Occurrence of and Reasons for “Missing Events” in Mobile Dietary Assessments: Results From Three Event-Based Ecological Momentary Assessment Studies

• DOI: 10.2196/15430
• Publication Date: 2020-08-04T01:24:01Z
• Authors: Katrin Ziesemer; Laura Maria König; Carol Jo Boushey; Karoline Villinger; Deborah Ronja Wahl; Simon Butscher; Jens Müller; Harald Reiterer; Harald Thomas Schupp; Britta Renner
• Keywords: 0301 basic medicine; Technology; Missing events; 501002 Angewandte Psychologie; Ecological Momentary Assessment; dietary assessment; diet records; mobile phone; mobile applications; technology; adherence; compliance; missing events; Ecological Momentary Assessment; mHealth; Diet records; Information technology; MHealth; 03 medical and health sciences; Mobile applications; 501002 Applied psychology; Surveys and Questionnaires; Humans; Retrospective Studies; 2. Zero hunger; Original Paper; T58.5-58.64; Mobile Applications; Diet; Adherence; Dietary assessment; Public aspects of medicine; RA1-1270; Mobile phone; info:eu-repo/classification/ddc/004; Compliance

Background Establishing a methodology for assessing nutritional behavior comprehensively and accurately poses great challenge. Mobile technologies such as mobile image-based food recording apps enable eating events to be assessed in the moment real time, thereby reducing memory biases inherent retrospective records. However, users might find it challenging take images of they consume at every event over an extended period, which lead incomplete records (missing events). Objective Analyzing data from 3 studies that used varied their technical enrichment, this study aims assess how often (meals snacks) were missed period 8 days naturalistic setting by comparing number recorded with normative expected events, recollections missing events. Methods Participants event-based Ecological Momentary Assessment (EMA) using dietary assessments asked record all (study 1, N=38, 1070 events; 2, N=35, 934 3, N=110, 3469 Study 1 basic app; 2 included fixed reminder possibility add meals after actual occurred instead (addendum); reminders, addendum feature, option skipped meals. The recalled reasons semistructured interviews EMA (studies 2) daily questionnaires 3). Results Overall, 183 participants reported 5473 Although momentary adherence rate indexed comparison was generally high across studies, differential pattern results emerged higher logged more technically intensive 3. Multilevel models logging trajectories showed significant, albeit small, decline time (b=−.11 −.14, Ps<.001, pseudo-R²=0.04-0.06), mainly because drop snacks between 2. Intraclass coefficients indicated 38% or less observed variance individual differences. most common competing activities issues, whereas situational barriers important. Conclusions Three different indicators (normative, stability, events) consistently given nature diet protocols, effect sizes rather small remarkably stable. Moreover, individual’s state context seemed exert greater influence on rates than stable differences, emphasizes need nuanced understanding factors affect adherence.