Bridging the Gap: Evaluating Tools for Adherence to Dietary‐Related Behavioral Recommendations After Metabolic Bariatric Surgery—A Scoping Review

ABSTRACT

Background and Aims

Adherence to dietary‐related behavioral recommendations following metabolic bariatric surgery (MBS) is important for achieving optimal surgical outcomes, but standardized definitions and tools are lacking. This scoping review aimed to map existing tools for assessing adherence to dietary‐related behavioral recommendations after MBS, evaluate their content and psychometric properties, and identify gaps to guide future research and tool development.

Methods

The review was registered on the Open Science Framework and followed PRISMA‐ScR guidelines. Searches were conducted in PubMed and Embase from inception through April 2024, supplemented by hand‐searching the reference lists of included reports. The review included studies of adults who underwent MBS, using tools with at least two items assessing adherence to dietary‐related recommendations, with detailed tool descriptions, and, when available, information on validity and reliability. Screening and data extraction were performed independently by two reviewers, with discrepancies resolved by a third reviewer.

Results

From 3223 publications, 16 reports from 10 countries were included. Tools assessing dietary adherence post‐MBS varied widely in content and behavioral targets, with most robust psychometric properties lacking. Reliability was assessed via Cronbach's alpha and test–retest methods and validity via face, content, construct, and criterion measures. Standardized recall periods and comprehensive scoring systems were notably absent.

Conclusions

Existing tools for assessing adherence to dietary‐related behavioral recommendations after MBS show significant variability, with most lacking standardized psychometric properties and recall periods, limiting their utility. Future research should focus on standardizing dietary‐related “core principles,” facilitating the development of new instruments in research and clinical settings.

1. Introduction

Patients undergoing metabolic bariatric surgery (MBS) are encouraged to adhere to specific behavioral principles after surgery to minimize preventable complications and achieve optimal post‐surgical health outcomes [1, 2, 3, 4, 5, 6]. These include dividing food intake into structured meals throughout the day, consuming high‐protein foods, chewing food slowly and thoroughly, ending meals when feeling “comfortably full,” avoiding carbonated drinks, separating liquids from solids, taking daily dietary supplements, engaging in physical activity, and adhering to a follow‐up regimen with the multidisciplinary team [7, 8]; however, behavioral recommendations may vary over time and across centers [3]. Adherence is influenced by a range of factors spanning psychological aspects such as motivation and emotional eating, physical and social barriers like access to care and socioeconomic status, and cognitive demands like understanding and following a complex set of recommendations [9, 10]. Adherence to dietary‐related health behaviors can be measured using various methods, such as self‐report questionnaires, electronic monitoring, pharmacy refill data, and blood tests, each offering distinct advantages and limitations [3, 11, 12]. Presently, there is significant variability in how adherence to the dietary‐related behavioral recommendations after MBS is defined and measured, with no universally accepted gold standard measure or instrument currently available for measuring adherence [3]. The lack of consistency in measuring adherence to MBS recommendations poses challenges in determining the impact of non‐adherence on various clinical outcomes within this specific population and ultimately designing and testing the relevant interventions to improve such outcomes [3]. To develop and test a validated instrument to assess such aspects in the future, a critical initial step is to map out existing tools designed for assessing adherence to the MBS dietary‐related behavioral recommendations. Accordingly, we conducted a review to systematically map and summarize available research, identify the range of tools currently in use, assess their psychometric properties, and indicate the main gaps for further research. Specifically, a scoping review was chosen for its ability to broadly explore and map research on adherence tools, without the narrow focus or strict quality assessments of a systematic review [13].

2. Materials and Methods

This review was prospectively registered on the Open Science Framework (https://osf.io/2ekqy) and has been reported according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses extension for Scoping Reviews (PRISMA‐ScR) Checklist [14].

2.1. Inclusion Criteria

To be eligible for inclusion, studies had to meet specific criteria of including adult participants (≥ 18 years) who underwent MBS, using tools with at least two items assessing adherence to dietary‐related recommendations, with detailed tool descriptions, while information on validity and reliability was extracted when available (Table 1).

TABLE 1.

Eligibility criteria according to the participant, intervention, comparator, outcome, and study design (PICOS) framework.

Criteria	Included	Excluded
Participant	Adults (≥ 18 years) who have undergone MBS If a study included pre‐operative patients, it was only included if it also tracked them post‐operatively or included post‐operative individuals	Animal, < 18 years of age, non‐MBS surgery populations (including endoscopic procedures), pregnancy
Indicator	Assessment tool with at least two items specifically measuring adherence to dietary‐related behavioral recommendations for MBS If an assessment tool was initially developed for a different population but had been adapted, tested, or validated for use in MBS populations, it was included	Not presenting an assessment tool or not an assessment tool measuring adherence to a dietary‐related recommendation related to MBS Studies measuring adherence solely to physical activity or follow‐up without dietary‐related recommendations were excluded Tools designed specifically for diagnosing eating or psychiatric disorders were not considered
Comparator	Any
Outcomes	Assessment tool description The description of the tool must include the individual items Data on validity and reliability were extracted when available	Does not include assessment tool description
Study design	Cross‐sectional, prospective or retrospective cohort, case–control, diagnostic accuracy, interventional studies, and protocols	Reviews, qualitative research, case reports
Other	Peer‐reviewed original contribution literature	Other publication type including guidelines, reports, book chapters, theses, editorials, letters to the editor, commentaries, conference abstracts, and preprints Full text unavailable or could not be translated into English

Abbreviation: MBS, metabolic bariatric surgery.

