Ecological Momentary Assessment in Eating Disorders Research: A Qualitative Examination of Participant Experience and Recommendations for Future Studies

ABSTRACT

Objective

Ecological momentary assessment (EMA) is a widely‐used research method for investigating temporal relationships among eating disorder (ED) symptoms. Though EMA has many methodological advantages (e.g., reducing retrospective recall bias), little is known about the experience and effects of participating in this type of study from the perspective of individuals with EDs. The present study aimed to examine the experience of participants with EDs after completing an EMA study, with the goal of elucidating potential positive and negative effects of EMA methodology.

Method

A heterogeneous sample of participants with EDs (N = 192) completed clinical interviews, questionnaires, and an EMA protocol (five surveys/day for 14 days). A subsample of these participants (n = 16) completed a qualitative interview exploring their experience participating in the study. A reflexive thematic analysis was conducted using Nvivo software.

Results

The following themes were identified: (1) Self‐awareness, mindfulness, and reflection; (2) Behavioral change; (3) Rewarding aspects of the study; (4) Challenging aspects of the study; (5) Study design (including facilitators and barriers to participating); and (6) Suggestions for future studies.

Discussion

Although participants reported some challenging aspects of the study, most described their experience as positive (or at least neutral), and many noted direct benefits of participating. Future EMA research may benefit from integrating the perspectives of those with lived experience into study design, potentially reducing participant burden, improving the quantity and quality of data collected, and increasing benefits for participants.

Summary.

• This qualitative study investigated participants' experience of an ecological momentary assessment study.

• Results indicate that participating in this type of research can be both challenging and beneficial.

• Researchers should consider factors that mitigate negative effects and promote adherence and therapeutic benefit when designing research studies.

• Individuals with lived experience of both having an eating disorder and being a research participant have a valuable perspective that can help improve study design.

Ecological momentary assessment (EMA) is a research methodology permitting repeated measurement of constructs in daily life (Shiffman et al. 2008). Modern EMA studies have participants record experiences via their mobile devices (Zapata‐Lamana et al. 2020) after an event occurs (i.e., event‐contingent) and/or in response to periodic prompts (i.e., signal‐contingent). EMA studies have exploded in recent years, with EMA being used to study various psychological processes across non‐clinical and clinical populations, including eating disorders (EDs). EMA has the potential to significantly improve understanding of EDs (e.g., via examining temporal associations between symptoms) and can be harnessed as a tool to facilitate and inform treatment (Engel et al. 2016; Levinson et al. 2024; Smith and Juarascio 2019; Smith et al. 2019).

EMA has several advantages, including limiting retrospective bias and increasing ecological validity (for a review, see Doherty et al. 2020). These methodological advantages have contributed to enthusiasm for EMA among researchers. However, it is important to consider how this methodology is experienced by participants. Some older studies using mostly paper‐and‐pencil monitoring forms for recording ED and related symptoms as they occur found that this methodology was generally acceptable to participants experiencing eating pathology (Deaver et al. 2003; Redlin et al. 2002; Stein and Corte 2003; Stickney et al. 1999). However, these studies may not accurately reflect the experience of participating in modern EMA protocols. A study by Heron et al. (2023) found that women experiencing binge eating found an app‐based EMA protocol to be acceptable and feasible, assessed using quantitative measures. They also asked for participant feedback on instructional materials, specific EMA questions, and study procedure using an exit interview. While this study provides valuable guidance for study and item design, the interview used was fairly structured and did not focus on the perceived impact of participating in the study on the participant and their symptoms. Several recent qualitative studies have sought to explore participant experience in EMA research more broadly in populations including unhoused youth, veterans reporting non‐suicidal self‐injury, and adults experiencing chronic cancer‐related fatigue (Acorda et al. 2021; Bootsma et al. 2022; Gromatsky et al. 2022). Participants generally reported an overall positive experience in EMA research, with increased awareness being a commonly cited benefit of participation, though negative effects (e.g., disruptions to usual activities) and challenges (e.g., issues with technology, lack of personalized questions) were also noted (Acorda et al. 2021; Bootsma et al. 2022; Gromatsky et al. 2022). These studies also gathered recommendations from participants regarding improvement of future EMA study design.

Taking into consideration participant experience in the design of EMA studies may help reduce burden and negative impact on participants while potentially increasing the quality and quantity of data collected. EMA participation may be particularly beneficial for individuals with EDs given previously reported effects of EMA in increasing awareness (Gromatsky et al. 2022), as insight and emotional awareness have been identified as potential treatment targets in this population (Konstantakopoulos et al. 2011; Westwood et al. 2017). More broadly, difficulties with participant recruitment and retention in ED research suggest the need to integrate lived experience perspectives into ED research design (Ortiz et al. 2020). As such, this qualitative study investigated the experience of individuals with EDs who had previously participated in an EMA study.

