**RESEARCH Open Access** # Psychological aspects of grazing in adolescents: Psychometric properties and measurement invariance of the Rep(eat)-Q in community and clinical samples Sílvia Félix1,2\*, Sofa Ramalho3 , João Marôco4 , Andreia Ribeiro5,6, Nadine Afonso¹ , Sónia Gonçalves1 and Eva Conceição2,5,6 ## **Abstract** **Background** Grazing is a disordered eating behavior associated with poor weight control, increased eating disorder psychopathology, and psychological difculties in adults. Unfortunately, little is known about grazing in adolescence, which is aggravated by the lack of validated measures for this population. This study investigates the psychological aspects of grazing in adolescence and provides psychometric support for a brief self-report measure, the Rep(eat)-Q. **Methods** A community sample of middle/high school students (n=358, 55.6% females) and a clinical sample with overweight/obesity (n=204, 59.8% females), completed a set of self-report questionnaires assessing eating and psychological variables including the Rep(eat)-Q. Weight and height data were also collected (Community: Mz-BMI=0.39, SD=0.98; Clinical: Mz-BMI=2.39, SD=0.74). **Results** Confrmatory factor analysis revealed an adequate ft [(χ2(107)=389.77; *p*<0.001; CFI=0.99; TLI=0.99; NFI=0.99; SRMR=0.062; RMSEA=0.098] for a second-order (grazing) model with two frst-order factors (repetitive eating and compulsive grazing subscales) with good reliability (0.85<α<0.91). Metric and scalar invariance were confrmed, allowing comparisons between samples. Compared with clinical adolescents, community adolescents reported higher scores on the Rep(eat)-Q total (Community: M=1.86, SD=1.30; Clinical: M=1.53, SD=1.35; *t*(559)= −2.81; *p*=0.005) and repetitive eating subscale (Community: M=2.04, SD=1.44; Clinical: M=1.52, SD=1.35; *t*(560)= −4.24; *p*≤0.001). The Rep(eat)-Q total score and subscales scores were signifcantly positively correlated with eating disorder psychopathology and inversely correlated with intuitive eating, suggesting good convergent validity (0.113 and kurtosis |ku|>7 [\[15](#page-11-26), [32\]](#page-11-27). A confrmatory factor analysis (CFA) was executed to assess the dimensional structure of Rep(eat)-Q. Te Diagonally Weighted Least Squares estimation method was applied. Goodness-of-ft was evaluated using the NFI (Normed Fit Index), TLI (Tucker Lewis Index), CFI (Comparative Fit Index), RMSEA (Root Mean Square Error of Approximation), χ2 (chi-square statistic), χ2/ df (chi-square statistic/degrees of freedom) and SRMR (Standardized Root Mean Square Residual). Criteria for a well-ftting model were defned as NFI, TLI, and CFI values>0.95, and SRMR and RMSEA values<0.08 [\[3](#page-10-2), [5](#page-11-28), [26](#page-11-29), [33\]](#page-11-30). Te reliability of the scores was evaluated using the internal consistency estimators α ordinal, with estimates≥0.80 indicating satisfactory internal consistency evidence. Te average variance extracted (AVE) was also calculated and values of AVE≥0.50 were considered acceptable indicators of convergent evidence in terms of internal structure [[32\]](#page-11-27). Te measurement invariance among samples (i.e., community sample of adolescents vs. clinical sample of adolescents with overweight/obesity undergoing hospital treatment) was compared in several nested models: (1) confgural invariance, (2) loadings (metric invariance), (3) thresholds (scalar invariance). Tresholds in detriment of intercepts were used to test scalar invariance because indicators were ordinal [\[44\]](#page-12-4). Invariance was confrmed if values of ΔCFI were≤0.010 and/or the Δχ2 achieves non-signifcant *p*-values. After confrming scalar invariance, Rep(eat)-Q secondorder factor and Repetitive Eating and Compulsive Grazing subscales were compared across samples via T-tests for independent samples. Te efect sizes were considered small, medium or large if Cohen's *d* were≥0.20, 0.50, or 0.80, respectively [[14\]](#page-11-31). Pearson's correlation coefcients were performed to test associations between the Rep(eat)-Q (sub)scale(s) and other eating-related (convergent evidence) and psychological variables. Cohen's efect size benchmarks were used to assess efect sizes classes for the correlations (i.e., small:≥0.10; medium:≥0.30; large:≥0.50) [[14\]](#page-11-31). Finally, to determine whether sociodemographic/anthropometric (i.e., sex, age, and BMI z-score) and psychological variables [i.e., negative urgency (UPPS), anxiety and depression symptoms (DASS-21)] were predictors of grazing [Rep(eat)-Q total score], a linear regression analyze was performed considering community and clinical samples together. Te efect size was considered small, medium or large if r 2 was≥0.01, 0.09, or 0.25, respectively [\[14](#page-11-31)]. #### **Results** ## **Samples' characterization** A total of 574 adolescents participated in the primary studies. Seven adolescents from the community sample and six from the clinical sample were excluded for not having responded to the Rep(eat)-Q, resulting in a fnal sample of 562 participants (n=358 for the community sample, n=204 for the clinical sample). Te community sample included 199 (55.6%) girls and 159 (44.4%) boys. Te mean age was 15.60 (SD=1.75) years; 275 (77.0%) adolescents were attending high school and 83 (23.0%) were attending middle school. Te mean BMI was 22.17 kg/m2 (SD=3.50), with a BMI for age z-score of 0.41 (SD=0.98). According to the BMI for age z-score, most of the sample had normal weight (69.8%, n=241) but 20.7% (n=74) had overweight and 5% (n=18) had obesity. About 49.2% of mothers and 30.4% of fathers completed high school or college education and the majority of the parents reported having a full-time job (87.4% for fathers and 83.8% for mothers). Te clinical sample included 122 (59.8%) girls and 82 (40.2%) boys. Te mean