Experience Sampling as a dietary assessment method: a scoping review towards implementation

Background Accurate and feasible assessment of dietary intake remains challenging for research and healthcare. Experience Sampling Methodology (ESM) is a real-time real-life data capturing method with low burden and good feasibility not yet fully explored as alternative dietary assessment method. Methods This scoping review is the first to explore the implementation of ESM as an alternative to traditional dietary assessment methods by mapping the methodological considerations to apply ESM and formulating recommendations to develop an Experience Sampling-based Dietary Assessment Method (ESDAM). The scoping review methodology framework was followed by searching PubMed (including OVID) and Web of Science from 2012 until 2024. Results Screening of 646 articles resulted in 39 included articles describing 24 studies. ESM was mostly applied for qualitative dietary assessment (i.e. type of consumed foods) (n = 12), next to semi-quantitative dietary assessment (i.e. frequency of consumption, no portion size) (n = 7), and quantitative dietary assessment (i.e. type and portion size of consumed foods) (n = 5). Most studies used ESM to assess the intake of selected foods. Two studies applied ESM as an alternative to traditional dietary assessment methods assessing total dietary intake quantitatively (i.e. all food groups). ESM duration ranged from 4 to 30 days and most studies applied ESM for 7 days (n = 15). Sampling schedules were mostly semi-random (n = 12) or fixed (n = 9) with prompts starting at 8–10 AM and ending at 8–12 PM. ESM questionnaires were adapted from existing questionnaires, based on food consumption data or focus group discussions, and respond options were mostly presented as multiple-choice. Recall period to report dietary intake in ESM prompts varied from 15 min to 3.5 h. Conclusions Most studies used ESM for 7 days with fixed or semi-random sampling during waking hours and 2-h recall periods. An ESDAM can be developed starting from a food record approach (actual intake) or a validated food frequency questionnaire (long-term or habitual intake). Actual dietary intake can be measured by ESM through short intensive fixed sampling schedules while habitual dietary intake measurement by ESM allows for longer less frequent semi-random sampling schedules. ESM sampling protocols should be developed carefully to optimize feasibility and accuracy of dietary data. Supplementary Information The online version contains supplementary material available at 10.1186/s12966-024-01643-1.