1. Method

The consolidated criteria for reporting qualitative studies (COREQ; Tong et al. 2007) were used to guide the reporting of the current study.

1.1. EMA Study

Participants were drawn from an EMA study examining relationships among stress, emotion regulation, and ED symptoms (see https://osf.io/7e6ct/). As part of this study, 192 women with EDs were recruited via online advertisements and flyers in the community. Inclusion criteria were residing in Canada, identifying as a woman or assigned female at birth, having a North American phone number, and having a smartphone data plan to complete EMA surveys. Exclusion criteria included current mania and/or psychosis. The study was comprised of virtually administered semi‐structured clinical interviews (used to assess ED diagnosis and exclusion criteria), an online questionnaire battery, and EMA.

During the EMA protocol, participants received five semi‐random Qualtrics surveys to their mobile device via SurveySignal (Hofmann and Patel 2015) each day for 14 consecutive days. Survey completion took approximately 5 min, with length depending on participant responses. Surveys were sent within time periods chosen by participants based on their sleep–wake cycles (Early: 8 am –10 pm, Middle: 9 am–11 pm, Late: 10 am–12 am). Surveys queried about ED symptoms, mood, impulsive behaviors, recent stressful events, and use of emotion regulation strategies since the last prompt. Participants were also asked to initiate a survey each time they ate; this survey took approximately 2 min. Event‐contingent survey questions pertained to the type of eating episode and reasons for over/under‐eating. Participants received up to $150CAD for their participation. Specifically, participants received $38 for completing clinical interviews and questionnaires and $1/prompt plus a $3/day bonus if they completed more than 80% of the day's prompts. Participants who did not complete a minimum of three prompts per day for three consecutive days or for a total of 5 days were discontinued from the study (n = 22).

1.2. Participants

A subset of participants from the EMA study (n = 29) who had agreed to be recontacted were invited via email to participate in the qualitative study described herein. Though the original study included non‐ED participants, only those with an ED were recruited for the present study. Participants were purposively selected using a maximum variation strategy (i.e., selecting for diversity in ED diagnosis, age, education, ED treatment history, and survey compliance rate; Coyne 1997; Marshall 1996). The present sample was broadly representative of the original study sample; please refer to File S1 comparing demographic characteristics of the larger sample to those of the subsample who participated in the qualitative study.

Thirteen participants either did not respond to emails or did not attend the interview; none withdrew during or after the interview. The final sample included 16 participants. The average survey compliance rate in this sub‐sample from the EMA study was 81.61% (SD = 13.43). Please refer to Table 1 for demographic information.

TABLE 1.

Sociodemographic characteristics of participants.

Characteristic	M (SD)/n (%)
Age (years)	29.9 (8.4)
Diagnosis
Anorexia nervosa	4 (25.0)
Bulimia nervosa	6 (37.5)
Binge eating disorder	2 (12.5)
Avoidant/restrictive food intake disorder	2 (12.5)
Low frequency bulimia nervosa/binge eating disorder	2 (12.5)
Gender
Female	15 (93.8)
Gender non‐binary/Genderfluid	1 (6.2)
Sexual orientation
Bisexual	2 (12.5)
Straight	8 (50.0)
Lesbian	1 (6.3)
Pansexual	1 (6.3)
Queer	3 (18.8)
Questioning or unsure	1 (6.3)
Ethnicity
White	13 (81.3)
Filipino	1 (6.3)
South Asian	1 (6.3)
Other	1 (6.3)
Highest level of education
High school	1 (6.3)
Some university (at least 1 year)	5 (31.3)
College certificate/diploma	2 (12.5)
Bachelor's degree	6 (37.5)
Master's degree	2 (12.5)
History of eating disorder treatment
Current	3 (18.8)
Past	5 (31.3)
Never	8 (50.0)

1.3. Procedure

The present study is part of a larger qualitative project exploring: (1) participants' experiences with EMA (present manuscript) and (2) their attitudes towards receiving EMA data as a personalized single‐session intervention (Lapadat et al. unpublished manuscript). The project received ethical approval from McGill University's Research Ethics Board‐II (#22‐07‐016). On average, the time between participation in the EMA study and the qualitative study was 7.7 months (SD = 4.8, Range = 1.6–15.1 months).

Interviews were conducted online as participants were located across Canada. Prior to starting the interview, informed consent was obtained. Interviews were conducted individually and lasted 1–1.5 h. Participants received $30 CAD as compensation. Authors 1, 2, and 3 (all women) completed the interviews and post‐interview comment forms detailing initial reactions. Author 1 was a postdoctoral fellow and clinical psychologist with expertise in EDs. Author 2 was a graduate student in clinical psychology with prior qualitative research experience. Author 3 was a post‐baccalaureate research coordinator with several years of experience conducting ED research. Authors met several times prior to the start of recruitment to discuss the aims of the study, biases and assumptions, and to develop the study protocol and semi‐structured interview questions based on the EMA study, identified gaps in the literature, etc. Interview prompts included: “To start, I would like you to tell me what you remember about the past study. Please share anything that comes to mind.” and “Now I would like to hear about your personal experience participating in this study. What was it like for you to complete the study?” Please refer to File S2 for the complete interview protocol.