For clarity, the following definitions for validity (i.e., the degree to which a questionnaire measures what it intends to measure) and reliability (i.e., the degree to which a questionnaire yields consistent results) were used; face validity refers to the extent to which respondents subjectively perceive the questionnaire as clear and suitable for accurate responses; content validity refers to the extent to which a questionnaire fully represents the domain of the construct it aims to assess; construct validity refers to the extent to which questionnaire scores relate or unrelate to other measures as expected based on theoretical hypotheses; criterion validity refers to the extent to which questionnaire scores correlate with an external criterion, often considered a gold standard or real‐world outcome; internal consistency refers to the extent to which questionnaire items are intercorrelated and measure the same construct; and test–retest reliability refers to the extent to which a questionnaire produces consistent scores when administered repeatedly over time [15, 16, 17].

2.2. Search Method and Screenings

A search of PubMed and Embase was conducted from database inception through April 18, 2024, using a combination of keywords and controlled vocabulary (Table S1). All citations retrieved from database searches were imported into citation software (EndNote 20), where duplicates were automatically removed and then manually verified [18]. A piloting process was conducted, requiring two rounds of screening 30 reports to refine the eligibility criteria and achieve over 85% agreement before formal screening. Titles, abstracts, and full texts were then independently screened by two investigators (L.H., D.A., L.M.S., C.W., S.S.D., and T.B.P.) using Covidence software [19], with a third reviewer resolving disagreements (S.S.D. or T.B.P.). Reasons for excluding full texts were documented. The search was supplemented with hand‐searching reference lists of included articles to identify additional studies.

2.3. Data Extraction

Data extraction included study characteristics (i.e., author, country, year, study type, and sample size), participant details (i.e., age, gender, anthropometrics, and type of MBS), and information on the assessment tool (i.e., description [i.e., language, purpose, item content and development, recall period, and scoring], types of validity [i.e., face, content, construct, and criterion], and types of reliability [i.e., internal consistency and test–retest reliability]). Two reviewers independently performed the extraction (L.H., D.A., C.W., and S.S.D.), resolving discrepancies with a third reviewer (T.B.P.). All data were recorded using a standardized extraction form. Where more than one paper was published from the same trial, results were presented for the overall trial, including all relevant individual papers. The data synthesis followed a narrative approach, which involved summarizing, tabulating, and descriptively integrating the findings from the included studies.

3. Results

The PRISMA flowchart [20] presented in Figure 1 outlines the identification of N = 3223 publications. After duplicates (n = 1252) were excluded, n = 1971 reports underwent title and abstract screening, and a further n = 1870 reports were excluded. Subsequently, n = 101 reports were sought for retrieval, and n = 97 full texts were screened, leading to the inclusion of n = 16 reports. No additional relevant publications were identified by hand‐searching the reference lists of reports included.

FIGURE 1.

PRISMA flow diagram of the literature search and study selection.

3.1. Studies Characteristics and Design

Population characteristics, study objectives, and study design of included studies are detailed in Table 2. Studies were conducted in the United States (n = 3) [21, 22, 23], Italy (n = 2) [24, 25], the Netherlands (n = 2) [26, 27], Norway (n = 2) [28, 29], Germany (n = 2) [30, 31], Belgium (n = 1) [32], Israel and Portugal (n = 1) [33], Israel (n = 1) [34], Saudi Arabia (n = 1) [35], and Sweden (n = 1) [36]. Study designs included cross‐sectional (n = 9) [23, 24, 25, 26, 31, 32, 33, 34, 35], longitudinal (n = 5) [21, 22, 27, 29, 36], a randomized controlled trial (n = 1) [28], and a single‐arm, pre‐post, prospective mixed‐methods study (n = 1) [30].

TABLE 2.

Population characteristics, study objectives, and study design of included studies.