age was 15.04 (SD=1.63) years old; 51.5% (n=105) were attending middle school and 28.5% (n = 99) were attending high school. Te mean BMI was 31.12 kg/m2 (SD=6.01), with a BMI for age z-score of 2.39 (SD=0.74). Approximatly 27% of the participants from clinical sample were assessed at their frst medical appointment for weight loss treatment, whereas the remaining participants were already engaged in ongoing follow-up, with heterogeneous durations. Te mean follow-up duration was 787.36 days (SD=1178.54). Twenty-fve percent (n=51) of the sample had overweight and 75% (n=153) had obesity. Concerning parents' information, 28.2% of mothers and 22.6% of fathers completed high school or college education and most parents reported having a full-time job (81% for fathers and 69.7% for mothers). Table [1](#page-5-0) displays the means and stantard deviations for all the self-reported measures applied to the community and clinical samples. ## **Evidence of the psychometric properties of the Rep(eat)‑Q** Each of the 12 items of the Rep(eat)-Q presented responses in all categories of the ordinal scale (0–6). Te items did not present severe violations of univariate normality (i.e., |sk| and |ku| values below 3 and 7, respectively). Item 12 (i.e., "Snacked on food when you were anxious, bored or lonely, or under other emotions") had the highest mean value (M=2.09; SD=2.00) and item 8 (i.e., "Felt driven to eat") had the lowest mean value (M=1.34; SD=1.68). Table [2](#page-6-0) presents the items' distribution properties for both samples. **Table 1** Descriptive statistics of the self-reported measures assessed per sample | | | Community
sample | | Clinical sample | | | | |-------------------------------|-------|---------------------|-------|-----------------|--|--|--| | | M | SD | M | SD | | | | | Rep(eat)-Q total score | 1.86 | 1.30 | 1.53 | 1.35 | | | | | Rep(eat)-Q repetitive eating | 2.04 | 1.44 | 1.52 | 1.35 | | | | | Rep(eat)-Q compulsive grazing | 1.67 | 1.43 | 1.54 | 1.54 | | | | | EDE-Q-7 total score | 1.94 | 1.70 | – | – | | | | | EDE-Q-7 restraint | 1.51 | 1.81 | - | - | | | | | EDE-Q-7 SWO | 2.12 | 2.12 | – | – | | | | | EDE-Q-7 BD | 2.41 | 2.11 | – | – | | | | | ChEAT total score | – | – | 16.08 | 9.10 | | | | | ChEAT RPB | – | – | 2.66 | 3.13 | | | | | ChEAT FP | – | – | 2.99 | 2.41 | | | | | ChEAT FGF | – | – | 9.77 | 5.78 | | | | | ChEAT SPE | – | – | 0.30 | 0.95 | | | | | IES-2 total score | – | – | 3.31 | 0.46 | | | | | IES-2 EPER | – | – | 3.45 | 0.90 | | | | | IES-2 RHSC | – | – | 3.20 | 0.40 | | | | | IES-2 BFC | – | – | 3.47 | 0.79 | | | | | DASS-21 depression | 6.25 | 5.21 | 5.52 | 5.43 | | | | | DASS-21 anxiety | 5.78 | 5.22 | 4.31 | 4.47 | | | | | DASS-21 stress | – | – | 6.29 | 5.24 | | | | | UPPS—NU | 2.43 | 0.57 | 2.51 | 0.60 | | | | | DERS total score | 92.01 | 24.26 | – | – | | | | | DERS strategies | 19.01 | 7.49 | – | – | | | | | DERS non-acceptance | 13.72 | 6.11 | – | – | | | | | DERS awareness | 16.97 | 5.22 | – | – | | | | | DERS impulsivity | 14.60 | 5.83 | – | – | | | | | DERS goals | 15.13 | 4.92 | – | – | | | | | DERS clarity | 12.58 | 4.45 | – | – | | | | | FSCRS inadequate self | 14.91 | 8.91 | – | – | | | | | FSCRS hated self | 4.08 | 4.61 | – | – | | | | | FSCRS reassured self | 19.06 | 7.48 | – | – | | | | | TEXI working memory | 2.66 | 0.78 | – | – | | | | | TEXI inhibition | 2.86 | 0.70 | – | – | | | | | PedsQL total score | – | – | 77.32 | 15.31 | | | | | PedsQL psycho health | – | – | 75.18 | 15.58 | | | | | PedsQL physical health | – | – | 81.33 | 16.57 | | | | nCommunity sample=357; nClinical sample=204 *M* Mean; *SD* Standard deviation; *Rep(eat)-Q* Repetitive eating questionnaire; *EDE-Q-7* Eating disorder examination questionnaire 7-item form; *SWO* Shape/weight overvaluation; *BD* Body dissatisfaction; *TEXI* Teenage executive functioning inventory; *DERS* Difculties in emotion regulation scale; UPPS-NU Urgency, premeditation, perseverance, sensation seeking scale—negative urgency subscale; *FSCRS* The forms of self-criticizing/attacking and self-reassuring scale; *DASS* Depression, anxiety, and stress scales; *ChEAT* Children's eating attitudes test; *IES* Intuitive eating scale; *RPB* Restrictive and purgative behaviors; *FP* Food preoccupation; *FGF* Fear of getting fat; *SPE* Social pressure to eat; *EPER* Eating for physical rather than emotional reasons; *RHSC* Reliance on hunger an satiety cues; *BFC* Body-food congruence; *PedsQL* Pediatric quality of life inventory; *Psycho Health* Psychosocial health #### *Validity evidence based on internal structure* Confrmatory factor analysis was used to test a secondorder structure with two factors for the entire sample (considering clinical vs. community sample as the multigroup analysis factor). Tis second-order structure was chosen because the existence of a large amount of shared variance between the factors (i.e., repetitive eating and compulsive grazing) was previously shown to be better explained by the existence of a second-order factor (i.e. grazing) [\[40](#page-11-2)]. An adequate ft to the data was found (χ2(107)=389.77, *p*<0.001; χ2 /df=3.64; CFI=0.99; TLI=0.99; NFI=0.99; SRMR=0.062; RMSEA=0.098; P(RMSEA≤0.05)≤0.001; 95% CI [0.087; 0.108]). #### *Evidence of construct validity and reliability* Both frst-order factors showed acceptable evidence of internal consistency (compulsive grazing: α=0.91; AVE=0.69, and repetitive eating: α=0.85; AVE=0.58). Te second-order factor also presented good internal consistency evidence (Rep(eat)-Q total: α=0.86; AVE=0.58). ## **Measurement invariance and comparison between community and clinical samples** Figure [1](#page-6-1) depicts the loadings for each sample. Table [3](#page-6-2) presents the measurement invariance tests. Te secondorder model showed adequate ft indices, supporting confgural invariance and indicating that the model fts