Interviews were audio‐ and video‐recorded, and the audio was uploaded to Descript software for transcription. Trained undergraduate research assistants reviewed transcripts to ensure accuracy, remove potentially identifiable information (e.g., treatment center names), and add notation (for pauses, laughter, etc.). Transcripts were not given to participants for comment; however, participants were re‐contacted and offered the opportunity to meet with the interviewer for a presentation of the study's main findings. Those that participated in this meeting (n = 5; 31.25%) reported that results were generally consistent with their experience.

1.4. Data Analysis

Interview data were analyzed in NVivo 14 using reflexive thematic analysis, as outlined in Braun and Clarke (2006). Analysis was inductive, or data driven, meaning codes and themes were created based on understanding of the data, rather than in an attempt to test an existing theoretical framework (Braun and Clarke 2006). Codes were independently generated from the transcripts, defined as the smallest element of the data that can be interpreted in a meaningful way. Next, codes were grouped into subthemes and themes, which were iteratively reviewed and refined by Authors 1 and 2. Themes were derived at a semantic level. Thematic analysis was conducted within a realist/essentialist approach, aiming to accurately capture participants' objective realities as directly expressed in their responses. Recruitment was stopped when saturation was reached based on consensus among the interviewers.

2. Results

Six themes and 23 sub‐themes were identified in the data. Exemplary quotes are reported in Table 2.

TABLE 2.

Themes, sub‐themes, frequencies, and exemplary quotes.

