- Email: [email protected]
Impulsivity and test meal intake in obese binge eating women
Appetite 43 (2004) 303–307 www.elsevier.com/locate/appet
Impulsivity and test meal intake in obese binge eating women Jennifer A. Nassera,*, Marci E. Glucka, Allan Geliebtera,b a
Department of Medicine, NY Obesity Research Center, St Luke’s/Roosevelt Hospital Center, Columbia University College of Physicians and Surgeons, WH-1020, 1111 Amsterdam Ave, New York, NY 10025, USA b Department of Psychiatry, NY Obesity Research Center, St. Luke’s/Roosevelt Hospital Center, Columbia University College of Physicians and Surgeons, New York, NY 10025, USA Received 5 March 2004; revised 13 April 2004; accepted 15 April 2004
Abstract Greater impulsivity has been observed in those with chemical (cocaine, marijuana, alcohol) and behavioral addictions (gambling, sex, shopping), as well as in individuals with personality and conduct disorders. Greater impulsivity has also been described in those with Bulimia Nervosa and attributed to aberrations in serotonin, as has eating in response to negative affect.However, less is known about the impact of impulsivity on eating behavior in obese humans in general, and in those who meet sub-clinical and full clinical criteria for Binge Eating Disorder (BED) in particular. Using a laboratory test meal paradigm, we demonstrated: (1) greater Motor Impulsivity (Barratt Impulsivity Scale (BIS) ðp ¼ 0:05Þ in those with BED ðn ¼ 11Þ as compared to those without BED ðn ¼ 11Þ; (2) a positive correlation between BED criteria and BIS scores ðp , 0:01Þ; (3) a positive correlation between test meal duration and Zung Depression Score, and (4) a positive correlation between Motor Impulsivity and mood rated before consuming the test meal. These associations suggest potential aberrations in serotonin transmission in BED, and a possible target for pharmacotherapy of BED especially in those who are resistant to Cognitive Behavioral Therapy. q 2004 Elsevier Ltd. All rights reserved. Keywords: Impulsivity; Binge eating; Obesity; Test meal intake
Introduction Current literature define impulsivity as (1) the inclination to choose small, immediately available rewards over larger, delayed rewards, and/or (2) the inclination to respond rapidly without forethought and/or attention to consequences (Evenden, 1999; Swann, Bjork, Moeller, & Dougherty, 2002). Greater impulsivity has been reported in those with chemical (cocaine, alcohol, nicotine) (Allen, Moeller, Rhoades, & Cherek, 1998; Mitchell, 1999; Richards, Zhang, Mitchell, & de Wit, 1999) and behavioral addictions (gambling, sex, shopping) as well as in those with personality and conduct disorders (Hollander & Evers, 2002), where it correlates with severity of dependence or disordered state. Some investigators, using treatment program patients, report greater impulsivity in individuals with eating disorders compared to normal controls (Fahy & Eisler, 1993; Keel & Mitchell, 1997; Newton, Freeman, * Corresponding author. E-mail address: [email protected] (J.A. Nasser). 0195-6663/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2004.04.006
& Munro, 1993; Wolfe, Jimerson, & Levine, 1994). Bushnell et al. (1996) reports that rates of impulsivity in those with substance use disorder or Bulimia Nervosa (BN), who are not in treatment programs, are no greater than in the general population. Furthermore, he cautions that eating disorder patients in treatment programs, who often have multiple co-morbid conditions, may show greater impulsivity and therefore not be representative of individuals with eating disorders in general. The lack of consistent association of impulsivity with treatment response lends support to Bushnell’s suggestion for caution. Wolfe et al. (1994) report that in BN patients, impulsivity did not correlate with frequency of binge eating episodes, while Fahy and Eisler (1993) and Keel and Mitchell (1997) found association between poor treatment response and high impulsivity in patients with BN. Steiger et al. (2001) demonstrated that impulsivity in women with BN is related to decreased serotonin functioning, and Fischer, Smith, and Anderson (2003) suggest that impulsivity in those with BN, correlates with eating in response to negative affect, which is also related to aberrant
304
J.A. Nasser et al. / Appetite 43 (2004) 303–307
serotonin functioning. The lack of association between impulsivity and severity of binge eating in BN may be related to variations in serotonin aberrations among individuals with BN. Another possibility as Fischer et al. (2003) suggest, is that a different component of impulsivity, i.e. ‘urgency’ impulsivity (the tendency to act rashly when experiencing negative affect) rather than the more commonly assessed ‘lack of planning’ impulsivity, correlates positively with bulimia symptoms. Fischer et al. results would seem to agree with those of Steiger, Lehoux, and Gauwin (1999) who reported that the ‘urge to binge’ correlates with ‘dietary cognitive control’ in BN individuals with low impulsivity but not in those with high impulsivity. Prior work of Heatherton and Baumeister (1991) and Herman and Polivy (1980) demonstrated a correlation between ‘dietary cognitive restraint’ and incidence of binge eating, and high dietary cognitive restraint is common in those who are obese (Herman & Polivy, 1980). Very little has been noted about impulsivity and Binge Eating Disorder (BED) in obese individuals, and available instruments to measure impulsivity (Eysenck and Eysenck, 1977, Barratt Impulsivity Scale, Patton, Stanford, & Barratt , 1995) do not specifically include items related to eating behavior. Barratt (1993) describes the impulsive person as someone who acts without thinking, acts on the spur of the moment, is restless when required to sit still, likes to take chances, is happy-go-lucky, has difficulty in concentrating, and is a doer and not a thinker.The clinical definition (DSM-IV) for BED includes a number of criteria which could be considered descriptive of impulsive behavior, namely: (1) intake of an abnormally large amount of food in a short period of time, (2) the frequency of binge episodes, and (3) being embarrassed and disgusted about overeating which leads to eating in isolation, (4) lack of control over eating during a binge episode, (5) eating more rapidly, (6) eating when not physically hungry, (7) eating until one feels uncomfortably full. Consequently, we aimed to investigate the relationship between impulsivity, laboratory test meal intake (which is increased in those with BED, Geliebter, Hassid, and Hashim (2001), and correlates with their larger gastric capacity, Geliebter, Yahav, Gluck, & Hashim, 2004), self-reported binge eating symptoms, and negative affect in obese women enrolled in a weight loss research study. We hypothesized that women meeting criteria for BED would show greater impulsivity as measured by the three subscales of the Barratt Impulsivity Scale (BIS, Patton et al., 1995). The three subscales of the BIS include: Nonplanning Impulsivity, characterized as ‘present orientation or a lack of futuring’; Motor Impulsivity, defined as acting without thinking; and Cognitive Impulsivity, the making of quick decisions. Additionally, we hypothesized that measures of food consumption from a laboratory test meal study (intake, duration) as well as selected BED criteria would correlate positively with BIS scores, and that impulsivity and performance in the test meal would be positively correlated with negative affect.