Background
Research on health and nutrition relies on accurate assessment of dietary intake [1].However, dietary intake is a complex exposure variable with high interand intra-variability existing of different components ranging from micronutrients, macronutrients, food groups, meals to the dietary pattern as a whole.Therefore, measuring dietary intake accurately and feasibly is challenging for both researchers and healthcare professionals [2][3][4].Only few established nutritional biomarkers are available and, therefore, no objective method exist to reflect true dietary intake or the dietary pattern as a whole in epidemiological research [2,3].Instead, most dietary assessment methods rely on self-report.Food records, referred to as the "golden standard", together with 24-h dietary recalls provide most detailed dietary data while Food Frequency Questionnaires (FFQ) reflects habitual (i.e.long-term usual intake) dietary intake which is the variable of interest in most diet-disease research [4][5][6].Food records, 24-h dietary recalls, and FFQs have known limitations and challenges including recall bias, social-desirability bias, misreporting, and burdensomeness contributing to inherent measurement error in dietary intake data [2,6].A review of Kirkpatrick et al. showed that feasibility, including cost-effectiveness and ease-of-use, is the main determinant for researchers in selecting a dietary assessment method instead of appropriateness for study design and purpose at the expense of data quality and accuracy [7].To advance nutritional research and enhance the quality of dietary data, exploring the implementation of new methodologies is warranted to improve feasibility and overcome the limitations of current dietary assessment methods.
Experience Sampling Methodology (ESM), an umbrella term including Ecological Momentary Assessment (EMA), ambulatory assessment, and structured diary method, refers to intensive longitudinal assessment and real-time data-capturing methods [8].Participants are asked to respond to short questions sent through smartphone prompt messages or beeps at random moments during the day to assess experiences or behaviors and moment-to-moment changes in daily life [9].Originating from the field of psychology and behavioral sciences, ESM typically assesses current mood, cognitions, perceptions, or behaviors and descriptors of the momentary context (i.e.location, company) [9].Usually, assessments are collected in a random time sampling protocol yet, assessments can also be triggered by an event (event-contingent sampling), at fixed time points, or random within fixed time intervals (semi-random).ESM questionnaires are usually designed to be completed in under 2 min consisting of open-ended questions, visual analogue scales, checklists, or self-report Likert scales.Several ESM survey applications (i.e.m-Path, PsyMate, PocketQ) are currently available in which the sampling protocol and questionnaires can be customized to the study design and aim [10,11].It was shown that ESM reduces recall bias, reactivity bias, and misreporting in psychology and behavioral research by its design through unannounced, rapid, real-life, real-time repeated assessments [12].For this reason, Experience Sampling might be an interesting new methodology to explore as an alternative dietary assessment methodology.The design of ESM could overcome recall bias, reactivity bias, social desirability bias, and misreporting seen in traditional dietary assessment methods.However, the application of ESM for dietary assessment is new.Defining and balancing ESM methodological considerations, i.e. study duration, frequency and timing of sampling (signaling technique), formulation of questions and answer options, is a delicate matter and crucial in balancing feasibility with data accuracy [13].
The application of ESM in the field of dietary assessment has not been fully explored yet.Schembre et al. reviewed ESM for dietary behavior for the first time [12].However, it has not yet been assessed how ESM could be implemented as an alternative dietary assessment method aiming to estimate daily energy, nutrient, and food group intake quantitatively.
Therefore, this scoping review investigates how Experience Sampling Methodology can be implemented to develop an Experience Sampling-based dietary assessment method as an alternative to traditional dietary assessment methods to measure daily energy, nutrient, and food group intake quantitatively.This review aims to map ESM sampling protocols and questionnaire designs used to assess dietary intake.Additionally, the findings of this review will be combined with best practices to develop ESMs and dietary assessment methods to formulate key recommendations for the development of an Experience Sampling-based Dietary Assessment Method (ESDAM).The following questions will be answered: i) How is ESM applied in literature to assess dietary intake -focusing on methodological considerations (i.e.development and formulation of questions and answers, selection and consideration of prompting schedule (timing and frequency))?ii) How can ESM specifically be applied for quantitative assessment of total dietary intake (i.e. as an alternative to traditional dietary assessment method)?

Design
This scoping review followed the methodological framework for scoping reviews of Arksey and O'Malley which was further developed by Levac et al. [14,15].A scoping review approach was chosen to explore and map the design aspects and considerations for developing experience sampling methods to assess dietary intake as an alternative to traditional dietary assessment methods, which is novel.Moreover, this review will formulate design recommendations to apply ESM as a dietary assessment method and will serve as starting point to develop an ESDAM.An a priori protocol was developed based on the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) and the Joanna Briggs Institute Scoping Review protocol template (Supplementary Material) [16,17].According to Arksey and O'Malley methodological framework, the iterative nature of scoping reviews may include further refinement of the search strategy and the inclusion and exclusion criteria during the initial review process due to the unknown breadth of the topic [14].Therefore, adaptations made to the methodology described in the a priori protocol based on initial searches are described below.This scoping review was reported according to the PRISMA extension for scoping reviews (PRISMA-ScR) [18].