Themes/sub‐themes	Frequency (n)	Exemplary quotes
Theme 1: Self‐awareness, mindfulness, and reflection (n = 16)
Increased self‐awareness	15	“So I was kind of like seeing the up and down of my own moods and how that could possibly relate to the eating and the exercise and the other habit related questions that might have been asked in the survey.”—Serena “I think with the stress levels, like, I don't know, sometimes I wasn't really realizing the correlations with stress and [ED] behaviors and stuff. And like, I feel like actually putting the two together like that did kind of make me more aware of that.”—Claire
Increased mindfulness and “checking in”	11	“So it was helpful to kind of just be forced to take those pauses in the day, even if it was just like one minute to just have like a check in almost of like, how am I feeling”—Patricia “The survey […] allowed me to just live in the moment of those experiences. And so that was good for me. Yeah. The check‐ins right after eating, it's like staying present and allowing for, those ups and downs to happen without holding onto them.”—Savannah “I was in the middle of a recovery or treatment program while I was doing it. It really helped me to kind of stay on track and be accountable for my eating. And yeah, just really reminding myself to have meals and […] I would usually have like a food diary or a journal pretty similar to what the survey was asking.”—Tanya
Negative consequences of increased self‐ awareness and mindfulness	7	“Maybe I would feel a little negative if I had like an unhealthy eating episode and I would definitely reflect on it a lot more, just thinking about the way it made me feel and how I feel after”—Charlotte “I kind of went into it [with] the mindset, ‘oh, well you're tracking my binge eating’. So, I was kind of hyper‐aware of my eating more so than I would have been before.”—Jay “Like it's helpful because now you're more cognizant of it. But what are you doing with that increased cognition of those behaviors? Like are you making changes that are gonna be positive or is this gonna bring deleterious effects? Are you gonna now restrict more or something like that?”—Darcy
Theme 2: Behavioral change (n = 14)
Accountability	5	“I have a meal plan, so staying accountable to eating what is on my meal plan. So if I didn't have something to where I had to write down and say, yes, I ate this, or no, I did not, I'm more likely to skip it and then not forget about the fact that I skipped it, but be like, ‘oh, no one's gonna hold me responsible for not eating that only really myself’. Whereas I found it helpful that I had another, you know, quote unquote third party that I was [accountable to] and it wasn't like I was actually held accountable by the study, but it felt that way and that helped.”—Abby “Accountability for me, it's not supposed to be punishing. Especially if I'm holding myself accountable and nobody else is like holding me accountable or whatever. It's giving myself autonomy to, you know, choose the right actions for me. Um and so with this, it was holding me accountable in like, at least trying to break some habits I had been forming with my eating disorder and symptoms.”—Kylo
Changes to eating habits/mental health practices during study participation	4	“It kept me on a regular schedule, whereas before my eating would be kind of sporadic, more so towards the end of the day. Whereas with this, I was consciously making an effort to have a morning meal, maybe a snack after, you know, between lunch and morning and then I'd have lunch. And so it kept me on a set schedule, which was nice.”—Jay “I did notice, like I was trying to binge less. I think it's funny cause I say I was trying to binge less even [though] I already tried to binge less, but like, I was slightly more successful because I was like, ‘I have to like admit to it.’”—Belle
Long(er)‐term change	7	Long‐term change: “I'm still trying to eat more consistently, so I've been trying ever since that time. Like, what was it, March or something you said? Yeah, since that time I've been trying to eat more consistently throughout the day.”—Kylo Change lasting beyond study, but not maintained: “For a week or so, I was still doing the regular meals like throughout the day, the set schedule. But again, it started to taper off towards the end when it was just back to my regular habits.”—Jay
Changes during participation due to burden	4	“I don't think my eating habits changed that significantly. It was just more like, instead of eating a handful of nuts every time I walk by the kitchen, I was like, ‘Okay, let me get a bowl and like be a civilized human about this and like have this counted [as] one entry.’”—Darcy
Potentially reinforcing ED behaviors	2	“But otherwise I don't even notice that if I'm not working out today. And does it even matter? But it did matter to me when I was like, jotting it down and keeping a note of all my activities. I thought that if I'm not aggressively working out, maybe I should go for a walk or do something, because I haven't done anything.”—Kayla
Theme 3: Rewarding aspects of the study (n = 14)
Gave space, catalyzed change, and reinforced recovery	8	“And I don't know [it] felt like it was fun to not, fun is not the word, but it felt like…, like, um, validating to send my [response] because it was like, I have this big taboo, but now I'm just giving it to strangers.”—Miguel “But I think that [doing the study] still like, put a little […] seed, like a kernel. Of like openness and […] so I think that helped.”—Miguel “Taking part in this was important to me personally because like […] I haven't seen, a mental health professional or received any explicit support around eating or disordered eating or body image. […] and it was something I wanted to learn more about myself. And I think the explicit conversations and the interview and then the follow‐up sort of two week survey check‐in period reinforced for me, you know, yes. […] And over time, it wasn't like immediately after the study, but you know, over that summer, like so over the past summer I was able to take those little steps to, you know, change the habits that were or like make adjustments to the habits that were, you know, less helpful for me and like contributing to, some more difficulties.”—Serena “It was fun and a little bit, you know, rewarding to share my experience about this. Cause of course, […] eating disorders aren't something that's typically spoken very openly about or shared very openly about.”—Gabriella “Doing the surveys, for example, really helped to remind me about the strategies and techniques I learned from the treatment program that I was in.”—Tanya
Helping others through participation in research	9	“I also felt good cause […] even though this isn't like a counseling session or a thing that's necessarily benefiting me directly, I hope that the research can contribute towards helping other people. And so that feels good to me.”—Savannah “I tried not to like change anything, [as] I feel like that would've made the results more like skewed or whatever. And I tried to be as honest as possible. I feel like with surveys it's easy for people to be like, ‘oh, like I don't wanna admit to doing something, so I'm just gonna say that I didn't do it’. But I tried not to do that as best I could?”—Claire
Theme 4: Challenges aspects of the study (n = 15)