both community and clinical samples. According to the criteria of Δχ2 and/or ΔCFI used to determine the ft of nested models, there is support for metric and scalar invariance. Metric invariance indicates that factor loadings and factor structure are identical across samples, and scalar invariance posits that thresholds are similar across groups, which allows the instrument scores to be compared among community and clinical samples. Community adolescents scored signifcantly higher than clinical adolescents in the Rep(eat)-Q total score [*t* (559)= −2.81; *p*=0.005; Cohen's *d*= −0.25] and in the Repetitive Eating Subscale [*t* (560)= −4.24; *p*≤0.001, Cohen's *d*= −0.37], but the efect sizes were small. No diferences were found between the samples for Compulsive Eating subscale [*t*(560)= −1.01; *p*=0.312, Cohen's *d*= −0.09]. # **Association with psychological/behavioral variables: convergent validity** Table [4](#page-7-0) shows the correlations between the Rep(eat)-Q and the other variables under study for community and clinical samples. **Table 2** Items' distributional properties | Item | | Combined samples | | | | | | | | Community sample | | | | | | Clinical sample | | | | | | |---------|------|------------------|-----|-----|-----|------|-------|------|------|------------------|-----|-----|------|--------|------|-----------------|-----|-----|-----|------|-------| | | M | SD | Min | Mdn | Max | sk | ku | M | SD | Min | Mdn | Max | sk | ku | M | SD | Min | Mdn | Max | sk | ku | | Item 1 | 1.95 | 1.99 | 0 | 1 | 6 | 0.82 | −0.60 | 2.27 | 2.03 | 0 | 2 | 6 | 0.56 | −0.99 | 1.41 | 1.80 | 0 | 1 | 6 | 1.45 | 1.09 | | Item 2 | 1.87 | 1.79 | 0 | 1 | 6 | 0.84 | −0.26 | 2.05 | 1.84 | 0 | 2 | 6 | 0.62 | −0.67 | 1.57 | 1.67 | 0 | 1 | 6 | 1.30 | 0.34 | | Item 3 | 1.90 | 1.87 | 0 | 1 | 6 | 0.80 | −0.48 | 2.02 | 1.81 | 0 | 2 | 6 | 0.58 | −0.74 | 1.67 | 1.96 | 0 | 1 | 6 | 1.17 | 0.15 | | Item 4 | 1.80 | 1.88 | 0 | 1 | 6 | .89 | −0.36 | 2.01 | 1.85 | 0 | 2 | 6 | 0.64 | −0.75 | 1.44 | 1.88 | 0 | 1 | 6 | 1.44 | 0.93 | | Item 5 | 1.54 | 1.87 | 0 | 1 | 6 | 1.07 | −0.03 | 1.50 | 1.83 | 0 | 1 | 6 | 1.01 | −0.19 | 1.61 | 1.94 | 0 | 1 | 6 | 1.15 | .14 | | Item 6 | 1.40 | 1.71 | 0 | 1 | 6 | 1.25 | 0.62 | 1.42 | 1.71 | 0 | 1 | 6 | 1.17 | 0.39 | 1.36 | 1.72 | 0 | 1 | 6 | 1.39 | 1.09 | | Item 7 | 1.69 | 1.78 | 0 | 1 | 6 | 1.01 | −0.00 | 1.75 | 1.76 | 0 | 1 | 6 | 0.89 | −0.29 | 1.60 | 1.82 | 0 | 1 | 6 | 1.23 | 0.55 | | Item 8 | 1.34 | 1.68 | 0 | 1 | 6 | 1.23 | 0.64 | 1.41 | 1.69 | 0 | 1 | 6 | 1.08 | 0.19 | 1.22 | 1.65 | 0 | 1 | 6 | 1.54 | 1.67 | | Item 9 | 1.76 | 1.85 | 0 | 1 | 6 | 0.89 | −0.31 | 1.85 | 1.90 | 0 | 1 | 6 | 0.76 | −0.56 | 1.59 | 1.77 | 0 | 1 | 6 | 1.14 | 0.32 | | Item 10 | 1.85 | 1.77 | 0 | 1 | 6 | 0.82 | −0.26 | 2.08 | 1.84 | 0 | 2 | 6 | 0.57 | −0.70 | 1.44 | 1.57 | 0 | 1 | 6 | 1.39 | 1.50 | | Item 11 | 1.70 | 2.06 | 0 | 1 | 6 | 1.00 | −0.40 | 1.58 | 2.00 | 0 | 1 | 6 | 1.01 | −0.38 | 1.90 | 2.13 | 0 | 1 | 6 | 0.96 | −0.50 | | Item 12 | 2.09 | 2.00 | 0 | 2 | 6 | 0.63 | −0.85 | 2.39 | 1.99 | 0 | 2 | 6 | 0.38 | −01.08 | 1.55 | 1.91 | 0 | 1 | 6 | 1.21 | 0.30 | *N* 561; *M* Mean; *SD* Standard deviation; *Min* Minimum; *Mdn* Median; *Max* Maximum; *sk* Skewness; *k* Kurtosis ![](https://assets.reachrx.ai/images/10.1186%E2%98%8Ds40337-025-01344-5/_page_6_Figure_6.jpeg) **Fig. 1** Factor loadings of the second-order model for the community and clinical sample **Table 3** Tests of measurement invariance of Rep(eat)-Q for community and clinical samples | Model invariance | χ2 | df | CFI | Δχ2 | p(Δχ2) | ΔCFI | |------------------|---------|-----|-------|---------|--------|-------| | Confgural | 389.772 | 107 | 0.990 | – | – | – | | Metric | 401.494 | 116 | 0.990 | 11.722 | 0.229 | 0.000 | | Scalar | 605.129 | 173 | 0.985 | 203.635 | ≤.001 | 0.005 | *N* 561; *χ2* Chi-square statistic; *df* Degrees of freedom; *CFI* Comparative Fit Index **Table 4** Pearson's correlations between the Rep(eat)-Q and sociodemographic variables, BMI z-score, and eating and psychological variables | | Rep(eat)-Q total score | Repetitive eating subscale | Compulsive
grazing subscale | |------------------------|------------------------|----------------------------|--------------------------------| | Community sample | | | | | Age | 0.121* | 0.109* | 0.106* | | z-score BMI | 0.078 | −0.052 | 0.196*** | | EDE-Q-7 total score | 0.286*** | 0.107* | 0.411*** | | EDE-Q-7 restraint | 0.173*** | 0.001 | 0.313*** | | EDE-Q-7 SWO | 0.332*** | 0.196*** | 0.409*** | | EDE-Q-7 BD | 0.252*** | 0.110* | 0.349*** | | TEXI working memory | 0.379*** | 0.305*** | 0.383*** | | TEXI inhibition | 0.443*** | 0.355*** | 0.450*** | | DERS total score | 0.329*** | 0.226*** | 0.370*** | | DERS strategies | 0.360*** | 0.258*** | 0.394*** | | DERS non-acceptance | 0.299*** | 0.217*** | 0.324*** | | DERS awareness | −0.071 | −0.089 | −0.040 | | DERS impulsivity | 0.232*** | 0.164** | 0.257*** | | DERS goals | 0.345*** | 0.261*** | 0.367*** | | DERS clarity | 0.148** | 0.075 | 0.195*** | | UPPS—NU | 0.428*** | 0.326*** | 0.453*** | | FSCRS inadequate self | 0.370*** | 0.253*** | 0.420*** | | FSCRS hated self | 0.346*** | 0.227*** | 0.402*** | | FSCRS reassured self | −0.154*** | −0.098 | −0.181*** | | DASS-21 depression | 0.378*** | 0.281*** | 0.406*** | | DASS-21 anxiety | 0.369*** | 0.272*** | 0.399*** | | Clinical sample | | | | | Age | 0.061 | 0.095 | 0.013 | | z-score BMI | 0.063 | 0.058 | 0.059 | | ChEAT total score | 0.330*** | 0.199** | 0.401*** | | ChEAT RPB | −0.157* | −0.200** | −0.099 | | ChEAT FP | 0.605*** | 0.485*** | 0.633*** | | ChEAT FGF | 0.313*** | 0.184* | 0.30.85*** | | ChEAT SPE | 0.108 | 0.135 | 0.071 | | IES-2 total score | −0.313*** | −0.292*** | −0.291*** | | IES-2 EPER | −0.441*** | −0.390*** | −0.429*** | | IES-2 RHSC | 0.041 | 0.043 | 0.034 | | IES-2 BFC | −0.141* | −0.181* | −0.088 | | PedsQL total score | −0.461*** | −0.350*** | −0.499*** | | PedsQL psycho health | −0.447*** | −0.329*** | −0.494*** | | PedsQL physical health | −0.386*** | −0.314** | −0.400*** | | UPPS—NU | 0.413*** | 0.366*** | 0.400*** | | DASS-21 depression | 0.381*** | 0.265*** | 0.434*** | | DASS-21 anxiety | 0.337*** | 0.227* | 0.392*** | | DASS-21 stress | 0.327*** | 0.217** | 0.383*** | nCommunity Sample=357; nClinical Sample=204 Sex (1=Female; 2=Male); *BMI* Body mass index; *EDE-Q-7* Eating disorder examination questionnaire 7-item form; *SWO* Shape/weight overvaluation; *BD* Body dissatisfaction; *TEXI* Teenage executive functioning inventory; *DERS* Difculties in emotion regulation scale; *UPPS-NU* Urgency, premeditation, perseverance, sensation seeking scale—negative urgency subscale; *FSCRS* The forms of self-criticizing/attacking and self-reassuring scale; *DASS* Depression, anxiety, and stress scales; *ChEAT* Children's eating attitudes test; *IES* Intuitive eating scale; *RPB* Restrictive and purgative behaviors; *FP* Food preoccupation; *FGF* Fear of getting fat; *SPE* Social pressure to eat; *EPER* Eating for physical rather than emotional reasons; *RHSC* Reliance on hunger an satiety cues; *BFC* Body-food congruence; *PedsQL* Pediatric quality of life inventory; *Psycho Health* Psychosocial health Correlation with medium or large efect size were highlighted in bold ^\* *p*<0.05; \*\**p*<0.005; \*\*\**p*≤0.001 Concerning associations with dimentions of eating behavior, as expected, the Rep(eat)-Q total score and its subscales mostly presented signifcant small to moderate correlations with eating disorder psychopathology in the community (EDE-Q-7) and clinical (ChEAT) samples and with intuitive eating (IES-2) in the clinical sample, showing evidence of good convergent validity. Tere were also signifcant correlations with cognitive/behavioral measures for both community and clinical samples. In community samples, higher values of the Rep(eat)-Q total score and subscales, were associated (medium efect sizes) with greater difculties in working memory, inhibition (TEXI), and negative urgency (UPPS). Similarly, the Rep(eat)-Q total score and compulsive grazing were positively correlated with self-criticism (hated self and inadequate self subscales of the FSCRS). Emotion regulation was only assessed in the community sample. Higher values of the Rep(eat)-Q total score and its subscales were associated (small-to-medium efect sizes) with greater difculties in DERS total score and its subscales of strategies, non-acceptance, goals, impulsivity and clarity (except for repetitive eating). Regarding psychological distress, in both the community and clinical samples, higher values of the Rep(eat)- Q total score and its subscales were associated (with small-to-medium efect sizes) with depression, anxiety, and stress symptoms (DASS-21). Tere were also positive medium efect size correlations between Rep(eat)-Q total score and the health-related quality of life (PedsQL) scores for the clinical sample. # **Grazing in adolescents: the role of sociodemographic/ anthropometric, and psychological variables** To investigate the role of sociodemographic, anthropometric and psychological variables in explaining grazing eating behavior the variables collected for the entire sample (i.e., combining community and clinical samples) were tested together as indicators of grazing (total sample size N=574). Sex, age, and BMI z-score were entered in Step 1, followed by anxiety (DASS-21), depression (DASS-21), and negative urgency (UPPS) in Step 2. Te results are displayed in Table [5.](#page-8-0) Te frst block, which included the sociodemographic and anthropometric variables sex, age, and BMI z-score, was not statistically signifcant [F(3, 493)=2.22, *p*=0.085, R2=0.013; small efect size]. When negative urgency, depression and anxiety symptoms were added to Block 2, the overall model was signifcant [F(6, 490)=22.87, *p*≤0.001)] and explained 21.8% of the variance in grazing (R2=0.218; medium efect size). Tis represents a statistically signifcant improvement in the explained variance [ΔR²=0.025; F(3, 490)=*p*≤0.001]. Only the Block 2 variables were statisticaly signifcant indicators (i.e., depression, anxiety, and negative urgency). Interestingly, negative urgency was the strongest indicator, and for each unit increase in negative urgency symptoms, the grazing scores increased by 0.688 (β=0.305). Age, sex and BMI z-score were not signifcant in the model. ## **Discussion** Tis is the frst study to investigate the psychometric properties of a measure to assess grazing eating behavior, the Rep(eat)-Q, in adolescents. Overall, the fndings indicate that the Rep(eat)-Q presents adequate psychometric **Table 5** Linear regression analysis to test sociodemographic/anthropometric and psychological variables as predictors of grazing [Rep(eat)-Q total score)] for the complete sample (i.e., combining community and clinical samples) | B | SE | β | t | 95% CI | R2 | F | |--------|-------|--------|---------|-----------------|-------|----------| | | | | | | | | | −0.140 | 0.117 | −0.054 | −1.20 | [−0.141; 0.035] | 0.013 | 2.22 | | 0.071 | 0.035 | 0.092 | 2.02* | [0.003 0.181] | | | | −0.025 | 0.045 | −0.025 | −0.055 | [−0.115; .065] | | | | | | | | | | | | 0.188 | 0.108 | 0.071 | 1.73 | [−0.010; 0.154] | 0.212 | 22.87*** | | 0.032 | 0.032 | 0.041 | 1.00 | [−0.040; 0.123] | | | | −0.063 | 0.040 | −0.029 | −1.56 | [−0.145; 0.017] | | | | 0.038 | 0.016 | 0.146 | 2.44* | [0.029; 0.263] | | | | 0.030 | 0.015 | 0.125 | 2.02* | [0.003; 0.246] | | | | 0.688 | 0.104 | 0.305 | 6.64*** | [0.214; 0.395] | | | | | | | | | | | *N* 504; Sex (1=Female; 2=Male); *BMI* Body mass index; *DASS* Depression, anxiety, and stress scales; *UPPS-NU* Urgency, premeditation, perseverance, sensation seeking scale—negative urgency subscale ^\* *p*<0.05; \*\* *p*<0.005; \*\*\**p*≤0.001 