Methods Participants Obese individuals were recruited through local advertisements for participation in an outpatient weight loss program at the New York Obesity Research Center of St Luke’s/Roosevelt Hospital. During the initial phone interview, participants were screened for serious illness such as heart disease, cancer, and diabetes. Women could not be pregnant or lactating. A physical examination including medical history, ECG, and blood tests, was performed to ensure otherwise good health. The study protocol and consent form had been approved by the IRB. Participants completed the Questionnaire on Eating and Weight Patterns (QEWP; Spitzer, Devlin, & Walsh, 1992). A clinical interview was also conducted to confirm diagnostic questions from the QEWP. Three classification groups based on severity of binge eating behavior were: Controls (no binge eating episodes, n ¼ 11), subthreshold BED or Binge Eaters (fewer than 2 binge episodes/week for 6 months, n ¼ 11) and BED (full DSM -IV criteria for BED, including 2 or more binge episodes/week for 6 months, n ¼ 11). The three groups were compared in age and BMI, as binge size in BED may be related to BMI (Guss, Kissileff, Devlin, Zimmerli, & Walsh, 2002), and Table 1 reveals that there were no group differences for age (F ¼ 0:63; p ¼ 0:54) or BMI F ¼ 0:18; p ¼ 0:84). Procedures Before beginning the weight-loss phase of the program, participants were given their previously determined preferred flavor of a complete liquid test meal intake after fasting for 8 h. Prior to the test meal, participants completed a battery of psychological scales. First, they completed the Zung Depression Self-Rating Scale (Zung, 1965). For this study, questions 5 and 7 were automatically scored ‘1’ (‘none or a little of the time’) because they addressed decreased eating and weight due to depression, not applicable to overweight participants seeking weight loss. Therefore the raw score total (possible range 20 –74) was used for analysis, rather than the adjusted SDS index score. Participants also completed the Restraint Scale (Herman & Polivy, 1980), the Barratt Impulsivity Scale (Patton et al., 1995), and a subset ðn ¼ 15Þ of the participants completed the Binge Eating Scale (BES, Gormally, Black, Daston, & Rardin, 1982). Participants also completed Visual Analogue Scale (VAS) ratings of their mood before and after consuming the test meal. After completing the scales, participants were presented with the test meal to determine spontaneous meal intake. The liquid test meal consisted of Boost (Mead Johnson), a nutritionally complete food, which was prepared 1:1 with water. The test meal (as served) had an energy density of 2.1 J/g (0.5 kcal/g). They ingested the meal through a straw
J.A. Nasser et al. / Appetite 43 (2004) 303–307
305
Table 1 Characteristics and test scores of the female participants by group Mean ^ SD Subject variable
Controls n ¼ 11
Binge eaters n ¼ 11
Full BED n ¼ 11
Age BMI Weight BES* Restraint scale Zung raw score þ Mood before meal
34.3 (8.1) 34.5 (5.1) 89.7 (14.4) 32.7 (6.2)* ðn ¼ 4Þ 16.9 (4.9) 33.3 (5.5) þ 59.4 (14)
28.2 (6.9) 34.2 (3.5) 92.5 (6.5) 32.3 (7.2)* ðn ¼ 6Þ 20.4 (3.9) 35.7 (7.7) 58.7 (11.5)
29.0 (8.3) 36.6 (6.2) 97.8 (18.8) 42.5 (2.7)* ðn ¼ 6Þ 20.5 (5.3) 42.3 (10.2) þ 56.3 (20.5)
Data analyzed by ANOVA; Post hoc analysis by LSD; ANOVA, * p ¼ 0:015: LSD, * p ¼ 0:019 (Control vs. BED) and * p ¼ 0:008 (Binge eaters vs. BED); ANOVA, þp ¼ 0:037; LSD þp ¼ 0:013 (control vs. BED).
from a large opaque cooler to prevent visual feedback. To emphasize that the test meal was to be considered a large meal, they were instructed to ingest it until they felt extremely full and not to consume any other food for 5 h. They completed various ratings of the test meal flavor afterwards, and were under the impression that it was the focus of the test meal study. To determine intake, the contents were weighed prior to and after consumption when the participant was not present. Data analysis Primary outcome measures were: test meal intake (grams), meal duration (minutes), scores on the BIS, Zung score and VAS mood ratings. Scores on the BES and Restraint scales were used to further characterize the groups and to serve as covariates in ANOVA. ANOVA was used to evaluate between-group differences in the outcome variables, followed by post hoc tests (LSD) if the overall ANOVA was significant. Linear regression was employed to examine the association of test meal intake with the outcome variables. Non-parametric correlations (Spearman’s rho) were used to examine the relationship of BIS scores with select criteria of binge eating behavior (yes/no responses to specific items of the QEWP). Data were analyzed with the Statistical Package for the Social Sciences (SPSS version 11, 2001, Chicago, IL). Two tailed p , 0:05 was needed for significance.