EMA prompts—Annoying, repetitive, interfering	11	“So it really felt like for that two weeks, like, my whole life was revolving around like, ‘okay, well we gotta make sure that the volume is on my phone’ and, ‘oh no, we shouldn't go there cuz we're gonna be like unavailable for two hours and I don't wanna miss a prompt.’ And that was a lot. So yeah, it was kind of a struggle and I think in the end I like just squeaked in, like I had one prompt that I could miss. And so I was like, glued to my phone because I was like, ‘I don't want to have done, you know, 13 days of work and then like lose it on the last day’ (laughs)”—Patricia “I was talking to my sister a bit about it and she's a nurse, so she works like 12 h shifts and she was like, ‘I wouldn't have even been able to participate at all.’ Because… you know, she has the type of job where you can't [take a break], you know. I work a desk job, so it was…fine for me most of the time to, you know, just take a two minute break from work. But, so I wonder like… what kind of… people maybe are not… participating? Like what sort of demographics I guess, is maybe getting missed because of how… many prompts there are during the day.”—Patricia
EMA prompts—Elicit negative emotions	7	“Yeah. You know, like I mentioned, I think the reporting on eating disorder behaviors, it's just really embarrassing and there's like, there's that guilt and shame associated with it.”—Gabriella “And sometimes the questions did make me think more about my body image and, where I was at. It made me think about it more, put focus on it, which also kind of did stress me out a bit […]. I remember there was one point where […] it came up and then I got the notification or whatever, the text, and then afterwards filling it out, I would like stare at myself in the mirror. So […] it could be triggering. It was a bit triggering for me? I don't know. But not all the time. Just occasionally.”—Tanya
EMA prompts—Negative effects of missing prompts	10	“Especially cuz I think it was like if you missed more than like two [EMA prompts] in a row or something or I remember if you missed a certain amount they would like kick you out ((laughs)). So, I was really paranoid about that.”—Claire “I think I maybe missed one [EMA prompt] just being like, ‘oh darn’. So yeah, a little disappointing, like [disappointed] in myself about not doing it as well as I possibly could.”—Poppy
Difficulties with interviews and questionnaires	7	“I definitely was a little uncomfortable talking in detail about my eating episodes and like what we would eat and, just the detail definitely would give me some negative thoughts about my eating habits and it's just a little embarrassing, I guess.”—Charlotte
Full picture not captured	4	“What I wish we had been able to do was just explain what we were saying from beginning to end. Rather than just, ‘oh, these questions’. Because [for] a lot of them [it was like] ‘well, I did this, but in a different way?’”—Rosalind “I think obviously also my perspective is biased by my own lived experiences and working in research and everything else. But […] Sometimes it's harder because I think [about] the standardized measures that we often use, sometimes don't capture [everything], they capture a lot, but sometimes by like trying to shove people into boxes, we like miss things or there's nuance that doesn't fit in the box.”—Darcy
Simultaneously harmful and beneficial	5	“And it's hard because you almost want to objectively get the data, but the only kind of way that you can do that is by self‐report. […] you're kind of in a catch‐22 that like, okay, if the only way we can get this is self‐report, but then self‐report is gonna make the person uncomfortable. Like, how do we get around that? And that is really hard because how can you get around that?”—Darcy
Theme 5: Study design (n = 16)
Facilitators	12	“As much as I found the questions repetitive, I did like how I think a lot of different feelings were encompassed in the questions. […] I don't know if I've seen it in other studies before or online somewhere, but the questions that I have had been asked before, seen before just weren't, never seemed [to] quite get to maybe the root of how I was feeling, and I thought a lot of the questions on this survey did. And I remember going through it the first time and being like, ‘well, that's the first time that someone has actually asked me that’.”—Abby
Barriers	13	“Sometimes I didn't know what the word meant and I was like, ‘uh’ and I just guessed ((laughs)). There was a couple of them, not gonna lie, that I had no idea what that word meant, or like, some of them were really similar and I was like, ‘Okay?’”—Claire “After more time went on, I became lazier. (laughs) Like at first I was like, ‘No, I'm gonna log everything. It's fine.’ And then I was like, ‘Ugh.’ Like probably after the [one] week mark?”—Darcy
Study‐ED interaction	8	“I guess I've done a lot of surveys and stuff and again with treatment, you have to do all that intake stuff and history. So […] it's a bit old hat. I'm used to it.”—Poppy “Just only a few months ago I was very deep into my eating disorder. And so I think definitely it would've been more triggering for me. Having to complete the surveys and also just having to be reminded constantly about my body image, my feelings, food. I think it definitely would've put me in a different situation. I don't know if it would have, been helpful for me to participate at that time.”—Tanya “But I can imagine for other people who are, like, very in the depths of their eating disorder, that it might be triggering or harder to kind of engage with that and have it so top of mind.”—Darcy
Theme 6: Suggestions for future studies (n = 13)
Reducing burden of EMA prompts	9	“I think maybe having a bigger window of time to be able to answer it, even if it was like an hour instead of half an hour. But I mean, I'm sure that there's good reasons that it was chosen to be half an hour. […] Even if the times were more set […] you know, maybe one is gonna come at nine o'clock and one is gonna come at noon, and two, like, if you had exact times, maybe that would help?”—Patricia “Or like if you miss one, maybe like if someone misses one, there's like an extra one at the end of the day to make it up. But only if you missed it.”—Claire “Maybe if there was an option to take one day off and do it another day. Like, so if you could turn tracking off for one day and then resume for another extra day.”—Jay
Facilitating accurate responding	8	“I think if there was something in the question, like in brackets, like, “By the way, this is what we mean.” Like something really, really quick. Just because it's 14 days, which is a long time for someone with ADHD brain to process and remember that information.”—Belle “I wish there was an option to maybe update how you were feeling in between the normal survey times as well? So maybe, you know, five minutes ago I felt awful about my body, but then maybe half an hour later before I got another survey, I started to feel a little bit better. So, I wish there was a way where you could actively track [how you were feeling] throughout the day as well on your own.”—Jay
Other areas worthy of exploration	6	“And so [its] just really interesting to see that so many women with eating disorders have ADHD and now this isn't the purpose of the study, ((sighs)) but maybe looking at some of the other things that are comorbid with eating disorders because I think one of the biggest challenges I've had in recovering from the eating disorder is that it's not just because it's this eating disorder, not just like, yeah, there's trauma in my past, but it's like, oh, my brain just operates in this way, that it gets hyper fixated and requires these dopamine hits.”—Gabriella
Reducing potential harm	5	“Yeah. So, if like I got my survey and it asked like, ‘have you engaged in a self‐care activity’? And I put down, ‘no’. It might prompt me like, ‘oh, I'm gonna be getting another survey in the next like four hours, maybe I should engage in a self‐care activity.’”—Poppy