properties in both community and clinical samples of adolescents with overweight/obesity. Specifcally, the instrument showed good validity evidence based on internal structure, construct validity, reliability, and its associations with other variables. ## **Factor structure and scores for community and clinical samples** Our fndings support a model with two frst-order factors (repetitive eating and compulsive grazing), and a secondorder latent factor of grazing as previously proposed by Teodoro and colleagues [[40\]](#page-11-2). Both the frst and secondorder factors showed good reliability estimates and all the original items were retained. Furthermore, the analysis confrmed measurement invariance across community and clinical adolescents, allowing comparisons of the Rep(eat) scores across these samples. Unexpectedly, the mean scores on the Rep(eat)-Q total (Community: M=1.86, SD=1.30; Clinical: M=1.53, SD=1.35) and repetitive eating subscale (Community: M=2.04, SD=1.44; Clinical: M=1.52, SD=1.35) were higher in the community sample than in the clinical sample. Adolescents in the clinical sample were undergoing hospital treatment for overweight/obesity, which primarily involved following a prescribed nutritional plan. Tis might have helped to reduce grazing by contributing to an increased awareness of undesirable behaviors. However, there were no signifcant diferences in the scores of the compulsive grazing subscale between these two samples (Community: M=1.67, SD=1.43; Clinical: M=1.54, SD=1.54). Tis outcome is likely due to compulsive grazing involving an increased degree of loss of control over eating, which makes this behavior more severe and more resistant to change [[18\]](#page-11-13), even for individuals undergoing medical or nutritional interventions. ## **Associations between grazing and psychological factors** Te Rep(eat)-Q total score and its subscales were signifcantly associated with the other eating-related measures included in the study. Specifcally, positive correlations were found with eating disorder psychopathology in both community and clinical samples, and negative correlations were found with intuitive eating in the clinical sample. Tese fndings support convergent validity, further substantiating grazing as a construct of disordered eating behavior. Signifcant correlations were also found with other psychological measures in both the community and the clinical samples. Consistent with previous studies [[8,](#page-11-1) [23,](#page-11-9) [24](#page-11-10), [37](#page-11-32)], this study revelead that grazing was associated with higher levels of depression, anxiety, and stress symptoms,negative urgency; and difculties in emotion regulation, as well as with lower quality of life in adolescents. Notably, our fndings introduce further novelty to the concept of grazing by exploring its correlation with new variables such as executive functions and self-criticism (assessed in the community sample). On the one hand, the existence of strong positive correlations between grazing and executive functions (i.e., working memory and inhibition) adds to the existing evidence that topdown mechanisms play a role in the regulation of eating behaviors [\[11](#page-11-33)]. On the other hand, the strong correlations between Rep(eat)-Q total and compulsive grazing with self-criticism (i.e., inadequate-self and hated-self subscales) align with previous evidence on binge eating (i.e., an eating behavior highly characterized by loss of control eating) [[12,](#page-11-34) [13\]](#page-11-35). Tis evidence strengthens the understanding that disordered eating behaviors are related to psychological, cognitive and emotional difculties. Overall, the compulsive grazing subscale exhibited stronger correlations with other eating behaviors and psychological variables than did the repetitive eating subscale in both samples. Te model proposed by Conceição and colleagues (2014) places various problematic eating behaviors on a continuum scale of loss of control over eating. According to this model, compulsive grazing is conceptualized as being higher in the spectrum of loss of control eating and eating-related psychopathology than grazing without loss of control, which has been supported by other subsequent studies [[9,](#page-11-12) [23\]](#page-11-9). Tus, our fndings bring further support for this conceptualization of grazing in a spectrum of disordered eating and extend the current literature to the developmental age of adolescents. ## **Importance of sociodemographic and antrophometric variables** Finally, the linear regression analysis emphasized the importance of psychological variables over sociodemographic and anthropometric characteristics in understanding grazing in adolescents. Specifcally, depression and anxiety symptoms, as well as negative urgency, were signifcantly associated with grazing, whereas sex, age, and BMI z-score were not. Among them, negative urgency was the strongest indicator, suggesting that adolescents scoring higher in grazing have a greater tendency to act impulsively in response to negative emotions. Tis fnding is consistent with prior studies highlighting the central role of impulsive traits and loss of control in disordered eating behaviors [[4](#page-11-36), [9](#page-11-12), [43\]](#page-12-5), and underscores the importance of addressing emotional aspects and impulsive traits in interventions aimed at managing grazing. Tis may be particularly relevant in adolescence because this age group is particularly vulnerable to developing disordered eating behaviors and weight problems [\[17,](#page-11-7) [22,](#page-11-8) [46](#page-12-0)]. #### **Limitations** One limitation of our study is that, although the overall model ft was acceptable, the RMSEA value was close to the upper threshold, which warrants caution in interpreting the results. Te