Author, year	Region and sample size	Population characteristics	Study objectives	Study design
Spaggiari, 2020 a	Italy n = 41	Patients who have undergone SG (56%) or RYGB (44%). Time since surgery: 57.3 ± 10.5 months. Characteristics: mean age of 52.2 ± 11.9 years, 78.0% females, mean pre‐surgery BMI=NR, and mean post‐surgery BMI = 30.2 ± 5.7 kg/m2.	‐ Create and validate a questionnaire to objectivize compliance with dietary and lifestyle suggestions provided after MBS.	Cross‐sectional study (observational)
Alsehemi, 2023 a	Saudi Arabia n = 390	Patients who have undergone several types of MBS (surgery type NR) Time since surgery: ≥ 3 years. Characteristics: mean age of 36.8 ± 9.7 years, 56.2% females, mean pre‐surgery BMI = 45.0 ± 7.7 kg/m2, and mean post‐surgery BMI = 28.7 ± 5.7 kg/m2.	‐ Translate the Eating Behavior after Bariatric Surgery (EBBS) questionnaire into Arabic. ‐ Validate the EBBS Arabic version questionnaire for use in clinical and research settings.	Cross‐sectional study (observational)
Sherf‐Dagan, 2023	Israel (n = 277) and Portugal (n = 111)	Patients who have undergone OAGB. Time since surgery ranges from 1 month to 5 years and was categorized into: 1–6 months, 6–12 months, and 1–5 years. Characteristics (Israel): pre‐surgery age of 41.6 ± 11.0 years, 75.8% females, mean pre‐surgery BMI = 41.2 ± 4.8 kg/m2, and mean post‐surgery BMI of 33.0 ± 4.2 kg/m2 (1–6 months post‐surgery), 27.3 ± 3.9 kg/m2 (6–12 months post‐surgery), and 27.2 ± 4.3 kg/m2 (1–5 years post‐surgery). Characteristics (Portugal): pre‐surgery age of 45.6 ± 12.3 years, 79.3% females, mean pre‐surgery BMI = 40.1 ± 5.6 kg/m2, and mean post‐surgery BMI of 30.1 ± 3.4 kg/m2 (1–6 months post‐surgery), 28.4 ± 3.8 kg/m2 (6–12 months post‐surgery), and 26.0 ± 3.6 kg/m2 (1–5 years post‐surgery).	‐ Gain information on nutritional and lifestyle parameters from two samples of OAGB patients living in different countries.	Multicenter, cross‐sectional study (observational)
Bäuerle, 2022	Germany n = 543	Patients who have undergone several types of MBS with SG (55.6%), RYGB (25.8%), and omega loop bypass (11.2%) being the most common. Time since surgery ranges between < 6 to > 24 months. Characteristics: current mean age of 46.8 ± 10.0 years, 80.1% females, mean pre‐surgery BMI = 51.2 ± 7.9 kg/m2, and mean post‐surgery BMI of 35.6 ± 7.8 kg/m2.	‐ Development and validation of an instrument assessing dietary behavior in patients after MBS.	Cross‐sectional study (observational)
Kafri, 2011	Israel n = 60	Patients who have undergone SG. Time since surgery was < 1 or > 1 year post‐surgery. Characteristics: current mean age of 41.2 ± 11.9 years, 81.7% females, mean pre‐surgery BMI = 44.5 ± 6.0 kg/m2, and mean post‐surgery BMI = 31.5 ± 4.8 kg/m2.	‐ Characterize health‐promoting behaviors, food selection, and food tolerance after SG at two different points post‐surgery.	Cross‐sectional study (observational)
Konings, 2020	The Netherlands n = 101	Patients who have undergone several types of MBS, including AGB (55%), VBG (39%), and RYGB (6%). Time since surgery: 4.6 ± 2.4 years post‐surgery. Characteristics: mean age of 44.5 ± 9.6 years, 82% females, mean pre‐surgery BMI = 46.5 ± 6.7 kg/m2, and mean post‐surgery BMI, NR.	‐ Assess postoperative changes in eating behavior and compliance with dietary guidelines and the effect of healthy behavior on weight loss. ‐ Assess the associations between compliance with dietary guidelines and satisfaction regarding their weight loss trajectory.	Cross‐sectional study (observational)
Lier, 2011	Norway n = 141	Patients with obesity (n = 141), n = 127 underwent RYGB, and n = 14 were excluded. Time since surgery: 1 year post‐surgery of follow‐up. Characteristics: mean age of 42.0 ± 10.4 years, 73% females, mean pre‐surgery BMI = 45.2 ± 5.3 kg/m2, and mean post‐surgery BMI, NR.	‐ Assess if attendance to a preoperative counseling program improved weight loss or adherence to treatment guidelines in patients who underwent MBS.	Randomized controlled trial
Mathews, 2023 b	USA n = 107 (of them, n = 57 took the survey more than once)	Patients who have undergone several types of MBS, including SG (52.3%), RYGB (42.1%), and conversion from SG to RYGB (5.6%). Time since surgery ranges from 1 month to 15 years post‐surgery and was categorized into short‐term (1–3 months), intermediate (4–12 months), and long‐term (> 1 year). Characteristics: mean age of 49 years, 84.1% females, mean pre‐surgery BMI = 45 kg/m2, and mean post‐surgery BMI, NR.	‐ Identify barriers to postoperative vitamin and mineral compliance in patients undergoing MBS. ‐ Develop clinical guidelines to improve supplementation postoperatively based on patient perspectives.	Longitudinal study (observational)
Sandhu, 2023 b	USA n = 107	Patients who have undergone several types of MBS, including RYGB (44.9%), and SG (55.1%), both primary and revision/conversion MBS. Time since surgery ranges from 1 month to 15 years post‐surgery Characteristics: < 40 years (29.9%) and > 40 years (69.2%), 84% females, mean pre‐surgery BMI = 45 kg/m2, and mean post‐surgery BMI, NR.	‐ Provide insights into the micronutrient biochemical profile in patients deemed compliant with supplementation following RYGB and SG.	Longitudinal study (observational)
Steenackers, 2022	Belgium n = 402	Patients who have undergone several types of MBS, including primary RYGB (67.7%), primary SG (17.4%), and secondary MBS (13.4%). Median time since surgery: 2.2 years (1.0–4.3). Characteristics: median age of 43.5 (34.0–52.0) years, 84.1% females, median pre‐surgery BMI = 41 (38.9–44.3) kg/m2, and median post‐surgery BMI = 28.0 (24.8–31.7) kg/m2.	‐ Explore supplement intake, compliance, and patients' perspectives towards nutritional supplementation after MBS. ‐ Examine whether compliance with nutritional supplementation can be predicted after MBS.	Cross‐sectional study (observational)
Welch, 2008 c	USA n = 300 (Clinic patient cohort: n = 100; Support group survey: n = 200)	Patients who have undergone RYGB. Mean time since surgery: 14.5 ± 13.9 months post‐surgery (for the support group sample). Characteristics (Clinic patient cohort): mean age of 42.6 ± 10.6 years, 87.9% females, mean pre‐surgery BMI and mean post‐surgery BMI, NR. Characteristics (Support group survey): mean age of 44.9 ± 9.7 years, 84.6% females, mean pre‐surgery BMI of = 53.5 ± 11.4 kg/m2, and mean post‐surgery BMI = 36.0 ± 8.3 kg/m2.	‐ Evaluate a new measure of postsurgical behaviors to assess patient adherence to recommended postsurgical behaviors. ‐ Investigate related aspects of patient adjustment to the postsurgical lifestyle. ‐ Examine predictors of weight loss over time, including demographic, clinical, behavioral, and psychosocial variables.	Cross‐sectional study (observational)
Pyykkö, 2023 c	The Netherlands n = 263	Patients who have undergone several types of MBS, including RYGB (89.4%), Omega loop Gastric Bypass (8.4%), and SG (2.3%). Time since surgery: 2‐year follow‐up. Characteristics: mean age of 47.7 ± 10.5 years, 75.7% females, mean pre‐surgery BMI = 39.0 ± 3.6 kg/m2, mean post‐surgery BMI = 27.5 ± 3.5 kg/m2 at 1‐year post‐surgery and 27.8 ± 3.9 kg/m2 at 2 years post‐surgery.	‐ Investigate the serial mediation between preoperative attachment and 2‐year post‐operative health behaviors through self‐esteem and health self‐efficacy. ‐ Provide insights and recommendations for aiding patients in adhering to healthier lifestyles post‐MBS.	Longitudinal, multicenter study (observational)
Santonicola, 2022	Italy n = 290	Patients who have undergone several types of MBS, including SG (59%), RYGB (31%), and AGB (7.2%). Time since surgery ranges from < 1 to 10 years post‐surgery. Characteristics: mean age of 39.5 ± 10.1 years, 81.4% females, mean pre‐surgery BMI = 45.1 ± 8.4 kg/m2, and mean post‐surgery BMI, NR.	‐ Evaluate the adherence to micronutrient supplementation in a cohort of Italian MBS patients. ‐ Identify predictors of adherence using a self‐administered, anonymous, internet‐based survey.	Cross‐sectional study (observational)
Spetz, 2024	Sweden Cohort 1: n = 120 Cohort 2: n = 211	Patients who have undergone several types of MBS, including RYGB (80.8%) and SG (19.2%) in Cohort 1 and RYGB (88.6%) and SG (11.4%) in Cohort 2. Time since surgery was at 1 year post‐surgery (cohort 1) and at 2 years post‐surgery (cohort 2). Characteristics (cohort 1): mean age of 42 ± 9 years, 81.7% females, mean pre‐surgery BMI of = 40 ± 6 kg/m2, and mean post‐surgery BMI, NR. Characteristics (cohort 2): mean age of 43 ± 11 years, 78.7% females, mean pre‐surgery BMI of = 40 ± 5 kg/m2, and mean post‐surgery BMI, NR.	‐ To assess the accuracy of the 5‐item Medication Adherence Report Scale (MARS‐5) in measuring adherence to vitamin and mineral supplementation post‐MBS.	Longitudinal study (observational)
Sundgot‐Borgen, 2024	Norway n = 73	Patients who have undergone several types of MBS, including RYGB (98.6%) and SG (1.4%). Time since surgery: 1–5 years post‐surgery. Characteristics: mean age of 53.9 ± 9.0 years, 80.8% females, mean pre‐surgery BMI = NR, and mean post‐surgery BMI = 32.1 ± 4.9 kg/m2.	‐ Explore changes in and associations between adherence to physical activity and general dietary recommendations after MBS. ‐ Determine whether physical activity and dietary behaviors interact to predict weight recurrence.	Longitudinal study (observational)
Yang, 2022	Germany n = 26	Patients who have undergone several types of MBS, including RYGB (65.4%) and SG (34.6%). Time since surgery: up to 1 year post‐surgery. Characteristics: mean age of 41.2 ± 10.1 years, 73.1% females, mean pre‐surgery BMI = 47.2 ± 7.1 kg/m2, and mean post‐surgery BMI, NR.	‐ Evaluate the usability of a smartphone application‐based follow‐up program in patients after MBS.	Single‐arm, pre‐post, prospective mixed‐methods study