Search strategy and screening
The search strategy was developed based on key words and Mesh terms for "dietary assessment" and "experience sampling" (Supplementary Material).The term "ecological momentary assessment" was included as a synonym of ESM.Electronic databases PubMed (including MEDLINE) and Web of Science were searched for relevant literature published between January 2012 and February 9th 2024.The year 2012 was chosen as lower limit for inclusion since this review focuses on the use of ESM by digital tools (i.e.smartphones, web-based or mobile applications) which has emerged especially since the introduction of applications for smartphones since 2008.Therefore, the time frame of this review is focused on literature published in the last 12 years.The reference lists of all included articles were screened for additional studies.The initial search strategy described in the protocol was developed based on the assumption that research using ESM as an alternative to traditional dietary assessment was limited.Therefore, initially, research using ESM in the broader field of health research was included to obtain more evidence on methodological considerations of application of ESM.In line with the Arksey and O'Malley methodological framework, inclusion criteria were adapted following initial searches along with discussion and consensus between the reviewer (JV) and principal investigator (CM).Therefore, inclusion criteria were adapted to research applying ESM to measure dietary intake quantitatively or qualitatively since literature was also available in the field of dietary behaviour in relation to contextual factors (Table 1).Studies measuring dietary behaviour (i.e.cravings, hunger, eating disorder behaviour, dietary lapses) only, without assessing dietary intake, were excluded.Event-based ESM as dietary assessment method was excluded since this was deemed a similar methodology as the food record and, therefore, not serving the purpose of this review to explore a new methodology for dietary assessment to overcome limitations of traditional dietary assessment methods.All inclusion and exclusion criteria are presented in Table 1.
All records were exported and uploaded into the review software Rayyan.Duplicates were identified through the software followed by a manual screening of the reviewer for confirmation and removal of duplicates.One reviewer (JV) screened the retrieved articles first by title and abstract followed by a full text screening [19][20][21].In case of hesitancy on inclusion of articles, the reviewer (JV) consulted the principal investigator (CM) to reach consensus.In line with established scoping review methods, methodological quality assessment was not performed [14,18].Since this review aims to shed light on design

Literature search and study characteristics
The electronic databases search resulted in 701 articles of which 55 duplicates were identified and removed.Next, 646 articles were screened by title and abstract of which 591 were excluded according to the exclusion criteria (Fig. 1).The remaining 55 articles were screened by full text.After exclusion of 16 articles following full text screening, 39 articles were selected for inclusion (Table 2).The included articles describe 24 individual studies of which the Mother's and Their Children's Health (MATCH) study was described most frequently (n = 12, 25%).Most studies were published in 2018 (n = 7), followed by 2020 (n = 6) and 2022 (n = 6).Students, including both high school and higher education students, were the study population in most EMA or ESM studies included (n = 10, 43%).Two studies applied the ESM methodology to assess dietary behaviour including dietary variables of children with mothers as proxy.
Five studies referred to their methodology using the terminology 'ESM' while the other studies used 'EMA' as terminology.

Application of ESM for dietary assessment in literature Dietary variables measured through ESM
Most studies assessed consumption of specific foods only [22-39, 42, 44-55].[39,55].Two studies assessed the occurrence of food consumption only without assessing type of foods consumed [40,41].The study of de Rivaz et al. assessed the largest type of meal consumed in between signals [56].Three studies aimed to assess total dietary intake of which the study of Lucassen et al. evaluated approaches to assess both actual and habitual dietary intake using ESM [43,[57][58][59].

Qualitative versus quantitative dietary assessment through ESM
As shown in Table 2, twelve studies performed qualitative dietary assessment (i.e.assessing type of foods consumed without quantification) (Table 2).Seven studies performed semi-quantitative dietary assessment (i.e.assessing frequency of meals/eating occasions or number of servings of food categories not allowing nutrient calculation) [44,49,50,[52][53][54][55][56] (Table 3).Quantitative dietary assessment, in line with the aim of traditional dietary assessment methods (i.e.assessment of both type and quantity of foods consumed allowing to estimate nutrient intake), was performed in four studies of which  4).

Study duration, ESM timing and signaling technique
Study duration of ESM dietary assessment varied from four to thirty days of which most studies (n = 15) had a study duration of seven days of ESM dietary assessment.
The study of Piontak et al. had the longest duration of 30 days of ESM assessment [35].The semi-random sampling scheme (i.e.random sampling within multiple fixed time-intervals) was applied most frequently (n = 12), followed by the fixed sampling scheme (i.e.sampling at fixed times) (n = 9).Random sampling (i.e.completely random sampling) was chosen in three studies [34,36,55].A mixed sampling approach was applied in three studies of which Lucassen et al. tested and compared both a fixed sampling and a semi-random sampling approach to assess overall dietary intake [22,42,57,59].Two studies applied different sampling schemes during the weekend compared to weekdays [22][23][24][25][26][27][28][29][30][31][32][33].Sampling time windows were adapted to the daily structure of the study population, i.e. shifts of shift-workers, school hours of students or (selfreported) waking hours (Table 2).The sampling time window of the included studies started between 6 and 10 AM and ended between 8 PM and midnight.One study applied a 24-h sampling time window since the study population were nurses working in shifts [39].

