Elsevier

Appetite

Volume 142, 1 November 2019, 104380
Appetite

Factor structure and psychometric properties of the Inflexible Eating Questionnaire in a sample of adult women

https://doi.org/10.1016/j.appet.2019.104380Get rights and content

Abstract

The Inflexible Eating Questionnaire (IEQ) is a recently developed measure that assesses an individual's inflexible adherence to rigid eating rules, along with the tendency to respectively feel empowered or distressed when such rules are or are not followed. At present, evidence supporting the unidimensional structure and psychometric properties of the IEQ is limited to one specific sample of Portuguese adults. Establishing whether the IEQ is a valid and reliable measure in a different sample and by an independent research team is needed. We sought to examine the factor structure and psychometric properties of the IEQ in large sample (n = 1000) of Australian female adults. A unidimensional structure was replicated and evidence of internal consistency (α = .89) was found. IEQ scores were significantly and moderately correlated with various eating restraint measures and intuitive eating, providing evidence of convergent validity. IEQ scores also predicted incremental variance in global eating disorder symptomatology and psychosocial impairment after controlling for intuitive eating, flexible control, and rigid dietary control. Present findings offer further support for the validity and reliability of the IEQ in a non-clinical sample of women. A brief measure that assesses the inflexible adherence to eating rules may be valuable for validating current models of eating disorder psychopathology. Furthermore, incorporating the IEQ into the assessment of future randomized trials of eating disorder prevention or treatment programs may be beneficial for elucidating these interventions mechanisms of change.

Introduction

Numerous etiological (Stice, Nemeroff, & Shaw, 1996) and maintenance (Fairburn, 2008) models of eating disorders propose that dietary restraint increases susceptibility to disordered eating through both physiological and psychological mechanisms. However, dietary restraint is a multifaceted construct comprised of distinct forms that cannot be categorized as entirely adaptive or maladaptive. Indeed, Westenhoefer, Stunkard, and Pudel (1999) proposed that dietary restraint can be classified into a flexible and rigid form. Flexible dietary control involves a graded approach to dieting, defined by behaviors such as eating a wide variety of foods while still paying attention to one's weight and compensating at later meals (consuming “healthier” foods) if “unhealthier” foods were consumed earlier. Conversely, rigid dietary control involves an all-or-none approach to dieting, defined by behaviors such as disciplined calorie counting, avoiding desired calorie-dense foods, and fasting or skipping meals for weight-loss purposes (Westenhoefer et al., 1999). A solid body of evidence has shown that these restraint forms are differentially related to key outcomes, with rigid dietary control consistently, strongly, and positively related to various disordered eating behaviors and body image concerns and flexible dietary control being negatively related to these outcomes (Linardon, 2018a, Linardon, 2018b; Linardon & Mitchell, 2017; Smith, Williamson, Bray, & Ryan, 1999; Westenhoefer et al., 1999, 2013). This suggests that dietary restraint may be particularly problematic only when it is practiced inflexibly.

Existing measures of dietary restraint have only focused on assessing the behavioral methods (e.g., skipping meals for weight loss, avoiding so-called forbidden foods) used to restrict food intake. A rigid approach to dieting, however, also entails various psychological processes that govern eating behavior, including the extent to which one believes they must follow at all times a set of arbitrary diet-rules, the tendency to feel empowered when such rules are obeyed, and the perceived failure in self-control and emotional distress experienced when such rules have been “broken”. The importance of these psychological processes are highlighted in existing models of eating disorder psychopathology, where it is theorized that binge eating behavior results from the tendency for one to react negatively (e.g., feeling like a failure, lacking self-control etc.) after any transgression from a self-imposed diet rule (Fairburn, 2008). Thus, an instrument that captures these psychological processes of rigid dietary control would be valuable for further testing and validating these models of eating disorders, and for identifying which facets of dietary restraint may be particularly harmful.