Abbreviations: AGB, adjustable gastric banding; EBBS, Eating Behavior after Bariatric Surgery; MARS‐5, 5‐item Medication Adherence Report Scale; MBS, metabolic bariatric surgery; NR, not reported; OAGB, one anastomosis gastric bypass; RYGB, Roux‐en‐Y gastric bypass; SG, sleeve gastrectomy; VBG, vertical banded gastroplasty.

^a Alsehemi (2023) is a translated version of Spaggiari (2020).

^b Mathews (2023) and Sandhu (2023) are based on the same study tool but address different research questions.

^c Pyykkö (2023) used the eating behavior subscale published by Welch (2008).

3.2. Mapping of Identified Tools

Tools description, reliability, and validity are presented in Table 3. Studies examining self‐assessment tools for dietary‐related behavioral recommendations adherence following MBS reveal inconsistencies in reliability, validity, and item relevance. Several tools demonstrated reliability through internal consistency, as assessed by Cronbach's alpha [23, 25, 27, 29, 35, 36] and consistency by test–retest reliability [23]. Validity testing varied across tools, with some employing face [24], content [31, 32, 35], construct [23, 25, 31, 35], and criterion validity [23, 25, 31, 35, 36]. Table 4 outlines dietary‐related behavioral recommendations for MBS included in the tools. Item relevance varied, with certain tools focusing broadly on dietary behaviors, while others included specific adherence aspects such as supplementation intake.

TABLE 3.

Tool description, reliability, and validity.