Discussion
This review reveals that ESM has been applied to assess dietary intake in various research settings using different design approaches.However, most studies assessed consumption of specific foods only focusing on the foods of interest related to the research question.Especially snack consumption and, in general, unhealthy foods were the foods of interest for which ESM was used most often to measure its consumption.Due to its momentary nature, ESM may be especially suitable to measure these specific foods which are often (unconsciously) missed or underreported using traditional dietary assessment methods.Findings from our review show that ESM applied to assess dietary intake shows both features of 24-h dietary recalls (24HRs) and food frequency questionnaires (FFQ).Aside from the recall-based reporting and multiple choice assessment of specific foods, found in 24HRs and FFQs respectively, the ESM is a new methodology compared to traditional dietary assessment methods.ESM shows to lends itself well to assess the total dietary intake quantitatively as well albeit less explored yet according to our review.Moreover, most studies using ESM for dietary assessment were behavioral science research (i.e.psychological aspects of eating behavior) which highlights the novelty and need of ESM specifically designed for dietary assessment and research on diet-health associations.

Recommendations to develop an Experience Sampling-based Dietary Assessment Method
The implementation of ESM will differ depending which health behavior is being measured and in which research field it is being applied [13,60].This section describes recommendations of the methodological implementation of ESM as an alternative dietary assessment methodology to measure total dietary intake quantitatively based on the findings of this review, recommendations of the open handbook for ESM by Myin-Germeys et al. and practices in traditional dietary assessment development [13].

Recommendations for study duration, ESM timing and frequency
All ESM study characteristics (study duration, sampling frequency, timing, recall period) are interrelated and cannot be evaluated individually.
ESM study duration (i.e.number of days) and sampling frequency (i.e.number of prompts per day) should be reconciled and should be inversely adapted to one another (i.e.short study duration allows for higher sampling frequency per day and vice versa) to maintain low burden and good feasibility.
Our review showed an ESM study duration of 7 days is most common however reporting fatigue might arise from day 4 onwards in case of high sampling frequency (i.e.fixed sampling every 2 h) similarly as experienced with food records [61].
Frequency and timing of ESM prompts should be adapted to waking hours covering the typical eating episodes of the target study population.Typically, studies used waking hours starting around 7 AM till 10 PM however a preliminary short survey can identify feasible and accurate waking hours of the target study population and allow to adapt accordingly.
Waking hours, and consequently sampling frequency, could be different on weekend days (i.e. more frequent, longer waking hours) as seen in some studies in our review.Short recall periods (i.e. last hours or previous day) are suggested to be better than longer recalls of weeks or months [62].Aiming to obtain more accurate dietary intake data, lower recall bias and social desirability bias by reducing the awareness of being measured requests short recall periods of 1 up to 3.5 h, with a 2-h recall most commonly applied, as demonstrated by our review.In this way, ESM allow for near real-time measurements of dietary intake.
Furthermore, study duration, sampling frequency and timing should be adapted and differs when aiming to measure actual dietary intake or habitual dietary intake.