temporal stability of the Rep(eat)-Q was not assessed in this study, despite our expectation of good test–retest reliability based on the original study. Additionally, the use of self-reported measures did not address social desirability bias, potentially biasing (by underreporting) the responses of patients from the clinical sample. Te use of diferent measures across the two samples limits direct comparability. Also, the clinical sample was assessed at diferent stages of intervention, which may have infuenced their psychological and behavioral profles. Finally, approximately 23% of community sample participants fulflled the questionnaires in paper–pencil format due to difculties with online access, which compromised the uniformity of the procedures, but increased the validity across diferent response systems. ## **Conclusion** Te present study supports the use of the Rep(eat)-Q as a self-report measure with adequate psychometric properties to assess grazing in both community and clinical samples of adolescents. Our fndings support a model with two frst-order factors (repetitive eating and compulsive grazing), and a second-order latent factor of grazing. Te scales possessed validity evidence based on internal structure, good reliability, and construct validity. Tis work also provides further evidence that grazing is a problematic eating behavior with clinical relevance among adolescents because of its association with other eating-related and psychological variables, which are independent of age, gender and BMI for this age group. Tis is also the frst study to show that grazing is positively correlated with poor executive functioning and increased self-criticism, strengthening the role of cognitive and emotional variables in the understanding of disordered eating behaviors. Te use of the Rep(eat)- Q in this population can help to screen grazing patterns and explore underlying mechanisms related to the development and maintenance of eating disorders and overweight/obesity. Future research should employ longitudinal designs to explore the prospective validity of grazing as a risk factor for the development of other disordered eating behaviors, psychological difculties and/ or weight fuctuations over time. ## **Author contributions** Authors SF, SR, and EC designed the study. Authors SF, SR, and NA collected the data. Authors SF, JM, and AR conducted the statistical analysis. Author SF wrote the frst draft of the manuscript. Authors SF, SR, JM, AR, NA, SG, and EC participated in the interpretation of fndings and revised/edited the manuscript. All authors have approved the fnal manuscript. #### **Funding** This study was conducted at the Psychology Research Centre (CIPsi/UM) School of Psychology, University of Minho, supported by the Foundation for Science and Technology (FCT) through the Portuguese State Budget (UIDB/01662/2020), and by a grant to Sílvia Félix (2020.07384.BD). This work was also supported by FCT by project reference UID/04375/Centro de Investigação em Psicologia para o Desenvolvimento and DOI identifer 10.54499/ UIDB/04375/2020 [\(https://doi.org/10.54499/UIDB/04375/2020](https://doi.org/10.54499/UIDB/04375/2020)) - Sofa Ramalho. #### **Data availability** The data that support the fndings of this study are available from the corresponding author upon reasonable request. ## **Declarations** #### **Conflict of interest** The authors declare no competing interests. #### **Consent to participate** The present study included adolescents from two independent broader studies (one with a community sample and the other with a clinical sample). Both studies were approved by the University of Minho ethical committee [Community sample: CEICSH 059/2021; Clinical sample: SECVS142/2015], and the clinical study was also approved by the hospital ethical committees [São João Hospital Center (CHSJ): 2015.192(164-DEFI/153-ES; Oporto Hospital Center— Northern Maternal and Child Center (CMIN): 19/06/2015]. All participants and their parents signed an informed consent document prior to data collection. #### **Ethical approval** All participants (and their parent/legal guardian) gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the ethical committee for research in social and human sciences from University of Minho, hospital ethical committees (ethical committee for health from São João University Hospital Center, and ethical committee from Santo Antonio University Hospital Center), and pedagogic councils of the involved school. #### **Author details** Psychotherapy and Psychopathology Research Lab, Psychology Research Centre, School of Psychology, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal. 2 Faculty of Nutrition and Food Sciences of the University of Porto, Porto, Portugal. 3 Psychology for Development Research Center (CIPD), University of Lusíada, Porto, Portugal. 4 Intrepid Lab & CETRAD. ECEO., Lusófona University, Lisbon, Portugal. 5 Faculty of Psychology and Education Sciences, University of Porto, Porto, Portugal. 6 Center for Psychology at University of Porto, Porto, Portugal. Received: 28 January 2025 Accepted: 13 July 2025 #### **References** - 1. Anafarta-Şendağ M, Tığrak A, Özbek-Şimşek D. Conceptualization of grazing: the psychometric properties of the repetitive eating questionnaire (Rep[eat]-Q) Turkish form. Turk J Psychiatry. 