Note: Quotes have been edited for clarity (e.g., removing words like “um,” word repetition, etc.). Names provided are pseudonyms selected by either the participant or the interviewer (depending on participant preference).

2.1. Theme 1: Self‐Awareness, Mindfulness, and Reflection (n = 16)

Participants noted that participating in the study helped them become more self‐aware, mindful (i.e., present in the moment), and gain a better understanding of themselves and their ED.

2.1.1. Experience of Increased Self‐Awareness (n = 15)

Nearly all participants described experiencing increased self‐awareness with study participation, citing higher awareness of ED behaviors and emotions. Participants generally described being more self‐aware as positive and beneficial; for example, noting the value of increased awareness of symptoms that had become automatic. Further, participants described realizations they made due to increased awareness, for example, noticing links between emotional state and ED behaviors.

2.1.2. Increased Mindfulness and “Checking in” (n = 11)

Participants described the prompts as an opportunity to be mindful and ‘check‐in’ with themselves, particularly their emotions. Benefits of mindfulness were noted, including helping to reflect, change perspective, and take note of progress. Comparisons were made between this effect of the prompts and different apps and therapeutic interventions (e.g., food journals).

2.1.3. Negative Consequences of Increased Self‐Awareness and Mindfulness (n = 7)

Downsides to increased self‐awareness and mindfulness were also reported, such as feeling negatively or ruminating when aware of certain behaviors. It was also noted that increased self‐awareness could potentially lead to behavioral change, which could either be adaptive or maladaptive (i.e., used in service of perpetuating the ED).

2.2. Theme 2: Behavioral Change (n = 14)

Participants reported different types of (mostly short‐term) behavioral change during and following study participation.

2.2.1. Accountability (n = 5)

Participants indicated that reporting on behaviors throughout the day increased their sense of accountability to their goals, ED recovery, and/or the research project.

2.2.2. Changes to Eating Habits/Mental Health Practices During Study Participation (n = 4)

During the EMA phase of the study, participants across the diagnostic spectrum noted that they ate more consistently, which was often attributed to the prompts serving as reminders to eat or as a source of structure for eating. Some participants also noted that they were less likely to restrict or binge eat during the study.

2.2.3. Long(er)‐Term Change (n = 7)

Behavioral changes that were maintained (for at least a short period) past the end of the study or were initiated following participation were also reported. These included eating more regularly, reduced binge eating, and seeking out hobbies to reduce negative emotions.

2.2.4. Changes due to Burden (n = 4)

Participants noted that completing the prompts became more difficult over time and that they sometimes changed their behavior (e.g., skipping a snack, having a larger snack rather than “grazing,” not binge eating) to avoid having to complete a prompt or to prevent the experience of negative emotions (e.g., guilt, shame).

2.2.5. Potentially Reinforcing ED Behaviors (n = 2)

Two participants reported that completing the prompts helped them to engage in “healthier” behaviors (e.g., increasing exercise, eating more vegetables). For one participant, it appeared that these “healthier” behaviors were actually in line with the ED.

2.3. Theme 3: Rewarding Aspects of the Study (n = 14)

In addition to increased self‐awareness and some behavior change, participants reported other rewarding elements associated with study participation. Participants broadly described their study participation positively or neutrally.

2.3.1. Gave Space, Catalyzed Change, and Reinforced Recovery (n = 8)

Participants reported that the study (interviews and EMA portion) provided them with a non‐judgmental space in which they could talk about their ED, something they noted as being difficult to come by in everyday life. Participants also described participating in the study as supportive of recovery, viewing their experience as a catalyst for change (e.g., prompting them to contact a health care provider) and/or as reinforcing recovery maintenance.

2.3.2. Helping Others Through Participation in Research (n = 9)

Participants described benefits to society of participating in research studies and how “giving back” made them feel good, reporting a strong desire to help other people with similar difficulties. They indicated that it felt good to participate in the study to help others and to advance understanding of EDs, even if their participation would not directly benefit them. Further, participants expressed belief that this research could help improve treatment, supplement care for those without current access, and reduce stigma surrounding EDs. Finally, participants reported preoccupation regarding the quality of research data, noting that they made specific efforts to not “skew” results or expressing concern that they had unintentionally given inaccurate information (due to the influence of self‐monitoring on behavior).

2.4. Theme 4: Challenging Aspects of the Study (n = 15)

Participants reported several negative consequences of participation, including negative effects of the study procedures on their emotional state and day‐to‐day lives.

2.4.1. EMA Prompts—Annoying, Repetitive, Interfering (n = 11)

Participants reported feeling like their lives revolved around the EMA prompts and that they had to change their habits (e.g., wake up early, take extra breaks at work, miss family time) to complete them. Prompts were described as repetitive and time‐consuming. It was also noted that, due to the requirements of EMA study design (e.g., the ability to check one's phone throughout the day), research may exclude certain groups.