Table 2 shows the test meal measures and BIS Impulsivity scores in the three groups. There was no significant difference between groups with respect to test meal intake (F ¼ 0:37; p ¼ 0:69), or duration (F ¼ 0:65; p ¼ 0:53). Statistically controlling for BES, Restraint Score, Zung score or mood rating did not increase significance of the differences in test meal intake and duration between the groups. There was a significant difference in the BIS-Motor subscale (F ¼ 3:3; p ¼ 0:05) between the groups but not in the Non-planning (F ¼ 1:3; p ¼ 0:28) or Cognitive subscales (F ¼ 1:4; p ¼ 0:27). Post hoc analysis (LSD) revealed a significantly higher score in the BIS-Motor scale in the full BED group compared to the control group ðp ¼ 0:01Þ: This relationship was still significant after controlling for BMI (F ¼ 4:96; p ¼ 0:01). A significant negative correlation was observed between BIS Non-planning subscore and test meal intake (r ¼ 20:44; p , 0:01) and duration (r ¼ 20:50; p , 0:01), for the whole sample of 33 participants. Table 3 shows that there were also positive significant correlations ðp , 0:05Þ; between Total Impulsivity BIS score and the BED criteria ‘Loss of Control during a binge’ ðRho ¼ 0:47Þ; ‘Eating when not physically hungry’ (Rho ¼ 0.50) and ‘Eating alone due to embarrassment’ (Rho ¼ 0.53).
Table 2 Impulsivity and laboratory test meal measures by group Mean ^ SD
Results
Measures
Controls
Binge eaters
BED
Demographic characteristics (Section 2) of the groups are displayed in Table 1. A significant difference was observed between the groups in the BES score ðp ¼ 0:015Þ with BED . Binge Eaters , Controls, and in the Zung score ðp ¼ 0:037Þ with BED . Binge Eaters ¼ Controls. No significant difference was observed in Restraint score between the groups or in age, weight, BMI, or rating of mood before the test meal.
Test meal intake (g) þ Meal duration (min) Total impulsivity I-Motor* I-Cognitive I-Nonplanning
916 (322) 6.9 (7.2) 62.0 (9.8) 20.7 (4.1) þ 15.7 (3.5) 25.5 (4.2)
1032 (416) 10.2 (9.0) 63.0 (8.6) 23.4 (3.3) 14.9 (3.4) 24.5 (4.4)
1029 (342) 9.8.0 (5.1) 70.9 (12) 26.2 (6.8)** 17.3 (3.5) 27.4 (4.2)
Data analyzed by ANOVA, * p ¼ 0:05; post hoc analysis by LSD, * * p ¼ 0:01: This relationship was still significant after controlling for BMI: F ¼ 4:96; p ¼ 0:01; þ Test meal contains 2.1 J/g (0.5 kcal/g).
306
J.A. Nasser et al. / Appetite 43 (2004) 303–307
Table 3 Correlation (Spearman Rho) of impulsivity subscores with select criteria of Binge eating disorder BIS impulsivity scores
Loss of control during a Binge
Eating when not physically hungry
Eating alone embarrassed
Non-planning Motor Cognitive Total
Rho ¼ 0.57, p ¼ 0:003* * Rho ¼ 0.27, p ¼ 0:18 Rho ¼ 0.28, p ¼ 0:17 Rho ¼ 0.47, p ¼ 0:02*
Rho ¼ 0.35, p ¼ 0:15 Rho ¼ 0.56, p ¼ 0:01* * Rho ¼ 0.12, p ¼ 0:63 Rho ¼ 0.50, p ¼ 0:03*
Rho ¼ 0.07, p ¼ 0:78 Rho ¼ 0.45, p ¼ 0:05* Rho ¼ 0.71, p ¼ 0:001* * Rho ¼ 0.53, p ¼ 0:02*
Table 4 shows an evaluation of the laboratory test meal as a ‘surrogate binge’. Significant positive Spearman Rho correlations were observed between BED status and test meal duration (Rho ¼ 0.375, p ¼ 0:03) and between BED status and Motor Impulsivity (Rho ¼ 0.41, p ¼ 0:02). There was also a positive correlation between Zung Score and test meal duration (r ¼ 0:35; p ¼ 0:04). After controlling for BMI and BED status, we also observed a significant positive correlation between ‘mood rating before the meal’ and Motor Impulsivity (r ¼ 0:52; p ¼ 0:003).
Discussion We observed a significant difference in the BIS-Motor subscale, with greater Motor Impulsivity in BED than in controls, as well as a significant correlation between BED status and duration of the test meal, a positive correlation between Zung Depression score and duration of the test meal, and a positive correlation between mood rating before the meal and Motor impulsivity. Since the Motor subscale of the BIS measures the tendency to ‘act without thinking’ or rather, to react; these observations suggest that ‘binge’ food consumption in individuals with BED could occur as a response to a negative mood (as was reported by Fischer et al. (2003) in those with BN). Negative correlations were observed between BIS-Non-planning and both test meal intake and test meal duration in the full sample, as was reported by Fischer et al. (2003) in those in BN. This was in contrast to the positive correlation between Non-planning impulsiveness and ‘Loss of Control during a Binge’. It is possible that the ‘planned’ nature of the laboratory test meal paradigm, with instructions to consume the test meal ‘until you feel extremely full’ gave the subject some control over the eating, which is somewhat different than binge eating. Although test meal intakes were in Table 4 Correlations among test meal duration, BED status, negative affect, and motor impulsivity Test meal duration
BIS-Motor Imp
BED status Rho ¼ 0.375, p ¼ 0:03* Rho ¼ 0.41, p ¼ 0:03* Zung Score R ¼ 0:35; p ¼ 0:04* R ¼ 20:02; p ¼ 0:89 Mood before test meala R ¼ 0:05; p ¼ 0:78 R ¼ 0:52; p ¼ 0:003* * a
Correlations with nood rating after controlling for BED status and BMI.