2023. [https://doi.org/10.5080/](https://doi.org/10.5080/u27276) [u27276](https://doi.org/10.5080/u27276). - 2. Bonder R, Davis C, Kuk JL, Loxton NJ. Compulsive " grazing " and addictive tendencies towards food. Eat Weight Disord Stud Anorex Bulim Obes. 2018.. - 3. Boomsma A. Reporting analyses of covariance structures. Struct Equ Modeling Multidiscip J. 2000;7(3):461–83. [https://doi.org/10.1207/S1532](https://doi.org/10.1207/S15328007SEM0703_6) [8007SEM0703\\_6](https://doi.org/10.1207/S15328007SEM0703_6). - 4. Bruzas MB, Tronieri JS, Chao AM, Jones E, McAllister C, Gruber K, McCuen-Wurst C, Berkowitz RI, Wadden TA, Allison KC. Binge size and loss of control as correlates of eating behavior and psychopathology among individuals with binge eating disorder and higher weight. J Behav Med. 2022;45(4):603–12. . - 5. Byrne BM. Structural equation modeling with AMOS. England: Psychology Press; 2013. - 6. Castilho P, Pinto-gouveia J, Duarte J. Exploring self-criticism: confrmatory factor analysis of the FSCRS in clinical and nonclinical samples. Clin Psychol Psychother. 2013;22:153–64. [https://doi.org/10.1002/cpp.1881.](https://doi.org/10.1002/cpp.1881) - 7. Conceição EM, Mitchell JE, Engel SG, Machado PPP, Lancaster K, Wonderlich SA. What is "grazing"? Reviewing its defnition, frequency, clinical characteristics, and impact on bariatric surgery outcomes, and proposing a standardized defnition. Surg Obes Relat Dis. 2014;10(5):973–82. [https://](https://doi.org/10.1016/j.soard.2014.05.002) [doi.org/10.1016/j.soard.2014.05.002](https://doi.org/10.1016/j.soard.2014.05.002). - 8. Conceição EM, Mitchell JE, Machado PPP, Vaz AR, Pinto-Bastos A, Ramalho S, Brandão I, Simões JB, de Lourdes M, Freitas AC. Repetitive eating questionnaire [Rep(eat)-Q]: enlightening the concept of grazing and psychometric properties in a Portuguese sample. Appetite. 2017;117:351–8. [https://doi.org/10.1016/j.appet.2017.07.012.](https://doi.org/10.1016/j.appet.2017.07.012) - 9. Conceição EM, Pinto-bastos A, de Lourdes M, Vaz AR, Brandão I, Ramalho S. Problematic eating behaviors and psychopathology in patients undergoing bariatric surgery : the mediating role of loss of control eating. Int J Eat Disord. 2018;51(6):507–17. [https://doi.org/10.](https://doi.org/10.1002/eat.22862) [1002/eat.22862](https://doi.org/10.1002/eat.22862). - 10. Coutinho J, Dias PC, Ribeiro E. The Portuguese version of the difculties in emotion regulation scale and its relationship with psychopathological symptoms Versão portuguesa da Escala de Difculdades de Regulação Emocional e sua relação com sintomas psicopatológicos. Rev Psiquiatr Clin. 2009;37(4):145–51. - 11. Dohle S, Diel K, Hofmann W. Executive functions and the self-regulation of eating behavior: a review. Appetite. 2018;124:4–9. [https://doi.](https://doi.org/10.1016/j.appet.2017.05.041) [org/10.1016/j.appet.2017.05.041](https://doi.org/10.1016/j.appet.2017.05.041). - 12. Duarte C, Pinto-Gouveia J, Ferreira C. Escaping from body image shame and harsh self-criticism: exploration of underlying mechanisms of binge eating. Eat Behav. 2014;15(4):638–43. [https://doi.org/10.](https://doi.org/10.1016/j.eatbeh.2014.08.025) [1016/j.eatbeh.2014.08.025](https://doi.org/10.1016/j.eatbeh.2014.08.025). - 13. Duarte C, Stubbs JR, Gilbert P, Stalker C, Catarino F, Basran J, Horgan G, Morris L. The weight-focused forms of self-criticising/attacking and self-reassuring scale: confrmatory factor analysis and associations with control, loss of control of eating and weight in overweight and obese women. Psychol Psychother Theory Res Pract. 2019;92(4):539–53. [https://doi.org/10.1111/papt.12196.](https://doi.org/10.1111/papt.12196) - 14. Ellis PD. The essential guide to efect sizes. Cambridge: Cambridge University Press; 2010. - 15. Finney SJ, DiStefano C. Non-normal and categorical data instructural equation modeling. In: Hancock GR, Mueller RO, editors. Structural equation modeling: a second course. 2nd ed. Charlotte: Information Age Publishing; 2013. p. 439–92. - 16. Gilbert P, Clarke M, Hempel S, Miles JNV, Irons C. Criticizing and reassuring oneself: an exploration of forms, styles and reasons in female students. Br J Clin Psychol. 2004;43:31–50. - 17. Golden NH, Schneider M, Wood C, Daniels S, Abrams S, Corkins M, De Ferranti S, Magge SN, Schwarzenberg S, Critch J, Hubbard V, Scanlon K, Soto V, Burrowes D, Haro T, Braverman PK, Adelman W, Alderman EM, Breuner CC, Lindros J. Preventing obesity and eating disorders in adolescents. Pediatrics. 2016. [https://doi.org/10.1542/peds.2016-1649.](https://doi.org/10.1542/peds.2016-1649) - 18. Goodpaster KPS, Marek RJ, Lavery ME, Ashton K, Rish JM, Heinberg LJ. Graze eating among bariatric surgery candidates: prevalence and psychosocial correlates. Surg Obes Relat Dis. 2016;12(5):1091–7. [https://](https://doi.org/10.1016/j.soard.2016.01.006) [doi.org/10.1016/j.soard.2016.01.006](https://doi.org/10.1016/j.soard.2016.01.006). - 19. Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difculties in emotion regulation scale. J Psychopathol Behav Assess. 2004;30(4):41–54. - 20. Grilo CM, Reas DL, Hopwood CJ, Crosby RD. Factor structure and construct validity of the eating disorder examination-questionnaire in college students: further support for a modifed brief version. Int J Eat Disord. 2015;48(3):284–9. [https://doi.org/10.1002/eat.22358.](https://doi.org/10.1002/eat.22358) - 21. Harrison AN, James Bateman CCB, Younger-Coleman NOM, Williams MC, Rocke KD, Clato-Day Scarlett SC, Chang SM. Disordered eating - behaviours and attitudes among adolescents in a middle-income country. Eat Weight Disord. 2020;25(6):1727–37. [https://doi.org/10.](https://doi.org/10.1007/s40519-019-00814-5) [1007/s40519-019-00814-5](https://doi.org/10.1007/s40519-019-00814-5). - 22. Hayes JF, Fitzsimmons-Craft EE, Karam AM, Jakubiak J, Brown ML, Wilfey DE. Disordered eating attitudes and behaviors in youth with overweight and obesity: implications for treatment. Curr Obes Rep. 2018;7(3):235–46.. - 23. Heriseanu AI, Hay P, Corbit L, Touyz S. Grazing in adults with obesity and eating disorders: a systematic review of associated clinical features and meta-analysis of prevalence. Clin Psychol Rev. 2017;58:16–32. [https://doi.org/10.1016/j.cpr.2017.09.004.](https://doi.org/10.1016/j.cpr.2017.09.004) - 24. Heriseanu AI, Hay P, Touyz S. Grazing behaviour and associations with obesity, eating disorders, and health-related quality of life in the Australian population. Appetite. 