Author, year	Tool description	Reliability	Validity
Spaggiari, 2020 a	Tool: Eating Behavior after Bariatric Surgery (EBBS) Questionnaire Language: Italian. Purpose: To quantify compliance with dietary and lifestyle suggestions after MBS. Items: 11 items (a modified version of the first version which included a 16‐item tool) with four functional domains: domain A (foods), domain B (drinks), domain C (behaviors), and domain S (lifestyle). Item development: Based on interviews with patients who underwent MBS; refined from 16 to 11 items after pilot testing. Recall period: NR. Scoring: Scores range from 0 to 2 per item; higher scores indicate better adherence.	Internal consistency: by Cronbach's α of 0.743 overall, and Hotelling's T‐square test p < 0.001 (0.725 for males and 0.751 for females).	Construct validity: Correlations were performed between the EBBS questionnaire and psychological tests, with mixed results. Criterion validity: Individual items were significantly correlated with the weight‐related outcomes (p < 0.001). However, no statistically significant correlation was observed between the weight‐related endpoints and the total score of the EBBS questionnaire.
Alsehemi, 2023 a	Tool: The Arabic Version of the Eating Behavior After Bariatric Surgery (EBBS) Questionnaire Language: Arabic. Purpose: To evaluate behavior and compliance of patients regarding dietary recommendations after undergoing MBS. Items: Originally 11 items, but reduced to 10 items, after omitting the alcohol consumption item (due to low variance, low CVI score, and religion constraints). Four domains include foods (A), drinks (B), behaviors (C), and lifestyle (S). Item development: Translation followed WHO guidelines with expert panel reviews and pilot testing. Recall period: NR. Scoring: Scores range from 0 to 2 per item; higher scores indicate better adherence.	Internal consistency: the 10 items had a Cronbach's α of 0.851.	Content validity: The I‐CVI and S‐CVI were assessed. The I‐CVI ranged between 0.9 and 1.0 for all items, except for question five (alcohol consumption), which was 0.6. The S‐CVI/Ave was 0.94 and the S‐CVI/UA was 0.63. Construct validity: The EFA, PCA, and parallel analysis tests were applied. All the items were loaded differently from their intended latent factors. Criterion validity: Significant correlations were found between weight‐related outcomes and the tool's domains.
Sherf‐Dagan, 2023	Tool: Compliance with the MBS Eating and lifestyle recommendations. Language: Hebrew (Israel) and Portuguese (Portugal) (English translation is available). Purpose: To assess compliance with the MBS eating and lifestyle recommendations. Items: 8 grouped items for eating recommendations, plus single items for lifestyle recommendations. Item development: NR. Recall period: During the last month. Scoring: 0 (not done), 1 (partially done), or 2 (always done) per item (eating recommendations), and mixed scoring per item (lifestyle recommendations), without overall scoring calculation or interpretation.	NR	NR
Bäuerle, 2022	Tool: Dietary Behavior Inventory‐Surgery (DBI‐S). Language: German (English translation is available). Purpose: To assess adherence to dietary behavior recommendations and guidelines for patients after MBS. Items: Originally 19 items, but reduced to 13 items after excluding ones that had low variance and ceiling effects. Item development: based on clinical practice recommendations and theory based. Recall period: NR. Scoring: Items scored by the semantic differential from 1 (=like behavior A) to 5 (=like behavior B), while several items were inverted. The total score ranges from 13–65, with higher scores representing greater adherence.	NR	Content validity: Based on scientific recommendations and several rounds of interdisciplinary experts' review. Construct validity: Correlations were performed between DBI‐S, an attitude towards healthy food item, and an impulsivity questionnaire. Significant correlation between DBI‐S score and attitude towards healthy food item (r = 0.36, p < 0.001), and impulsivity questionnaire (r = −0.38, p < 0.001). Cluster analysis confirms the ability to distinguish between rather unhealthy and healthy dietary behavior. Criterion validity: Correlations were performed between DBI‐S, BMI, and a quality‐of‐life item. Significant correlation between DBI‐S score and pre/post‐surgery BMI difference (r = −0.14, p = 0.002), and quality of life item (r = 0.19, p < 0.001).
Kafri, 2011	Tool: Multidimensional questionnaire specifically constructed for the present study. Language: Hebrew. Purpose: To assess eating style and health‐related behaviors post‐SG. Items: A total of 14 items, including single items for eating style and health‐related behaviors. Item development: NR. Recall period: NR. Scoring: Eating style (3‐point scale) and health‐related behaviors (mixed scores), without overall scoring calculation or interpretation.	NR	NR
Konings, 2020	Tool: Multidimensional survey specifically constructed for the present study. Language: Dutch. Purpose: To assess changes in eating behavior, compliance with postsurgical dietary guidelines, compensatory behavior, and alcohol use. Items: A total of 18 items, including eating behavior (4 items), compensatory behavior (5 items), postsurgical dietary guidelines (6 items), and alcohol use (2 items). Item development: NR. Recall period: NR. Scoring: Eating behavior and postsurgical guidelines items (4‐point Likert scale [never, sometimes, often, always]), compensatory behavior, and alcohol use items (yes/no and 4 options [no, small improvement, large improvement, resolved]), without overall scoring calculation or interpretation. The survey ends with one open‐ended text field question regarding satisfaction with surgery.	NR	NR
Lier, 2011	Tool: Survey specifically constructed for the present study. Language: Norwegian. Purpose: To assess eating habits, vitamin use, and physical exercise. Items: A total of 3 items, including eating habits (1 item), vitamin use (1 item), and physical exercise (1 item). Item development: NR. Recall period: NR. Scoring: A mix of frequency and descriptive responses, without overall scoring calculation or interpretation.	NR	NR
Mathews, 2023 and Sandhu, 2023 b	Tool: Survey specifically constructed for the present study. Language: English. Purpose: To assess nutritional supplementation practices and barriers. Items: A total of 11 items, including brand/type/dose/form of supplements (6 items), frequency of use (1 item), barriers to supplementation (1 item), insurance coverage (1 item), cost of supplements (1 item), and strategies for remembering (1 item). Item development: NR. Recall period: NR. Scoring: dichotomous (yes/no), multiple choice, and open‐ended free‐response questions, without overall scoring calculation or interpretation.	NR	NR