the direction expected with larger intakes for the binge eaters than normal, they were not significant, unlike previous findings from this group (Geliebter et al., 2001), and are likely due to the relatively small numbers of participants per group in this study. In this study, we used three groups similar in number of participants, age, weight, BMI and sex in a laboratory test meal paradigm to identify for the first time a positive association between impulsivity and elements of BED status: Loss of Control during a binge, Eating when not physically hungry, and Eating Alone and being embarrassed by eating. Other studies, while reporting the co-incidence of disordered eating and impulsivity, have not found a correlation between binge frequency and impulsivity. Available impulsivity instruments (i.e. BIS) do not contain specific examples of impulsive eating behavior; however, our results suggest that incorporating eating behavior would be worth exploring. A weakness of this study is that it involves correlations, and therefore cannot establish causal sequence between impulsivity and BED. Although we do not have measures of serotonin (a neurotransmitter associated with impulsivity), the association between impulsivity, binge eating behavior and negative affect suggests that in addition to selective serotonin reuptake inhibitors (SSRI’s), it may be worthwhile to explore the potential use of serotonin agonists to reduce impulsivity and binge eating behavior. In conclusion, we have shown greater Motor impulsivity in obese individuals with BED, positive correlation between test meal duration and negative affect, and positive correlations between Non-planning impulsivity and Loss of Control over a Binge, Motor impulsivity and Eating when not physically hungry, and Cognitive impulsivity and Eating Alone because of being embarrassed. These associations suggest possible new targets for cognitive behavioral therapy and a potential role for serotonin directed pharmacotherapy especially in those who are resistant to CBT treatment of BED.
References Allen, T. J., Moeller, F. G., Rhoades, H. M., & Cherek, D. R. (1998). Impulsivity and history of drug dependence. Drug and Alcohol Dependence, 50, 137– 145. Barratt, E. S. (1993). Impulsivity: integrating cognitive, behavioral, biological and environmental data. In W. McCowan, & M. Shuce (Eds.), The impulsive client: Theory, research and treatment (pp. 39 –56). Washington, DC: American Psychological Association.
J.A. Nasser et al. / Appetite 43 (2004) 303–307 Bushnell, J. A., Wells, J. E., & Oakley-Browne, M. A. (1996). Impulsivity in disordered eating, affective disorders and substance use disorders. British Journal of Psychiatry, 169, 329 –333. Evenden, J. L. (1999). Varieties of impulsivity. Psychopharmacology, 146, 348–361. Eysenck, S. B., & Eysenck, H. J. (1977). The place of impulsiveness in a dimensional system of personality description. British Journal of Social Psychology, 16, 57–68. Fahy, T., & Eisler, I. (1993). Impulsivity and eating disorders. British Journal of Psychiatry, 162, 193–197. Fischer, S., Smith, G. T., & Anderson, K. G. (2003). Clarifying the role of impulsivity in bulimia nervosa. The International Journal of Eating Disorders, 33, 406–411. Geliebter, A., Hassid, G., & Hashim, S. A. (2001). Test meal intake in obese binge eaters in relation to mood and gender. The International Journal of Eating Disorders, 29, 488–494. Geliebter, A., Yahav, E., Gluck, M. E., & Hashim, S. A. (2004). Gastric capacity, test meal intake, and appetitive hormones in binge eating disorder. Physiology and Behavior, 81, 735–740. Gormally, J., Black, S., Daston, S., & Rardin, D. (1982). The assessment of binge eating severity among obese persons. Addictive Behaviors, 7, 47–55. Guss, J. L., Kissileff, H. R., Devlin, M. J., Zimmerli, E., & Walsh, B. T. (2002). Binge size increases with body mass index in women with Binge eating disorder. Obesity Research, 10, 1021–1029. Heatherton, T. F., & Baumeister, B. F. (1991). Binge eating as escape from self-awareness. Psychological Bulletin, 110, 86–108. Herman, C. P., & Polivy, J. (1980). Restrained eating. In A. J. Stunkard (Ed.), Obesity. Philadelphia: WB Saunders. Hollander, E., & Evers, M. (2002). New developments in impulsivity. Lancet, 358, 949– 950. Keel, P. K., & Mitchell, J. E. (1997). Outcomes in bulimia nervosa. American Journal of Psychiatry, 154, 313–321.
307
Mitchell, S. H. (1999). Measures of impulsivity in cigarette smokers and non-smokers. Psychopharmacology, 146, 455–464. Newton, T. R., Freeman, C. P., & Munro, J. (1993). Impulsivity and dyscontrol in bulimia nervosa: is impulsivity an independent phenomenon or a marker for severity. Acta Psychiatrica Scandinavica, 87, 389– 394. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt impulsiveness scale. Journal of Clinical Psychology, 51, 768– 774. Richards, J. B., Zhang, L., Mitchell, S. H., & de Wit, H. (1999). Delay or probability discounting in a model of impulsive behavior: effect of alcohol. Journal of the Experimental Analysis of Behavior, 71, 121– 143. Spitzer, R. L., Devlin, M. J., Walsh, B. T., et al. (1992). Binge eating disorder: a multisite field trial of the diagnostic criteria. The International Journal of Eating Disorders, 11, 191 –203. Steiger, H., Lehoux, P. M., & Gauwin, L. (1999). Impulsivity, dietary control, and the urge to binge in bulimic syndromes. The International Journal of Eating Disorders, 26, 261–274. Steiger, H., Young, S. N., Kin, N. M., Koerner, N., Israel, M., Lageix, P., & Paris, J. (2001). Implications of impulsive and affective symptoms for serotonin function in bulimia nervosa. Psychological Medicine, 31, 85– 95. Swann, A. C., Bjork, J. M., Moeller, F. G., & Dougherty, D. M. (2002). Two models of impulsivity: relationship of personality traits and psychopathology. Biological Psychiatry, 51, 988– 994. Wolfe, B. E., Jimerson, D. C., & Levine, J. M. (1994). Impulsivity ratings in bulimia nervosa: relationship to binge eating behavior. The International Journal of Eating Disorders, 15, 289–292. Zung, W. W. K. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63 –70.