2019;143:1–9. [https://doi.org/10.1016/j.](https://doi.org/10.1016/j.appet.2019.104396) [appet.2019.104396](https://doi.org/10.1016/j.appet.2019.104396). - 25. Hongxia H, Hui L, Qin X, Hanfei Z, Ningli Y. Reliability and validity of the Chinese version of repetitive eating questionnaire for patients undergoing bariatric surgery. Chin Gen Pract. 2023;26(07):810–5. - 26. Hoyle RH. Structural equation modeling: concepts, issues and applications. London: SAGE Publications; 1995. - 27. Lima L, Guerra MP, Lemos M. Adaptação da escala genérica do Inventário Pediátrico de Qualidade de Vida - Pediatric Quality of Life Inventory 4. 0 - PedsQL, a uma população portuguesa. Rev Port Saude Publica. 2009;8:83–96. - 28. Lovibond SH, Lovibond PF. Manual for the depression, anxiety, stress scales. 2nd ed. England: Psychology Foundation; 1995. - 29. Machado PPP, Grilo CM, Crosby RD. Replication of a modifed factor structure for the Eating Disorder Examination-Questionnaire: extension to clinical eating disorder and non-clinical samples in Portugal. Eur Eat Disord Rev. 2018;26(1):75–80. . - 30. Maloney MJ, McGuire JB, Daniels SR. Reliability testing of a children's version of the eating attitude test. J Am Acad Child Adolesc Psychiatry. 1988;27(5):541–3.. - 31. Mantzios M, Egan H, Bahia H, Hussain M, Keyte R. How does grazing relate to body mass index, self-compassion, mindfulness and mindful eating in a student population? Health Psychol Open. 2018;5:1–7. . - 32. Marôco J (2021) Análise de equações estruturais: fundamentos teóricos, software & aplicações. 3rd ed. ReportNumber. - 33. McDonald RP, Ho M-HR. Principles and practice in reporting structural equation analyses. Psychol Methods. 2002;7(1):64–82. [https://doi.org/](https://doi.org/10.1037/1082-989X.7.1.64) [10.1037/1082-989X.7.1.64.](https://doi.org/10.1037/1082-989X.7.1.64) - 34. Pais-Ribeiro JL, Honrado A, Leal I. Contribuição para o estudo da adaptação portuguesa das escalas de ansiedade, depressão e stress (EADS) de 21 itens de Lovibond e Lovibond. Psicol Saude Doencas. 2004;5(2):229–39. - 35. Ramalho SM, Saint-Maurice PF, Félix S, Conceição E. Intuitive eating scale-2: factor structure and associations with disordered eating, impulsivity and quality of life in adolescents with overweight/obesity. Eat Behav. 2022. [https://doi.org/10.1016/j.eatbeh.2021.101593.](https://doi.org/10.1016/j.eatbeh.2021.101593) - 36. Reas DL, Dahlgren C, Wonderlich J, Syversen G, Kvalem I. Confrmatory factor analysis and psychometric properties of the Norwegian version of the Repetitive Eating Questionnaire: further evidence for two distinct subtypes of grazing behaviour. Eur Eat Disord Rev. 2019;27(2):205–11. . - 37. Ribeiro A, Sinval J, Félix S, Guimarães C, Machado BC, Gonçalves S, de Lourdes M, Conceição EM. Food addiction and grazing—the role of difculties in emotion regulation and negative urgency in university students. Nutrients. 2023;15(20):4410. [https://doi.org/10.3390/nu152](https://doi.org/10.3390/nu15204410) [04410.](https://doi.org/10.3390/nu15204410) - 38. Rossi AA, Mannarini S, Semonella M, Castelnuovo G, Pietrabissa G. The association between grazing and food addiction: the Italian version of the Repetitive Eating Questionnaire (Rep(Eat)-Q) and its relationships with food addiction criteria. Nutrients. 2024;16(7):949. [https://doi.org/](https://doi.org/10.3390/nu16070949) [10.3390/nu16070949.](https://doi.org/10.3390/nu16070949) - 39. Teixeira MdelCB, Pereira ATF, Saraiva JMT, Marques M, Soares MJ, Bos SC, Valente J, de Azevedo MHP, de Macedo AJF. Portuguese validation of the children's eating attitudes test. Rev Psiquiatr Clin. 2012;39(6):189– 93. [https://doi.org/10.1590/S0101-60832012000600002.](https://doi.org/10.1590/S0101-60832012000600002) - 40. Teodoro MC, Conceição EM, Sinval J, de Lourdes M, Neufeld CB. Adaptation, confrmatory factor analysis, and psychometric properties of the - Brazilian version of the Repetitive Eating Questionnaire. Int J Eat Disord. 2023;56(4):747–57. - 41. Thorell LB, Lazarevic N, Milovanovic I, Bugarski I. Psychometric proper ties of the Teenage Executive Functioning Invesntory (TEXI): a freely available questionnaire for assessing defcits in working memory and inhibition among adolescents. Child Neuropsychol. 2020;26(6):857–64. - 42. Tylka TL, Kroon Van Diest AM. The intuitive eating scale-2: item refne ment and psychometric evaluation with college women and men. J Counsel Psychol. 2013;60(1):137–53. - 43. Van Swearingen KM, Noel NE. Impulsivity traits associated with disor dered eating and binge drinking among female college students. J Am Coll Health. 2024;72(2):509–14. [https://doi.org/10.1080/07448481.2022.](https://doi.org/10.1080/07448481.2022.2047696) [2047696](https://doi.org/10.1080/07448481.2022.2047696) - 44. Vandenberg RJ, Lance CE. A review and synthesis of the measurement invariance literature: suggestions, practices, and recommendations for organizational research. Organ Res Methods. 2000;3(1):4–70. [https://](https://doi.org/10.1177/109442810031002) [doi.org/10.1177/109442810031002](https://doi.org/10.1177/109442810031002) - 45. Whiteside SP, Lynam D, Miller J, Reynolds S. Validation of the UPPS impulsive behavior scale: a four factor model of impulsivity. Eur J Pers. 2005;19:559–74. - 46. Yoon C, Mason SM, Hooper L, Eisenberg ME, Neumark-sztainer D. Dis ordered eating behaviors and 15-year trajectories in body mass index: fndings from project eating and activity in teens and young adults (EAT). J Adolesc Health. 2019. [https://doi.org/10.1016/j.jadohealth.](https://doi.org/10.1016/j.jadohealth.2019.08.012) [2019.08.012](https://doi.org/10.1016/j.jadohealth.2019.08.012) # **Publisher's Note** Springer Nature remains neutral with regard to jurisdictional claims in pub lished maps and institutional afliations.