Steenacker, 2022	Tool: Survey specifically constructed for the present study which also includes a modified version of the ProMAS. Language: Dutch. Purpose: To assess information regarding the intake of nutritional supplements. Items: A final questionnaire of 58 items divided into five modules; one module questioned the intake of supplements based on the modified version of the ProMAS (18 items which were rephrased by converting “medicine” into “nutritional supplements” and by adjusting question 15 on prescriptions), one module questioned barriers and facilitators for supplement intake, and one module questioned the belief of post‐MBS patients towards nutritional deficiencies and supplementation. Item development: based on a literature review and existing compliance measures, followed by an iterative internal review to reach a consensus. Recall period: NR. Scoring: For most questions, multiple responses and patients could add information by free text, while the ProMAS items were scored on a binary answer scale (1 = yes, true; 0 = no, not true) and based on the total sum score (range: 0–18), categorized into low compliance (0 ≤ score ≤ 4), medium‐low compliance (5 ≤ score ≤ 9), medium‐high compliance, (10 ≤ score ≤ 14), and high compliance (15 ≤ score ≤ 18).	NR	Content validity: an independent panel of nine experts with experience related to MBS and a panel of 10 patients who underwent MBS assessed the developed items without the previously validated ProMAS items. The experts assessed the items quantitatively (CVI: CVI‐I, S‐CVI) and qualitatively, while overall the S‐CVI was 0.93. The patients were interviewed. **The questionnaire's readability was evaluated using the Flesch–Kincaid Grade Level, which was calculated to be 9.6.
Welch, 2008 c	Tool: Bariatric Surgery Self‐Management Questionnaire (BSSQ) Language: English. Purpose: To assess self‐management behaviors carried out over the previous week. Item: Total of 33 items across seven domains: eating behaviors (EB, 8 items); fluid intake (FI, 8 items); physical activity (PA, 3 items); dumping syndrome management (DSM, 4 items); supplement intake (SI, 4 items); fruit, vegetable, and whole grains intake (FVW, 3 items); protein intake (PI, 3 items). Item development: through a literature review and interviews with MBS team clinicians. Subscale items were developed through team consensus. Recall period: the past week. Scoring: Likert‐type scale (never, sometimes, always), with a converted subscale and total scores to a 0–100 range, with a higher score denoting better adherence.	Internal consistency: measured by Cronbach's α for the total score (α = 0.83), and per each subscale (EB α = 0.83, FI α = 0.81, PI α = 0.74, PA α = 0.70, DSM α = 0.79, FVW α = 0.63, SI α = 0.79). Test–retest reliability: Two‐week test–retest reliability measured by ICC for the total score (ICC = 0.71) and per subscale (EB [ICC = 0.72], FI [ICC = 0.68], PI [ICC = 0.60], PA [ICC = 0.54], DSM [ICC = 0.66], FVW [ICC = 0.46], SI [ICC = 0.66]).	Construct validity: Correlations among BSSQ, LDQ, PBQ, and WRSM scores were calculated using Spearman's correlation index. The BSSQ total test correlated significantly with LDQ (r = −0.22, p < 0.01), PBQ (r = 0.31, p < 0.01), and WRSM (r = −0.17, p < 0.05). Also, patients were classified into 4 time periods post‐surgery (< 6, 6–12, 12–18, and > 18 months) and BSSQ subscale scores were compared. The mean total BSSQ score was consistently in the 60%–70% range over time, with the mean BSSQ subscale scores ranging widely from 61% to 90%, over time. Criterion validity: Stepwise multiple linear regression was used to identify if BSSQ subscales and other variables predicted weight loss. Three variables in the final model accounted for 73.0% of the variance (time from surgery [p < 0.0001], baseline weight [p < 0.0001], and BSSQ PA subscale [p < 0.001]). **The Flesch–Kincaid reading level was in the 7th–8th grade range.
Pyykkö, 2023 c	Tool: Bariatric Surgery Self‐Management Questionnaire (BSSQ) eating behavior subscale. Language: Dutch (data on translation validation was NR) Purpose: Assess adherence to MBS eating recommendations. Items: 8 items, based on the eating behavior subscale of BSSQ. Item development: Using a subscale of a tool previously validated. Recall period: the past week. Scoring: Likert‐type scale (never, sometimes, always), with a converted score range of 0–100, with a higher score denoting better adherence.	Internal consistency: Cronbach's α = 0.68 for the 2‐year post‐operative assessment.	NR
Santonicola, 2022	Tool: Anonymous internet‐based survey specifically constructed for the present study Language: Italian. Purpose: Evaluate patients' compliance with the prescribed micronutrient supplementation post‐MBS. Items: A total of 23 items, including the frequency of follow‐up visits post‐surgery (1 item) and adherence to micronutrient supplementation (7 items). Item development: by a multidisciplinary team of experts based on pre‐study interviews with them and patients. Recall period: NR. Scoring: multiple choice, Likert‐scale, and open‐ended free‐response questions, without overall scoring calculation or interpretation.	NR	Face validity: Preliminary 30 interviews were performed to detect problems through discussion groups with interviewers with the aim of obtaining useful feedback on the issues of the questionnaire.
Spetz, 2024	Tool: 5‐item Medication Adherence Report Scale (MARS‐5) Language: Swedish. Purpose: To measure self‐reported adherence to vitamin and mineral supplementation post‐MBS. Item: 5 items on forgetting, changing dosage, stopping, skipping, and taking less than prescribed; 1 item assesses unintentional non‐adherence and 4 items assesses intentional non‐adherence. Item development: The Swedish translation of MARS‐5 has previously been validated for another target population. Recall period: NR. Scoring: Ranges from 5 to 25; higher scores indicate higher adherence.	Internal consistency: by Cronbach's α = 0.81 (cohort 2) and α = 0.95 (cohort 1).	Criterion validity: by comparing MARS‐5 results with pharmacy refill data for vitamin B12 and combined calcium/vitamin D as an objective reference, while Spearman correlation coefficients ranged from 0.49 to 0.54 (p < 0.001 for all).
Sundgot‐Borgen, 2024	Tool: Adherence to dietary recommendations. Language: Norwegian. Purpose: To measure adherence to dietary recommendations. Item: 6 items, including regular meal pattern (1 item), limitation of fat and/or sugar intake (3 items), whole grains intake (1 item), fruit and vegetable intake (1 item). Item development: based on previously used self‐developed questions built according to the dietary guidelines of the Norwegian Health Directorate. Recall period: the last 4 weeks. Scoring: On a Likert scale from 1 (little) to 7 (a lot), a composite score averages across items, with a higher score indicating a higher degree of adherence.	Internal consistency: by Cronbach's α = 0.72.	NR
Yang, 2022	Tool: A survey specifically constructed for the present study delivered by the MYONCARE app Language: German. Purpose: Follow‐up care for patients post‐MBS. Item: A total of 29 items (with 16 key items mentioned in the paper) including dietary habits (3 items), satisfaction with food intake (1 item), maladaptive eating pattern (1 item), dietary supplements (5 items), and physical activity (4 items). Item development: constructed based on the database of the German registry for obesity and metabolic surgery. Recall period: NR. Scoring: multiple choice, without overall scoring calculation or interpretation.	NR	NR