Impulsivity and test meal intake in obese binge eating women Jennifer A. Nassera,*, Marci E. Glucka, Allan Geliebtera,b a
Department of Medicine, NY Obesity Research Center, St Luke’s/Roosevelt Hospital Center, Columbia University College of Physicians and Surgeons, WH-1020, 1111 Amsterdam Ave, New York, NY 10025, USA b Department of Psychiatry, NY Obesity Research Center, St. Luke’s/Roosevelt Hospital Center, Columbia University College of Physicians and Surgeons, New York, NY 10025, USA Received 5 March 2004; revised 13 April 2004; accepted 15 April 2004
Abstract Greater impulsivity has been observed in those with chemical (cocaine, marijuana, alcohol) and behavioral addictions (gambling, sex, shopping), as well as in individuals with personality and conduct disorders. Greater impulsivity has also been described in those with Bulimia Nervosa and attributed to aberrations in serotonin, as has eating in response to negative affect.However, less is known about the impact of impulsivity on eating behavior in obese humans in general, and in those who meet sub-clinical and full clinical criteria for Binge Eating Disorder (BED) in particular. Using a laboratory test meal paradigm, we demonstrated: (1) greater Motor Impulsivity (Barratt Impulsivity Scale (BIS) ðp ¼ 0:05Þ in those with BED ðn ¼ 11Þ as compared to those without BED ðn ¼ 11Þ; (2) a positive correlation between BED criteria and BIS scores ðp , 0:01Þ; (3) a positive correlation between test meal duration and Zung Depression Score, and (4) a positive correlation between Motor Impulsivity and mood rated before consuming the test meal. These associations suggest potential aberrations in serotonin transmission in BED, and a possible target for pharmacotherapy of BED especially in those who are resistant to Cognitive Behavioral Therapy. q 2004 Elsevier Ltd. All rights reserved. Keywords: Impulsivity; Binge eating; Obesity; Test meal intake
Introduction Current literature define impulsivity as (1) the inclination to choose small, immediately available rewards over larger, delayed rewards, and/or (2) the inclination to respond rapidly without forethought and/or attention to consequences (Evenden, 1999; Swann, Bjork, Moeller, & Dougherty, 2002). Greater impulsivity has been reported in those with chemical (cocaine, alcohol, nicotine) (Allen, Moeller, Rhoades, & Cherek, 1998; Mitchell, 1999; Richards, Zhang, Mitchell, & de Wit, 1999) and behavioral addictions (gambling, sex, shopping) as well as in those with personality and conduct disorders (Hollander & Evers, 2002), where it correlates with severity of dependence or disordered state. Some investigators, using treatment program patients, report greater impulsivity in individuals with eating disorders compared to normal controls (Fahy & Eisler, 1993; Keel & Mitchell, 1997; Newton, Freeman, * Corresponding author. E-mail address: [email protected] (J.A. Nasser). 0195-6663/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2004.04.006
& Munro, 1993; Wolfe, Jimerson, & Levine, 1994). Bushnell et al. (1996) reports that rates of impulsivity in those with substance use disorder or Bulimia Nervosa (BN), who are not in treatment programs, are no greater than in the general population. Furthermore, he cautions that eating disorder patients in treatment programs, who often have multiple co-morbid conditions, may show greater impulsivity and therefore not be representative of individuals with eating disorders in general. The lack of consistent association of impulsivity with treatment response lends support to Bushnell’s suggestion for caution. Wolfe et al. (1994) report that in BN patients, impulsivity did not correlate with frequency of binge eating episodes, while Fahy and Eisler (1993) and Keel and Mitchell (1997) found association between poor treatment response and high impulsivity in patients with BN. Steiger et al. (2001) demonstrated that impulsivity in women with BN is related to decreased serotonin functioning, and Fischer, Smith, and Anderson (2003) suggest that impulsivity in those with BN, correlates with eating in response to negative affect, which is also related to aberrant
304
J.A. Nasser et al. / Appetite 43 (2004) 303–307
serotonin functioning. The lack of association between impulsivity and severity of binge eating in BN may be related to variations in serotonin aberrations among individuals with BN. Another possibility as Fischer et al. (2003) suggest, is that a different component of impulsivity, i.e. ‘urgency’ impulsivity (the tendency to act rashly when experiencing negative affect) rather than the more commonly assessed ‘lack of planning’ impulsivity, correlates positively with bulimia symptoms. Fischer et al. results would seem to agree with those of Steiger, Lehoux, and Gauwin (1999) who reported that the ‘urge to binge’ correlates with ‘dietary cognitive control’ in BN individuals with low impulsivity but not in those with high impulsivity. Prior work of Heatherton and Baumeister (1991) and Herman and Polivy (1980) demonstrated a correlation between ‘dietary cognitive restraint’ and incidence of binge eating, and high dietary cognitive restraint is common in those who are obese (Herman & Polivy, 1980). Very little has been noted about impulsivity and Binge Eating Disorder (BED) in obese individuals, and available instruments to measure impulsivity (Eysenck and Eysenck, 1977, Barratt Impulsivity Scale, Patton, Stanford, & Barratt , 1995) do not specifically include items related to eating behavior. Barratt (1993) describes the impulsive person as someone who acts without thinking, acts on the spur of the moment, is restless when required to sit still, likes to take chances, is happy-go-lucky, has difficulty in concentrating, and is a doer and not a thinker.The clinical definition (DSM-IV) for BED includes a number of criteria which could be considered descriptive of impulsive behavior, namely: (1) intake of an abnormally large amount of food in a short period of time, (2) the frequency of binge episodes, and (3) being embarrassed and disgusted about overeating which leads to eating in isolation, (4) lack of control over eating during a binge episode, (5) eating more rapidly, (6) eating when not physically hungry, (7) eating until one feels uncomfortably full. Consequently, we aimed to investigate the relationship between impulsivity, laboratory test meal intake (which is increased in those with BED, Geliebter, Hassid, and Hashim (2001), and correlates with their larger gastric capacity, Geliebter, Yahav, Gluck, & Hashim, 2004), self-reported binge eating symptoms, and negative affect in obese women enrolled in a weight loss research study. We hypothesized that women meeting criteria for BED would show greater impulsivity as measured by the three subscales of the Barratt Impulsivity Scale (BIS, Patton et al., 1995). The three subscales of the BIS include: Nonplanning Impulsivity, characterized as ‘present orientation or a lack of futuring’; Motor Impulsivity, defined as acting without thinking; and Cognitive Impulsivity, the making of quick decisions. Additionally, we hypothesized that measures of food consumption from a laboratory test meal study (intake, duration) as well as selected BED criteria would correlate positively with BIS scores, and that impulsivity and performance in the test meal would be positively correlated with negative affect.