Recognizing the importance of having an instrument that assesses these psychological processes of rigid dietary control, Duarte and colleagues recently developed the Inflexible Eating Questionnaire (IEQ; Duarte, Ferreira, Pinto-Gouveia, Trindade, & Martinho, 2017). The 11-item IEQ was developed to assess an individual's inflexible adherence to a set of self-imposed eating rules, along with the tendency for one to respectively feel empowered or distressed when such rules are or are not met. In a sample of 905 Portuguese adults, Duarte et al. (2017) found the IEQ to exhibit a unidimensional structure via a confirmatory factor analysis (CFA), and they also found evidence for internal consistency (α = .90), temporal stability (r = .84), and convergent validity (via IEQ's significant bivariate associations with other measures of eating behavior, including flexible control, rigid control, dietary intent, and intuitive eating). Further evidence for validity was found when participants with high versus low eating disorder psychopathology scores were differentiated by IEQ scores. Finally, IEQ scores moderated the relationship between dietary restraint and global eating disorder psychopathology. In particular, the relationship between dietary restraint and eating disorder psychopathology was strongest in those with high scores on the IEQ and was weakest in those with low IEQ scores. Taken together, initial findings suggest that the IEQ appears to be valid and reliable, and a useful measurement tool in eating behavior research.

At present, evidence for validity and reliability of the IEQ is limited to only one sample. Thus, replicating the IEQ's unidimensional structure and confirming its psychometric properties in a different sample and country, and by an independent research team, is needed before (a) making definitive conclusions about the psychometric performance of this measure and (b) recommending its use in research and clinical contexts. This study therefore aimed to confirm the factor structure and psychometric properties of the IEQ in a large sample of Australian adult women. We hypothesised that (a) the unidimensional structure of the IEQ would be upheld, (b) the IEQ would be internally consistent, (c) IEQ scores would be significantly correlated with rigid control, flexible control, dietary restraint, and intuitive eating (i.e., evidence for convergent validity), and (d) IEQ scores would explain a significant amount of unique variance in eating disorder psychopathology and psychosocial impairment over and above existing measures of eating restraint (flexible and rigid control) and intuitive eating (i.e., evidence for incremental validity).

Section snippets

Participants and procedure

One-thousand females participated in this study. The mean age was 29.43 years.

(SD = 8.20) and the average body mass index (BMI = kg/m2) was 26.80 (SD = 9.86). Most participants identified as Caucasian (78%). Other ethnicities reported were Asian (9%), Hispanic (7%), African (1%) and Aboriginal/Torres Strait (1%). The remaining participants did not provide their ethnicity. None of the participants self-reported an eating disorder diagnosis. We limited our sample to females because the sites of

Preliminary analyses

Six multivariate outliers were detected. Given these cases comprised less than 1% of our total sample, and that their inclusion versus exclusion had no impact on our results, we retained these cases in the analyses. None of the assumptions of the general linear model (i.e., linearity, no multicollinearity) appeared to be violated (Field, 2013).

Factor structure

Fit indices provided mixed support for a unidimensional model of the 11-item IEQ (χ2 [df = 55] = 4161.27, p < .001; CFI = .91, TLI = .89, RMSEA = .09,

Discussion

This study aimed to confirm the unidimensional structure and psychometric properties of the IEQ in a larger sample of Australian adult women. Several of our findings provide further support for the psychometric properties of the IEQ in a distinct sample of adult women. First, the IEQ exhibited a unidimensional structure, which was consistent with the structure observed in a sample of Portuguese participants (Duarte et al., 2017). Second, the IEQ exhibited good internal consistency, suggesting

References (24)

  • C.G. Fairburn

    Cognitive behavior therapy and eating disorders

    (2008)
  • C.G. Fairburn et al.

    Assessment of eating disorders: Interview or self-report questionnaire?

    International Journal of Eating Disorders

    (1994)
  • Cited by (0)

    View full text