocolate) in an effort to counteract anticipated weight gain. It has long been known that food restriction can lead to preoccupation with and craving for the avoided food (Channon & Hayward, 1990); women who restrict prior to menstruation may be setting themselves up to crave exactly those foods they are avoiding. Thus, an alternative hypothesis suggests that menstrual cravings are the result of unsuccessful attempts to manage cyclic weight fluctuations by restricting high-calorie foods. Findings presented here are consistent with this ‘‘restriction’’ hypothesis in that they point to more pronounced weight fluctuations, greater dietary restraint, and more guilt associated with eating chocolate in menstrual cravers. The relatively less flexible control over intake that characterizes menstrual cravers assessed in this study is also consistent with this account as it has previously been shown to be associated with greater susceptibility to overeating, greater self-reported energy intake, more guilt after consuming ‘‘forbidden’’ foods, and higher BMI (Westenhoefer et al., 1999). The possibility that menstrual craving is a response to restriction of specific foods at certain times of the cycle should be assessed further. There are some limitations inherent in the present study. Prior studies consistently found menstrual craving in approximately half of women interviewed; the present sample included a much lower number of self-identified menstrual cravers. This could be due to the fact that participants included here were all collegeaged women when menstrual craving may be more prevalent in older women. The exact reasons for these differences in prevalence of menstrual craving merit further attention. Furthermore, findings reported here are based exclusively on self-report of recalled past behaviors, provided via online questionnaires. Future studies might attempt to track chocolate craving and the correlates identified here longitudinally, for example via daily diaries. Problems with self-report via the internet have been documented in the past (Nosek, Banaji, & Greenwald, 2002); however, more recent evidence suggests that data collected via the internet is quite comparable to paper-and-pencil administration of the same measures (Hedman et al., 2010; Meyerson & Tryon, 2003; Zlomke, 2009). It can be concluded that chocolate craving that occurs cyclically at specific times of the menstrual cycle is associated with certain negative attitudes and behaviors related to weight and eating, an observation which may warrant attention, especially in clinical settings. Based on the evidence presented here it is recommended that future research on craving differentiate between menstrual and non-menstrual cravers. References Altabe, M., & Thompson, J. K. (1990). Menstrual cycle, body image, and eating disturbance. International Journal of Eating Disorders, 9(4), 395–401doi:10.1002/1098108X(199007)9:4. Benton, D., Greenfield, K., & Morgan, M. (1998). The development of the attitudes to chocolate questionnaire. Personality and Individual Differences, 24(4), 513– 520doi:10.1016/S0191-8869(97)00215-8. Bjoervell, H., Roennberg, S., & Roessner, S. (1985). Eating patterns described by a group of treatment seeking overweight women and normal weight women. Scandinavian Journal of Behavior Therapy, 14, 147–156. Bruinsma, K., & Taren, D. L. (1999). Chocolate. Food or drug? Journal of the American Dietetic Association, 99, 1249–1256doi:10.1016/S0002-8223(99)00307-7. Cartwright, F., & Stritzke, W. G. K. (2008). A multidimensional ambivalence model of chocolate craving. Construct validity and associations with chocolate consumption and disordered eating. Eating Behaviors, 9(1), 1–12.
J.M. Hormes, C.A. Timko / Appetite 57 (2011) 1–5 Cepeda-Benito, A., Gleaves, D. H., Williams, T. L., & Erath, S. A. (2000). The development and validation of the state and trait food-cravings questionnaires. Behavior Therapy, 31, 151–173doi:10.1016/S0005-7894(00)80009-X. Channon, S., & Hayward, A. (1990). The effect of short-term fasting on processing of food cues in normal subjects. International Journal of Eating Disorders, 9(4), 447–452. Dalvit, S. P. (1981). The effect of the menstrual cycle on patterns of food intake. American Journal of Clinical Nutrition, 34, 1811–1815. Fedoroff, I. C., Polivy, J., & Herman, C. P. (1997). The effect of preexposure to food cues on the eating behavior or restrained and unrestrained eaters. Appetite, 28, 33– 47doi:10.1006/appe.1996.0057. Garner, D. M., & Garfinkel, P. E. (1979). The eating attitudes test. An index of the symptoms of anorexia nervosa. Psychological Medicine, 9, 273–279. Hedman, E., Ljotsson, B., Ruck, C., Furmark, T., Carlbring, P., Lindefors, N., et al. (2010). Internet administration of self-report measures commonly used in research on social anxiety. A psychometric evaluation. Computers in Human Behavior, 26(4), 736–740. Herman, C. P., & Polivy, J. (1980). Restrained eating. In A. J. Stunkard (Ed.), Obesity. Philadelphia: Saunders. Hill, A. J., & Heaton-Brown, L. (1994). The experience of food craving. A prospective investigation in healthy women. Journal of Psychosomatic Research, 38, 801– 814doi:10.1016/0022-3999(94)90068-X. Hill, A. J., Weaver, C. F., & Blundell, J. E. (1991). Food craving, dietary restraint, and mood. Appetite, 17(3), 187–197. Hormes, J. M., & Rozin, P. (2009). Perimenstrual chocolate craving. What happens after menopause? Appetite, 53, 256–259doi:10.1016/j.appet.2009.07.003. Hormes, J. M., & Rozin, P. (2010). Does ‘‘craving’’ carve nature at the joints? Absence of a synonym for craving in most languages. Addictive Behaviors, 35, 459–463. Kales, E. F. (1990). Macronutrient analysis of binge eating in bulimia. Physiology & Behavior, 48, 837–840doi:10.1016/0031-9384(90)90236-W. Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the depression anxiety stress scales (2nd ed.). Sydney: Psychology Foundation. Lowe, M. R., Butryn, M. L., Didie, E. R., Annunziato, R. A., Thomas, G. J., & Crerand, C. E. (2009). The power of food scale. A new measure of the psychological influence of the food environment. Appetite, 53, 114–118doi:10.1016/j.appet.2009.05.016. Macdiarmid, J. I., & Hetherington, M. M. (1995). Mood modulation by food. An exploration of affect and cravings in ‘chocolate addicts’. British Journal of Clinical Psychology, 34, 129–138. Mercer, M. E., & Holder, M. D. (1997). Food cravings, endogenous opioid peptides, and food intake. A review. Appetite, 29, 325–352doi:10.1006/appe.1997.0100. Meyerson, P., & Tryon, W. W. (2003). Validating Internet research. A test of the psychometric equivalence of Internet and in-person samples. Behavior Research Methods, 35(4), 614–620. Michener, W., Rozin, P., Freeman, E., & Gale, L. (1999). The role of low progesterone and tension as triggers of perimenstrual chocolate and sweets craving. Some negative experimental evidence. Physiology & Behavior, 67(3), 417–420. Mumford, G. K., Evans, S. M., Kaminski, B. J., Preston, K. L., Sannerud, C. A., Silverman, K., et al. (1994). Discriminative stimulus and subjective effects of theobromine and caffeine in humans. Psychopharmacology, 115, 1–8doi:10.1007/BF02244744. Nosek, B. A., Banaji, M. R., & Greenwald, A. G. (2002). E-research. Ethics, security, design, and control in psychological research on the internet. Journal of Social Issues, 58(1), 161–176.
5
O’Herlihy, C., Robinson, H. P., & de Crespigny, L. J. (1980). Mittelschmerz is a preovulatory symptom. British Medical Journal, 280(6219), 986. Osman, J. L., & Sobal, J. (2006). Chocolate cravings in American and Spanish individuals. Biological and cultural influences. Appetite, 47, 290–301doi:10.1016/ j.appet.2006.04.008. Pelchat, M. L. (1997). Food cravings in young and elderly adults. Appetite, 28, 103– 113doi:10.1006/appe.1996.0063. Pelchat, M. L., & Schaeffer, S. (2000). Dietary monotony and food cravings in young and elderly adults. Physiology & Behavior, 68, 353–359doi:10.1016/S00319384(99)00190-0. Rodin, J., Mancuso, J., Granger, J., & Nelbach, E. (1991). Food cravings in relation to body mass index, restraint and estradiol levels. A repeated measures study in healthy women. Appetite, 17, 177–185doi:10.1016/0195-6663(91)90020-S. Rogers, P. J., & Jas, P. (1994). Menstrual cycle effects on mood, eating and food choice. Appetite, 23, 289. Rogers, P. J., & Smit, H. J. (2000). Food craving and food addiction. A critical review of the evidence from a biopsychosocial perspective. Pharmacology Biochemistry and Behavior, 66, 3–14. Rozin, P., Levine, E., & Stoess, C. (1991). Chocolate craving and liking. Appetite, 17, 199– 212doi:10.1016/0195-6663(91)90022-K. Smit, H. J., Gaffan, E. A., & Rogers, P. J. (2004). Methylxanthines are the pharmacologically active constituents of chocolate. Psychopharmacology, 176, 412– 419doi:10.1007/s00213-004-1898-3. Sobik, L., Hutchison, K., & Craighead, L. (2005). Cue-elicited craving for food. A fresh approach to the study of binge eating. Appetite, 44(3), 253–263. Stubbs, M. L., & Costos, D. (2004). Negative attitudes toward menstruation. Implications for disconnection within girls and between women. Women and Therapy, 27(3/4), 37–54doi:10.1300/J015v27n03_04. Stunkard, A. J., & Messick, S. (1985). The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. Journal of Psychosomatic Research, 29(1), 71–83doi:10.1016/0022-3999(85)90010-8. Weingarten, H., & Elston, D. (1990). The phenomenology of food cravings. Appetite, 15, 231–246. Weingarten, H., & Elston, D. (1991). Food cravings in a college population. Appetite, 17, 167–175. Westenhoefer, J., Stunkard, A. J., & Pudel, V. (1999). Validation of the flexible and rigid control dimensions of dietary restraint. International Journal of Eating Disorders, 26(1), 53–64. Yonkers, K. A., O’Brien, P. M. S., & Eriksson, E. (2008). Premenstrual syndrome. Lancet, 371, 1200–1210doi:10.1016/S0140-6736(08)60527-9. Zellner, D. A., & Edwards, E. M. (2001). Conditioning models of food craving. In M. M. Hetherington (Ed.), Food cravings and addiction (pp. 365–390). Surrey, UK: Leatherhead. Zellner, D. A., Garriga-Trillo, A., Centeno, S., & Wadsworth, E. (2004). Chocolate craving and the menstrual cycle. Appetite, 42, 119–121doi:10.1016/j.appet.2003.11.004. Zellner, D. A., Garriga-Trillo, A., Rohm, E., Centeno, S., & Parker, S. (1999). Food liking and craving. A cross-cultural approach. Appetite, 33, 61–70doi:10.1006/ appe.1999.0234. Zlomke, K. R. (2009). Psychometric properties of Internet administered versions of Penn State Worry Questionnaire (PSWQ) and Depression, Anxiety, and Stress Scale (DASS). Computers in Human Behavior, 25(4), 841–843.