2.4.2. EMA Prompts—Elicit Negative Emotions (n = 7)

Completing the prompts/reporting certain behaviors elicited negative emotions and were confronting for some. One participant indicated that completing the prompts brought up mixed emotions about her recovery.

2.4.3. EMA Prompts—Negative Effects of Missing Prompts (n = 10)

Missing prompts elicited negative emotions (e.g., disappointment with self) and worry about being “kicked out” for missing prompts. Participants also noted real‐world obstacles (e.g., work meetings) that resulted in missing prompts.

2.4.4. Difficulties With Interviews and Questionnaires (n = 7)

The most commonly mentioned negatives associated with the interviews and questionnaires were: (1) interviews were sometimes confronting, as they forced participants to talk about things they usually avoid talking about, and (2) interviews and questionnaires were lengthy.

2.4.5. Full Picture Not Captured (n = 4)

Participants described the limitations of standardized assessments, indicating that questions posed by EMA prompts, on questionnaires, and in the interviews were not able to fully capture the nuances of their ED or experiences. The fact that the prompts could not always capture fluctuations/experiences between prompts was also noted.

2.4.6. Simultaneously Harmful and Beneficial (n = 5)

Participants remarked that study participation (and any ED intervention) is likely to simultaneously elicit positive and negative effects. Conversely, several participants indicated that the study procedures had no effect on them whatsoever.

2.5. Theme 5: Study Design (n = 16)

Participants identified aspects of the study that either facilitated or hindered participation. State of mind, phase of recovery, and prior experiences as someone with an ED were described as influencing the experience of the study procedures.

2.5.1. Facilitators (n = 12)

Aspects of the study that were identified as having facilitated participation included: financial compensation (described as fair and motivating); explicit and well‐communicated commitment to confidentiality; researcher kindness; and clarity of study procedures. Participants also indicated that EMA prompts were generally easy and feasible to complete, that the questions were thorough and assessed a variety of relevant experiences, and that having a breadth of options to choose from when responding to EMA questions (e.g., emotion adjectives) also facilitated participation.

2.5.2. Barriers (n = 13)

Most identified barriers were related to the EMA prompts and included: forgetting the definitions of some words (e.g., distinctions between emotions) and uncertainty regarding how to answer some questions (e.g., if the question did not apply to them).

2.5.3. Study‐ED Interaction (n = 8)

Aspects of an individual's state of mind, phase of recovery, and prior experiences as someone with an ED were described as influencing study experience. Participants often imagined what participation would have been like at another time in their life. For example, participants indicated their experience of the study procedures may have been either more or less distressing depending on their mindset (e.g., depressed) or circumstances (e.g., undergoing treatment). Additionally, participants noted that study procedures were well tolerated or were experienced neutrally because they were accustomed to talking about their lives/EDs due to prior treatment or participation in other research studies, suggesting that they were unsure if this would have been the case in the past or for others without these experiences. Finally, participants explained how much more difficult they believe the study would have been if they had been in a more acute phase of their ED. It was also noted that the accuracy/honesty of responses may not have been the same when symptoms were most severe.

2.6. Theme 6: Suggestions for Future Studies (n = 13)

Almost all participants offered suggestions related to how the design of EMA research studies could be improved.

2.6.1. Reducing Burden of EMA Prompts (n = 9)

Suggestions designed to reduce participant burden during the EMA portion of the study included: allowing a larger window during which prompts can be completed; having prompts follow a set schedule instead of being semi‐random; giving the option to take a day off from the study; giving the option to “make up” a missed prompt at the end of the day; and establishing a cut‐off time for when prompts will be sent (not too early or too late in the day).

2.6.2. Facilitating Accurate Responding (n = 8)

Suggestions for improving the accuracy of how participants respond to questions were: greater precision in question formulation (e.g., define and differentiate constructs); addition of short‐answer questions; ability to initiate EMA surveys (beyond those related to eating episodes).

2.6.3. Other Areas Worthy of Exploration (n = 6)

Participants described elements that they felt could be relevant to better understanding EDs (e.g., developmental factors, comorbid conditions, coping strategies and strengths, hormones/menstrual cycle) and suggested including questions related to these factors for future studies.

2.6.4. Reducing Potential Harm (n = 5)

Actions that the researcher could take to ensure participants are not harmed during participation were suggested, such as checking in with participants to ensure they do not feel triggered by prompts. Participants also described ways in which coping strategies could be built into the prompts.

3. Discussion

In the present qualitative study, individuals with EDs described their experience of participating in an EMA study. The EMA protocol was generally well‐tolerated with few ill effects, consistent with past research in EDs (Heron et al. 2023) and other populations (Acorda et al. 2021; Bootsma et al. 2022; Gromatsky et al. 2022; Meglio et al. 2024). Participants also provided a wealth of ideas for improving EMA research design, which may benefit both participants and researchers.