Methods Participants Obese individuals were recruited through local advertisements for participation in an outpatient weight loss program at the New York Obesity Research Center of St Luke’s/Roosevelt Hospital. During the initial phone interview, participants were screened for serious illness such as heart disease, cancer, and diabetes. Women could not be pregnant or lactating. A physical examination including medical history, ECG, and blood tests, was performed to ensure otherwise good health. The study protocol and consent form had been approved by the IRB. Participants completed the Questionnaire on Eating and Weight Patterns (QEWP; Spitzer, Devlin, & Walsh, 1992). A clinical interview was also conducted to confirm diagnostic questions from the QEWP. Three classification groups based on severity of binge eating behavior were: Controls (no binge eating episodes, n ¼ 11), subthreshold BED or Binge Eaters (fewer than 2 binge episodes/week for 6 months, n ¼ 11) and BED (full DSM -IV criteria for BED, including 2 or more binge episodes/week for 6 months, n ¼ 11). The three groups were compared in age and BMI, as binge size in BED may be related to BMI (Guss, Kissileff, Devlin, Zimmerli, & Walsh, 2002), and Table 1 reveals that there were no group differences for age (F ¼ 0:63; p ¼ 0:54) or BMI F ¼ 0:18; p ¼ 0:84). Procedures Before beginning the weight-loss phase of the program, participants were given their previously determined preferred flavor of a complete liquid test meal intake after fasting for 8 h. Prior to the test meal, participants completed a battery of psychological scales. First, they completed the Zung Depression Self-Rating Scale (Zung, 1965). For this study, questions 5 and 7 were automatically scored ‘1’ (‘none or a little of the time’) because they addressed decreased eating and weight due to depression, not applicable to overweight participants seeking weight loss. Therefore the raw score total (possible range 20 –74) was used for analysis, rather than the adjusted SDS index score. Participants also completed the Restraint Scale (Herman & Polivy, 1980), the Barratt Impulsivity Scale (Patton et al., 1995), and a subset ðn ¼ 15Þ of the participants completed the Binge Eating Scale (BES, Gormally, Black, Daston, & Rardin, 1982). Participants also completed Visual Analogue Scale (VAS) ratings of their mood before and after consuming the test meal. After completing the scales, participants were presented with the test meal to determine spontaneous meal intake. The liquid test meal consisted of Boost (Mead Johnson), a nutritionally complete food, which was prepared 1:1 with water. The test meal (as served) had an energy density of 2.1 J/g (0.5 kcal/g). They ingested the meal through a straw
J.A. Nasser et al. / Appetite 43 (2004) 303–307
305
Table 1 Characteristics and test scores of the female participants by group Mean ^ SD Subject variable
Controls n ¼ 11
Binge eaters n ¼ 11
Full BED n ¼ 11
Age BMI Weight BES* Restraint scale Zung raw score þ Mood before meal
34.3 (8.1) 34.5 (5.1) 89.7 (14.4) 32.7 (6.2)* ðn ¼ 4Þ 16.9 (4.9) 33.3 (5.5) þ 59.4 (14)
28.2 (6.9) 34.2 (3.5) 92.5 (6.5) 32.3 (7.2)* ðn ¼ 6Þ 20.4 (3.9) 35.7 (7.7) 58.7 (11.5)
29.0 (8.3) 36.6 (6.2) 97.8 (18.8) 42.5 (2.7)* ðn ¼ 6Þ 20.5 (5.3) 42.3 (10.2) þ 56.3 (20.5)
Data analyzed by ANOVA; Post hoc analysis by LSD; ANOVA, * p ¼ 0:015: LSD, * p ¼ 0:019 (Control vs. BED) and * p ¼ 0:008 (Binge eaters vs. BED); ANOVA, þp ¼ 0:037; LSD þp ¼ 0:013 (control vs. BED).
from a large opaque cooler to prevent visual feedback. To emphasize that the test meal was to be considered a large meal, they were instructed to ingest it until they felt extremely full and not to consume any other food for 5 h. They completed various ratings of the test meal flavor afterwards, and were under the impression that it was the focus of the test meal study. To determine intake, the contents were weighed prior to and after consumption when the participant was not present. Data analysis Primary outcome measures were: test meal intake (grams), meal duration (minutes), scores on the BIS, Zung score and VAS mood ratings. Scores on the BES and Restraint scales were used to further characterize the groups and to serve as covariates in ANOVA. ANOVA was used to evaluate between-group differences in the outcome variables, followed by post hoc tests (LSD) if the overall ANOVA was significant. Linear regression was employed to examine the association of test meal intake with the outcome variables. Non-parametric correlations (Spearman’s rho) were used to examine the relationship of BIS scores with select criteria of binge eating behavior (yes/no responses to specific items of the QEWP). Data were analyzed with the Statistical Package for the Social Sciences (SPSS version 11, 2001, Chicago, IL). Two tailed p , 0:05 was needed for significance.