Recommendations ESM signaling technique for actual versus habitual dietary intake
Measuring actual dietary intake using an intensive prompting schedule can only be performed for short periods, preferably three to four days, due to the risk of responding fatigue as seen similarly in food records.As demonstrated by Lucassen et al. actual intake can be measured by ESM applying a fixed sampling approach which samples every time-window during the waking hours (i.e.sampling every 2 h between 7 AM and 10 PM on dietary intake during past 2 h) [58].
Habitual dietary intake can be measured by ESM applying a semi-random sampling approach which samples every time window during waking hours multiple times during a longer period (i.e.sampling three time-windows per day on dietary intake during past 2 h for two weeks until every time window is sampled three times) [58].Measuring habitual dietary intake by ESM using a less intensive sampling frequency allows for a longer study duration (i.e.multiple weeks).Lastly, a combination of fixed and (semi-)random sampling schedules can be applied.Both in case of measuring actual and habitual dietary intake, it is recommended to compose a sampling schedule with time windows covering all waking hours to ensure all eating occasions could be sampled [12].Additionally, the sampling schedule should cover weekend days next to week days to be able to sample the variability in dietary intake.More so, to capture variability of dietary intake several waves of ESM measurement periods could be implemented alternated with no-measurement periods.On the other hand, the application of multiple waves is associated with higher dropout rates especially with increased time in-between waves [13].
In conclusion, ESM signaling technique, frequency, timing, recall period and duration of sampling should be carefully adapted to one another to ensure accurate dietary intake data, low burden and optimal feasibility.As recommended by Myin-Germeys et al., a pilot study allows to evaluate all ESM design characteristics to obtain optimal data quality yet remain feasible [13].

Recommendations for ESM questions and response options
Questionnaires for ESM should be carefully developed and request methodological rigor [63].As stated by Myin-Germeys et al., there are currently no specific guidelines on how to develop questionnaires for ESM [63].However, according to our review most studies adapt existing questionnaires to implement in ESM research.Still, few studies in our review describe methodologically which or how adaptations are made to fit in the ESM format.First, a timeframe should be chosen on which the question will reflect.Although ESM is ideally consisting of questions on momentary variables, this is less suitable to measure dietary intake.As dietary intake does not continuously take place, momentary questions (i.e.What are you eating in this moment?)would lead to a large amount of missing data and, consequently, large measurement error on daily dietary intake estimations.Instead, time intervals lend itself better to assess dietary intake with ESM.The time interval on which the question reflects should be clearly stated (i.e.What did you eat during the last two hours?).As mentioned previously, in case of an interval contingent (semi-random) ESM approach, constitution of contiguous time intervals that cover the complete waking hour time frame (i.e.waking hours between 7 AM and 10 PM with semi-random ESM sampling by intervals of every two hours) is recommended to reduce risk of missing eating occasions [12].Therefore, following the latter approach, it is most feasible to choose the same time frame on which the question reflects as the time intervals of the prompts (i.e.semi random sampling in time intervals of two hours with question 'What did you eat since the last signal?').The time frame on which the question reflects should be chosen based on expected events of dietary intake (i.e.every two or three hours) and depends on dietary habits of the target population which is culture specific.Myin-Germeys et al. recommend to keep questions short and to the point so it fits the screen of the mobile device and allows for quick response [63].Furthermore, implicit assessments (i.e.Have you eaten since the last signal?)are recommended over explicit assessments (i.e.Did you eat fast food since the last signal?) to inhibit reactivity bias.Questionnaire length is important to consider as it is recommended to maintain a completion time of maximum three minutes to keep the burden low [63].Although in traditional ESM research questionnaires up to 30 items are accepted, in the field of dietary assessment, this would equivalent a short FFQ and can be considered too burdensome when presented all at once at every prompt reducing compliance.Moreover, ESM research in the field of psychology, where it originated from, uses most often scales (i.e.Likert scale, visual scales) as respond options.Unlike many psychological variables (i.e.mood, emotions), dietary intake can be assessed quantitatively and precise which allows for more specific response options.

Recommendations to develop ESM sampling scheme based on FFQ or food record
Questions and respond options for ESM dietary assessment could be adapted from existing questionnaires as demonstrated in the studies of our review.In the field of dietary assessment, ESM could therefore be applied to validated dietary assessment questionnaires such as validated Food Frequency Questionnaires (FFQ's) or (webbased) food records as proposed in Fig. 2.
Starting from the food record approach, a general open question (i.e.Did you eat anything since the last signal?)could be followed by a question to specify the consumed foods by an open field text box or food groups part originating from a National Food Consumption Database.Portion sizes of consumed foods could be provided by an open field text box with standard units (i.e.milliliters, grams) or common household measures (i.e.table spoons, glasses).
Starting from the FFQ approach, food groups assessed in FFQ's could be regrouped to a limited number and questions reformulated to assess dietary intake in near real time to design ESM questionnaires.Consumption of all food groups could be assessed at each prompt or consumption of a different set of food groups could be assessed at each prompt.In the latter case, the study needs to be designed so that consumption of each food group is assessed at each interval multiple times to account for unanswered prompts with missing data.Moreover, ordering of questions on consumption of food groups need to be considered as the consumption of specific food groups might need to be assessed at the same prompt to reduce ambiguity (i.e.fried food consumption needs to be assessed before consumption of fast food to avoid response overlap).Asking the same set of questions at each prompt may feel repetitive but might reduce burden [63].A control question can be added to assess careless responding.