Abbreviations: ASEQBS, Attitude‐Social influence‐Efficacy Questionnaire after Bariatric Surgery; BSSQ, Bariatric Surgery Self‐Management Questionnaire; DASBS, Dietary Adherence Scale after Bariatric Surgery; EFA, exploratory factor analysis; I‐CVI, Content Validity Index; LDQ, Lifestyle Distress Questionnaire; MARS‐5, 5‐item Medication Adherence Report Scale; MBS, metabolic bariatric surgery; NR, not reported; PBQ, Perceived Benefits Questionnaire; PCA, principal components analysis; ProMAS, Probabilistic Medication Compliance Scale; S‐CVI, Scale‐level Content Validity Index; S‐CVI/AVE, Scale‐level Content Validity Index/Average; S‐CVI/UA, Scale‐level Content Validity Index/Universal Agreement; WRSM, the weight‐related symptom measure; WHO, World Health Organization.

^a Alsehemi (2023) is a translated version of Spaggiari (2020).

^b Mathews (2023) and Sandhu (2023) are based on the same study tool but address different research questions.

^c Pyykkö (2023) used the eating behavior subscale published by Welch (2008).

TABLE 4.

Dietary‐related behavioral recommendations for metabolic and bariatric surgery included in the tool.

Dietary‐related behavioral recommendations for metabolic bariatric surgery	(Spaggiari, 2020) a	(Alsehemi, 2023) a	(Sherf‐Dagan, 2023)	(Bäuerle, 2022)	(Kafri, 2011)	(Konings, 2020)	(Lier, 2012)	(Mathews, 2023, and Sandhu, 2023) b	(Steenackers, 2022)	(Welch, 2008) c	(Pyykkö, 2023) c	(Santonicola, 2022)	(Spetz, 2024)	(Sundgot‐Borgen, 2024)	(Yang, 2022)
Expected
Regular meal pattern	●	●	●		●	●	●			●	●			●
Start eating when feeling hungry and/or end meals when feeling comfortable full	●	●	●		●					●	●
Portion control					●	●				●	●
Avoiding grazing and/or snacking				●	●	●									●
Eating slowly and/or chewing food extensively and/or taking small bites	●	●	●	●	●					●	●
Adequate protein intake by food and/or supplement intake			●	●	●					●					●
Avoiding high‐calorie and/or high‐sugar foods and drinks	●	●	●	●	●					●				●
Consuming fruits and vegetables	●	●		●	●					●				●
Keeping adequate hydration			●	●						●					●
Separating between solids and liquids	●	●	●	●		●				●
Avoiding carbonated beverages	●	●	●
Adhering to dietary supplement intake			●		●	●	●	●	●	●		●	●		●
Other recommended behaviors included
Performing physical activity	●	●	●		●		●			●					●
Attending regular follow‐up visits			●		●							●
Self‐monitoring weight	●	●	●
Avoiding alcohol	●		●			●

^a Alsehemi (2023) is a translated version of Spaggiari (2020).

^b Mathews (2023) and Sandhu (2023) are based on the same study tool but address different research questions.

^c Pyykkö (2023) used the eating behavior subscale published by Welch (2008).

4. Discussion

Adherence among patients undergoing MBS can be defined as the degree to which behaviors, such as dietary and lifestyle modifications, align with healthcare recommendations, approximating actual behavior on a continuum [37]. Importantly, greater adherence to postoperative dietary‐related behavioral recommendations has been linked to improved health outcomes after surgery [1, 2, 3, 4, 5, 6]. The current scoping review was conducted to map the existing research on tools assessing adherence to dietary‐related behavioral recommendations in MBS, with an emphasis on evaluating their content and psychometric properties. Collectively, 16 reports from various countries and cultures were identified, with overall large discrepancies in terms of the appropriateness of individual items, validity, and reliability. The tools varied widely, reflecting inconsistent definitions, from single‐ to multi‐dimensional assessments of adherence to dietary‐related and lifestyle behavioral recommendations.

The development of an assessment instrument constitutes a systematic process that requires meticulous attention to theoretical frameworks and methodological rigor [38]. Key measurement properties emphasized in standard psychometric assessments include validity and reliability [39]. Nevertheless, most of the tools reviewed failed to fully adhere to essential steps in the development process, with only a small fraction successfully establishing robust psychometric properties [23, 25, 31, 35, 36]. In addition, some tools include items such as the frequency of alcohol consumption or self‐weighing routines [25], even in the absence of formal recommendations for these behaviors following MBS, while others incorporate outdated recommendations, such as in the area of supplementation [23]. In this context, an operational definition of an item ensures measurable characteristics, clarity, and consistency [38]. Therefore, defining items based on evidence or established practice is essential to maintain their relevance. However, given the inconsistencies in the definition of the “core principles” for achieving optimal outcomes after MBS, future studies should focus on standardizing these recommendations. Most tools remained descriptive, lacking overall scoring or interpretative frameworks, with only some offering validated scoring systems [23, 25, 27, 29, 31, 35, 36]. Consequently, without addressing these barriers, the applicability and generalizability of these tools remain limited [38].