Table 2 shows the test meal measures and BIS Impulsivity scores in the three groups. There was no significant difference between groups with respect to test meal intake (F ¼ 0:37; p ¼ 0:69), or duration (F ¼ 0:65; p ¼ 0:53). Statistically controlling for BES, Restraint Score, Zung score or mood rating did not increase significance of the differences in test meal intake and duration between the groups. There was a significant difference in the BIS-Motor subscale (F ¼ 3:3; p ¼ 0:05) between the groups but not in the Non-planning (F ¼ 1:3; p ¼ 0:28) or Cognitive subscales (F ¼ 1:4; p ¼ 0:27). Post hoc analysis (LSD) revealed a significantly higher score in the BIS-Motor scale in the full BED group compared to the control group ðp ¼ 0:01Þ: This relationship was still significant after controlling for BMI (F ¼ 4:96; p ¼ 0:01). A significant negative correlation was observed between BIS Non-planning subscore and test meal intake (r ¼ 20:44; p , 0:01) and duration (r ¼ 20:50; p , 0:01), for the whole sample of 33 participants. Table 3 shows that there were also positive significant correlations ðp , 0:05Þ; between Total Impulsivity BIS score and the BED criteria ‘Loss of Control during a binge’ ðRho ¼ 0:47Þ; ‘Eating when not physically hungry’ (Rho ¼ 0.50) and ‘Eating alone due to embarrassment’ (Rho ¼ 0.53).
Table 2 Impulsivity and laboratory test meal measures by group Mean ^ SD
Results
Measures
Controls
Binge eaters
BED
Demographic characteristics (Section 2) of the groups are displayed in Table 1. A significant difference was observed between the groups in the BES score ðp ¼ 0:015Þ with BED . Binge Eaters , Controls, and in the Zung score ðp ¼ 0:037Þ with BED . Binge Eaters ¼ Controls. No significant difference was observed in Restraint score between the groups or in age, weight, BMI, or rating of mood before the test meal.
Test meal intake (g) þ Meal duration (min) Total impulsivity I-Motor* I-Cognitive I-Nonplanning
916 (322) 6.9 (7.2) 62.0 (9.8) 20.7 (4.1) þ 15.7 (3.5) 25.5 (4.2)
1032 (416) 10.2 (9.0) 63.0 (8.6) 23.4 (3.3) 14.9 (3.4) 24.5 (4.4)
1029 (342) 9.8.0 (5.1) 70.9 (12) 26.2 (6.8)** 17.3 (3.5) 27.4 (4.2)
Data analyzed by ANOVA, * p ¼ 0:05; post hoc analysis by LSD, * * p ¼ 0:01: This relationship was still significant after controlling for BMI: F ¼ 4:96; p ¼ 0:01; þ Test meal contains 2.1 J/g (0.5 kcal/g).
306
J.A. Nasser et al. / Appetite 43 (2004) 303–307
Table 3 Correlation (Spearman Rho) of impulsivity subscores with select criteria of Binge eating disorder BIS impulsivity scores
Loss of control during a Binge
Eating when not physically hungry
Eating alone embarrassed
Non-planning Motor Cognitive Total
Rho ¼ 0.57, p ¼ 0:003* * Rho ¼ 0.27, p ¼ 0:18 Rho ¼ 0.28, p ¼ 0:17 Rho ¼ 0.47, p ¼ 0:02*
Rho ¼ 0.35, p ¼ 0:15 Rho ¼ 0.56, p ¼ 0:01* * Rho ¼ 0.12, p ¼ 0:63 Rho ¼ 0.50, p ¼ 0:03*
Rho ¼ 0.07, p ¼ 0:78 Rho ¼ 0.45, p ¼ 0:05* Rho ¼ 0.71, p ¼ 0:001* * Rho ¼ 0.53, p ¼ 0:02*
Table 4 shows an evaluation of the laboratory test meal as a ‘surrogate binge’. Significant positive Spearman Rho correlations were observed between BED status and test meal duration (Rho ¼ 0.375, p ¼ 0:03) and between BED status and Motor Impulsivity (Rho ¼ 0.41, p ¼ 0:02). There was also a positive correlation between Zung Score and test meal duration (r ¼ 0:35; p ¼ 0:04). After controlling for BMI and BED status, we also observed a significant positive correlation between ‘mood rating before the meal’ and Motor Impulsivity (r ¼ 0:52; p ¼ 0:003).
Discussion We observed a significant difference in the BIS-Motor subscale, with greater Motor Impulsivity in BED than in controls, as well as a significant correlation between BED status and duration of the test meal, a positive correlation between Zung Depression score and duration of the test meal, and a positive correlation between mood rating before the meal and Motor impulsivity. Since the Motor subscale of the BIS measures the tendency to ‘act without thinking’ or rather, to react; these observations suggest that ‘binge’ food consumption in individuals with BED could occur as a response to a negative mood (as was reported by Fischer et al. (2003) in those with BN). Negative correlations were observed between BIS-Non-planning and both test meal intake and test meal duration in the full sample, as was reported by Fischer et al. (2003) in those in BN. This was in contrast to the positive correlation between Non-planning impulsiveness and ‘Loss of Control during a Binge’. It is possible that the ‘planned’ nature of the laboratory test meal paradigm, with instructions to consume the test meal ‘until you feel extremely full’ gave the subject some control over the eating, which is somewhat different than binge eating. Although test meal intakes were in Table 4 Correlations among test meal duration, BED status, negative affect, and motor impulsivity Test meal duration
BIS-Motor Imp
BED status Rho ¼ 0.375, p ¼ 0:03* Rho ¼ 0.41, p ¼ 0:03* Zung Score R ¼ 0:35; p ¼ 0:04* R ¼ 20:02; p ¼ 0:89 Mood before test meala R ¼ 0:05; p ¼ 0:78 R ¼ 0:52; p ¼ 0:003* * a
Correlations with nood rating after controlling for BED status and BMI.