Application of ESM as alternative dietary assessment method in literature
Most studies used ESM to measure food consumption qualitatively (i.e.type of foods consumed) or semiquantitatively (i.e.frequency of consumption of specific Fig. 2 Recommendations to implement experience sampling for actual and habitual dietary assessment foods) as opposed to quantitatively (i.e.type and quantity of foods consumed) to serve the same purpose as traditional dietary assessment methods.Questions were most often formulated using direct phrasing and asking about consumption of specific foods since the last signal.Answers were most often binary (i.e.yes/no indicating consumption of specific foods since last signal) combined with options to specify type and/or frequency or amount of foods consumed.Only the studies of Jeffers et al. and Lucassen et al. apply ESM to measure total dietary intake quantitatively of which Lucassen et al. evaluated ESM specifically as an alternative methodology for dietary assessment [57,58].
Although both event-contingent and signal-contingent approaches are being used for dietary assessment, signal-contingent ESM approaches might provide auspicious opportunities to overcome the limitations and biases of traditional dietary assessment methods [12].The near-real time data collection combined with (semi-)random sampling shows potential to reduce the burden for the participant both by its low intensity of registering and by its shorter questions with easy respond options.Moreover, the (semi-)random sampling technique might make the participant less aware of being measured resulting in possibly lower social-desirability bias leading, together with the short recall period, to more accurate data.In combination with modern technology such as mobile applications feasibility could be enhanced as well.Adapting questions and response options from either a validated FFQ or food record allow for relatively easy implementation of ESM as alternative dietary assessment method for total dietary intake (i.e.all food groups).However, validity and reliability need to be evaluated in the target population, similarly as traditional dietary assessment methods.
The systematic review and meta-analysis of Perski et al. states to have reviewed the use of ESM to assess five key health behaviors including dietary behavior [60].Similar to our findings, all four studies described by Perski et al. are assessing dietary intake through ESM of specific foods only instead of the total dietary pattern (i.e.all food groups).Moreover, Perski et al. included event-contingent sampling (i.e.registering dietary intake as it occurs) approaches as well.As highlighted by Schembre et al. event-contingent sampling entails similar limitations and biases such as social desirability bias and burden as the traditional dietary assessment methods [27].Not surprising, as event-contingent sampling can be seen as a similar approach as the traditional food record and serves for this reason not the purpose of this review to define a new methodology to overcome the limitations of current traditional dietary assessment methods.Similarly, photo-based methodologies (i.e. using images as food diary by event-based sampling) are unlikely to overcome the limitations of traditional dietary assessment methods due to the large measurement error in estimation of portion sizes and types of foods and were for this reason excluded in our review [3].Most importantly, the four included reviews on dietary behavior in the meta-analysis of Perski et al. lacked specific details on ESM design characteristics or methodological implication of ESM as alternative dietary assessment method.Still, the potential of ESM to obtain more accurate and reliable dietary data is highlighted together with the need for proper validation.
Altogether, the lacking details on important methodological aspects of ESM hinders drawing conclusions on common practices for implementation of ESM for quantitative dietary assessment.Nevertheless, Perski et al. emphasize the need for more elaboration on the methodological aspects in order to provide a summary of best practices on implementation of ESM for specific health behaviors including dietary behavior [60].Our scoping review meets this need with key methodological recommendations for developing an experience sampling dietary assessment method for total dietary intake next to elaboration on commonly applied ESM design characteristics.