One important factor to consider when discussing adherence to dietary‐related behavioral recommendations in MBS is the time elapsed since surgery, as patients with a longer postoperative duration often exhibit lower adherence to recommendations [29, 33]. This phenomenon may be attributed to physiological adjustments that occur over time following the surgery [40], as well as “behavioral fatigue,” where the initial motivation and commitment to adhere to the recommendations diminish with time [4]. Additionally, food quantity but not necessarily food quality may change throughout the post‐operative period [41, 42], and the behavioral recommendations provided to patients may change over time and between centers [3]. Nonetheless, most tools did not specify a target timeframe post‐surgery for responses, and only a few clearly defined a recall period [23, 27, 29, 33]. The recall period in a survey denotes the timeframe respondents are instructed to reference when answering questions. Clearly defining the recall period in questionnaires could impact the accuracy of self‐reported adherence [43], and therefore, is necessary. In addition, when assessing adherence, it is important to account for social desirability bias, which may arise when respondents provide answers they perceive as socially acceptable or likely to present them positively [44, 45]. Although this phenomenon is challenging to prevent in research and clinical practice, strategies such as providing a rationale for adherence assessment or including a preliminary normalizing statement have been suggested [3]. Nevertheless, the reviewed tools did not report on employing these strategies.

Incorporating behavioral science and patient‐oriented frameworks during the design phase of adherence tools can enhance their development by guiding preliminary research (e.g., patient interviews) to identify relevant domains and ensure comprehensive content validity. These approaches align tools with patient realities, addressing limitations observed across many existing tools, and improving their relevance and applicability in diverse settings [46]. Importantly, some of the included papers also went beyond assessing adherence to exploring its determinants and influences [21, 27, 32]. Recognizing the key determinants of adherence is essential to understanding the complex interplay of factors influencing dietary‐related behavioral recommendations following MBS [12]. Patients following MBS report barriers to adhering to recommendations including practical challenges (e.g., cost, time shortage, long travel distances to healthcare facilities, and insufficient guidance post‐surgery), physiological issues (e.g., food intolerance and gastrointestinal symptoms), and psychological factors (e.g., low self‐efficacy and motivation, and lack of social support [7, 12, 27, 33, 47, 48, 49].

Finally, future adherence tools might benefit from integrating digital health technologies, such as smartphone applications and wearable devices, to enhance data collection accuracy, enable continuous monitoring, and reduce reliance on self‐reported measures. These advancements could support a more objective and scalable approach to assessing adherence among patients undergoing MBS [50].

5. Strength and Limitations

A key strength of this scoping review is its comprehensive approach, offering a broad overview of tools for assessing adherence to dietary‐related behavior recommendations following MBS and identifying knowledge gaps and opportunities for future research, which a narrower systematic review might overlook. Nevertheless, some limitations should be mentioned. First, the included studies exhibited significant variation in study design, population characteristics, and objectives, which complicated the synthesis of findings and the direct comparison of the tools. Second, as MBS literature evolves with time, tools developed at different times may not fully align with current best practices or incorporate emerging recommendations, potentially limiting their relevance. Third, tools developed for different populations but adapted for MBS were included, which might lead to variability in their applicability to the specific needs of the MBS population. Fourth, key determinants of adherence were not included a priori in the inclusion criteria or the data extraction plan, which limited the assessment of how well each tool captured the full scope of adherence‐related dimensions.

6. Conclusion and Future Implications

Adherence to postoperative nutritional and lifestyle recommendations is considered important for achieving optimal surgical outcomes and minimizing preventable complications. In this context, the use of easy‐to‐use and cost‐effective validated tools to assess adherence is essential for accurately evaluating patient behaviors, identifying potential barriers and facilitators, and developing tailored interventions to enhance adherence. This scoping review mapped existing tools for assessing adherence to dietary‐related behavior recommendations after MBS, highlighting variability in their content and establishment of psychometric properties. While using questionnaires for data collection is common in the MBS literature, researchers must ensure they obtain the methodological framework [17] when creating new tools to assess adherence to dietary‐related behavior recommendations following MBS. Future studies should prioritize the standardization of dietary‐related “core principles” recommendations following MBS, facilitating the development of new instruments to assess dietary‐related behavioral adherence in research and clinical settings.

Author Contributions

Conceptualization, S.S‐D., T.B‐P., and C.W.; methodology, S.S‐D., T.B‐P., and C.W.; data curation, S.S‐D., T.B‐P., C.W., L.H., D.A., and L.M‐S.; writing – original draft preparation, S.S‐D., T.B‐P., and C.W.; writing – review and editing, S.S‐D., T.B‐P., L.H., D.A., and D.S.B.; project overseeing, co‐leading, and administration, S.S‐D. and T.B‐P. All authors have read and agreed to the published version of the manuscript.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

S.S.‐D. received research funds from Novo Nordisk. This funding is not related to the current manuscript. All other authors declare no conflicts of interest.

Supporting information

Table S1: Complete search strategy with record outcomes according to each database.

Sherf‐Dagan S., Wright C., Heusschen L., et al., “Bridging the Gap: Evaluating Tools for Adherence to Dietary‐Related Behavioral Recommendations After Metabolic Bariatric Surgery—A Scoping Review,” Obesity Reviews 27, no. 1 (2026): e70012, 10.1111/obr.70012.

Data Availability Statement

All data supporting this review are included in the article and its Supporting Information.