the direction expected with larger intakes for the binge eaters than normal, they were not significant, unlike previous findings from this group (Geliebter et al., 2001), and are likely due to the relatively small numbers of participants per group in this study. In this study, we used three groups similar in number of participants, age, weight, BMI and sex in a laboratory test meal paradigm to identify for the first time a positive association between impulsivity and elements of BED status: Loss of Control during a binge, Eating when not physically hungry, and Eating Alone and being embarrassed by eating. Other studies, while reporting the co-incidence of disordered eating and impulsivity, have not found a correlation between binge frequency and impulsivity. Available impulsivity instruments (i.e. BIS) do not contain specific examples of impulsive eating behavior; however, our results suggest that incorporating eating behavior would be worth exploring. A weakness of this study is that it involves correlations, and therefore cannot establish causal sequence between impulsivity and BED. Although we do not have measures of serotonin (a neurotransmitter associated with impulsivity), the association between impulsivity, binge eating behavior and negative affect suggests that in addition to selective serotonin reuptake inhibitors (SSRI’s), it may be worthwhile to explore the potential use of serotonin agonists to reduce impulsivity and binge eating behavior. In conclusion, we have shown greater Motor impulsivity in obese individuals with BED, positive correlation between test meal duration and negative affect, and positive correlations between Non-planning impulsivity and Loss of Control over a Binge, Motor impulsivity and Eating when not physically hungry, and Cognitive impulsivity and Eating Alone because of being embarrassed. These associations suggest possible new targets for cognitive behavioral therapy and a potential role for serotonin directed pharmacotherapy especially in those who are resistant to CBT treatment of BED.
References Allen, T. J., Moeller, F. G., Rhoades, H. M., & Cherek, D. R. (1998). Impulsivity and history of drug dependence. Drug and Alcohol Dependence, 50, 137– 145. Barratt, E. S. (1993). Impulsivity: integrating cognitive, behavioral, biological and environmental data. In W. McCowan, & M. Shuce (Eds.), The impulsive client: Theory, research and treatment (pp. 39 –56). Washington, DC: American Psychological Association.
J.A. Nasser et al. / Appetite 43 (2004) 303–307 Bushnell, J. A., Wells, J. E., & Oakley-Browne, M. A. (1996). Impulsivity in disordered eating, affective disorders and substance use disorders. British Journal of Psychiatry, 169, 329 –333. Evenden, J. L. (1999). Varieties of impulsivity. Psychopharmacology, 146, 348–361. Eysenck, S. B., & Eysenck, H. J. (1977). The place of impulsiveness in a dimensional system of personality description. British Journal of Social Psychology, 16, 57–68. Fahy, T., & Eisler, I. (1993). Impulsivity and eating disorders. British Journal of Psychiatry, 162, 193–197. Fischer, S., Smith, G. T., & Anderson, K. G. (2003). Clarifying the role of impulsivity in bulimia nervosa. The International Journal of Eating Disorders, 33, 406–411. Geliebter, A., Hassid, G., & Hashim, S. A. (2001). Test meal intake in obese binge eaters in relation to mood and gender. The International Journal of Eating Disorders, 29, 488–494. Geliebter, A., Yahav, E., Gluck, M. E., & Hashim, S. A. (2004). Gastric capacity, test meal intake, and appetitive hormones in binge eating disorder. Physiology and Behavior, 81, 735–740. Gormally, J., Black, S., Daston, S., & Rardin, D. (1982). The assessment of binge eating severity among obese persons. Addictive Behaviors, 7, 47–55. Guss, J. L., Kissileff, H. R., Devlin, M. J., Zimmerli, E., & Walsh, B. T. (2002). Binge size increases with body mass index in women with Binge eating disorder. Obesity Research, 10, 1021–1029. Heatherton, T. F., & Baumeister, B. F. (1991). Binge eating as escape from self-awareness. Psychological Bulletin, 110, 86–108. Herman, C. P., & Polivy, J. (1980). Restrained eating. In A. J. Stunkard (Ed.), Obesity. Philadelphia: WB Saunders. Hollander, E., & Evers, M. (2002). New developments in impulsivity. Lancet, 358, 949– 950. Keel, P. K., & Mitchell, J. E. (1997). Outcomes in bulimia nervosa. American Journal of Psychiatry, 154, 313–321.
307
Mitchell, S. H. (1999). Measures of impulsivity in cigarette smokers and non-smokers. Psychopharmacology, 146, 455–464. Newton, T. R., Freeman, C. P., & Munro, J. (1993). Impulsivity and dyscontrol in bulimia nervosa: is impulsivity an independent phenomenon or a marker for severity. Acta Psychiatrica Scandinavica, 87, 389– 394. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt impulsiveness scale. Journal of Clinical Psychology, 51, 768– 774. Richards, J. B., Zhang, L., Mitchell, S. H., & de Wit, H. (1999). Delay or probability discounting in a model of impulsive behavior: effect of alcohol. Journal of the Experimental Analysis of Behavior, 71, 121– 143. Spitzer, R. L., Devlin, M. J., Walsh, B. T., et al. (1992). Binge eating disorder: a multisite field trial of the diagnostic criteria. The International Journal of Eating Disorders, 11, 191 –203. Steiger, H., Lehoux, P. M., & Gauwin, L. (1999). Impulsivity, dietary control, and the urge to binge in bulimic syndromes. The International Journal of Eating Disorders, 26, 261–274. Steiger, H., Young, S. N., Kin, N. M., Koerner, N., Israel, M., Lageix, P., & Paris, J. (2001). Implications of impulsive and affective symptoms for serotonin function in bulimia nervosa. Psychological Medicine, 31, 85– 95. Swann, A. C., Bjork, J. M., Moeller, F. G., & Dougherty, D. M. (2002). Two models of impulsivity: relationship of personality traits and psychopathology. Biological Psychiatry, 51, 988– 994. Wolfe, B. E., Jimerson, D. C., & Levine, J. M. (1994). Impulsivity ratings in bulimia nervosa: relationship to binge eating behavior. The International Journal of Eating Disorders, 15, 289–292. Zung, W. W. K. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63 –70.