Limitations and strengths
An important limitation of this scoping review is, inherent to scoping reviews, the less rigor search strategy and screening process.This will have resulted in an incomplete overview of studies describing ESM for dietary assessment.Still, this review has not the aim to assess outcomes of studies but rather evaluate how ESM can be applied for dietary assessment methodologically.Therefore, its strength lies in the assessment and description of ESM approaches specifically to provide insight in its use for quantitative dietary assessment as an alternative method for the traditional dietary assessment methods.To our knowledge, this has only been performed by Schembre et al. previously [12].However, our scoping review is, to our knowledge, the first to describe practical recommendations for developing an ESM for total dietary assessment (i.e.all food groups).Additionally, only two studies were identified to have applied ESM for total dietary assessment.Consequently, limited evidence-based information was available in literature on the development of ESM characteristics (prompting schedule, duration, questionnaire design) for quantitative dietary assessment of total dietary intake.Nevertheless, studies on qualitative and semi-quantitative dietary assessment using ESM were described and form, together with the guidelines of Myin-Germeys et al., the base of practical guidelines of designing an ESM protocol for quantitative dietary assessment of total dietary intake.To our knowledge, this review is the first to discuss recommendations on the implementation of ESM for quantitative dietary assessment as an alternative for traditional dietary assessment methods.

Conclusions
This review shows that ESM is increasingly being applied in research to measure dietary intake.However, few studies applied ESM to assess total dietary intake quantitatively with the same purpose of traditional dietary assessment methods.Still, the methodological characteristics of ESM show auspicious possibilities to overcome limitations of the classic dietary assessment methods.This paper provides guidance and is the starting point for the development of an Experience Sampling Dietary Assessment Method to assess total dietary intake quantitatively based on recent literature and theoretical background.Thorough evaluation and validation studies are needed to test the full potential of ESM as a feasible and accurate alternative for traditional dietary assessment methods.

Fig. 1
Fig. 1 PRISMA flow diagram of the screening and selection process

Table 1
Inclusion and exclusion criteria of the scoping review adapted from the protocol following initial searches

Table 2
Methodological considerations of included studies a which applied ESM for qualitative dietary assessment

Table 2 (
continued) Fixed signalling: prompt messages sent at predefined time(s), SR Semi-random signalling: prompt messages sent at random moments during predefined time windows, EC Event contingent sampling: self-initiated registering/responding when predefined event occurs, EOTDQ End-of-the-day-questionnaire: questionnaire sent at the end of the day at predefined time (often in combination with signalling techniques) a Manuscripts describing research based on the same study protocol were aggregated and described as one ESM study protocol in the table b Signalling techniques commonly applied in Experience Sampling protocols: R Random signalling: completely random timing of prompt messages between pre-defined (waking) hours, F

Table 3
Methodological considerations of included studies a which applied ESM for semi-quantitative dietary assessment

Table 3
(continued) a Manuscripts describing research based on the same study protocol were aggregated and described as one ESM study protocol in the table b Signalling techniques commonly applied in Experience Sampling protocols: R Random signalling: completely random timing of prompt messages between pre-defined (waking) hours, F Fixed signalling: prompt messages sent at predefined time(s), SR Semi-random signalling: prompt messages sent at random moments during predefined time windows, EC Event contingent sampling: self-initiated registering/responding when predefined event occurs, EOTDQ End-of-the-day-questionnaire: questionnaire sent at the end of the day at predefined time (often in combination with signalling techniques)

Table 4
Methodological considerations of included studies a which applied ESM for quantitative dietary assessmentManuscripts describing research based on the same study protocol were aggregated and described as one ESM study protocol in the table a b Signalling techniques commonly applied in Experience Sampling protocols:

Table 4
[34]tinued) Forman et al. used prompts which requested to report snack intake between the last prompt of the previous day and falling asleep and between waking up and receiving the first prompt[49].The study of Bruening et al. combined both real-time prompts, to report what participants were doing the moment before receiving the prompt, and retrospective prompts to report what they were doing the past 3 h[34].