Aligning with previous research, increased self‐awareness and mindfulness were the most consistently mentioned effects of study participation (Acorda et al. 2021; Bootsma et al. 2022; Gromatsky et al. 2022; Piot et al. 2022). Many participants identified specific behaviors or patterns of which they were previously unaware. Self‐monitoring is a core component of evidence‐based treatment for EDs (Fairburn 2008), and EMA, a form of self‐monitoring using modern technology, may have similar therapeutic benefit. While the acceptability of digital self‐monitoring methods depends on a complex array of individual factors (Lindgreen et al. 2018), integrating EMA into assessment and treatment of EDs may be advantageous, as this may assist with tracking progress or guiding intervention (Levinson et al. 2024; Piot et al. 2022). For example, EMA data have been used to create personalized conceptualizations to guide selection of treatment targets (Levinson et al. 2021) and to deliver just‐in‐time adaptive interventions (Juarascio et al. 2018). Present findings suggest that EMA study participation on its own may serve a similar function, allowing people to monitor progress and “take stock” of gains.

Regarding EMA study design, participants identified several elements that promoted or facilitated participation. Future studies should not overlook the importance of fair financial compensation, recognizing the level of commitment EMA studies require. Further, it is unlikely that participants would have viewed the study context as one of the rare spaces where they could openly talk about their ED if not for the researchers' commitment to kindness and respect and detailed explanation of confidentiality. Indeed, negative researcher demeanor has previously been noted as a barrier to ED research participation (Ortiz et al. 2020). Additionally, though difficulties with technology are a commonly cited barrier in EMA research (Acorda et al. 2021; Gromatsky et al. 2022), our participants reported the EMA survey platform was easy to use. This highlights the importance of software choice; however, it may also reflect a high familiarity with technology in this sample. Finally, the thoroughness of the questions was identified as something that facilitated participation. Feeling like one's experience is represented in the questions likely increased engagement. There is a recognized need for the establishment of uniform and psychometrically strong EMA questions (Levinson et al. 2024). In line with work highlighting the value of patient‐oriented research (Hart and Wade 2020; Obeid et al. 2020), it is also important to include those with lived experience in developing assessment materials to ensure that aspects relevant to patients are captured. Participants suggested the inclusion of questions assessing potential precipitants of disordered eating (e.g., childhood experiences, trauma), comorbid conditions (that may interact with ED symptoms), and biological factors (e.g., effects of hormonal cycles on eating and mood). Participants also suggested integration of greater flexibility into EMA protocols in the form of short‐answer questions and the ability to initiate EMA surveys. These recommendations echo calls for greater personalization in EMA research and ED research more broadly (Acorda et al. 2021; Meglio et al. 2024; Ortiz et al. 2020; Piot et al. 2022).

The burden of EMA research on participants should not be ignored. In particular, the semi‐random nature of EMA prompts was described as interfering, requiring changes to behavior (e.g., modifying sleep schedules, navigating personal obligations while monitoring phone for prompts). Previous studies have also found timing of EMA prompts to be a barrier to adherence (Meglio et al. 2024). Further, participants reported that missing prompts negatively impacted them (e.g., eliciting guilt). Participants offered suggestions for mitigating this difficult aspect, including having prompts follow a schedule and/or offering a way to “make up” missed prompts. As such, researchers should consider the benefits and costs of randomizing the timing of prompts.

The temporary experience of negative emotions was a commonly reported negative effect of EMA surveys. This is consistent with past research finding that asking about negative events can increase negative mood and stress (Labott et al. 2013) and so is unlikely to be specific to EMA methodology. Though not agreeable, the experience of negative emotions is likely preferable to a pattern of ignoring strong emotional states. This relates to the observation in the present study that any study participation and treatment may simultaneously elicit positive and negative effects. Finally, though noted only by a small number of participants, EMA participation may be used to reinforce ED behaviors, particularly in those with low insight. Several participants also imagined that their study experience, what they put into it, and how they would respond to increased self‐awareness would likely have been different at more acute stages of their ED. Whether different levels of insight and/or ED symptom severity increase risk for negative effects of EMA is an empirical question worthy of further exploration. However, previous studies have found that, although the perceived burden of EMA participation is somewhat higher in those with than without affective disorders (van Genugten et al. 2020), participation was not found to elicit suicidal urges or non‐suicidal self‐injury even in a high‐risk population (Gromatsky et al. 2022) and seemed to decrease some risky behaviors (e.g., substance use; Acorda et al. 2021).

Participants recommended that safeguards be built into EMA studies to mitigate potential harm, such as researcher “check‐ins” with participants. Some participants recommended asking about coping strategies as both a way to obtain a more complete psychological profile (rather than merely focusing on symptoms) and as a reminder to use tools that have been helpful previously. Others expressed a desire for the EMA platform to recommend coping strategies depending on symptoms endorsed. Even without this explicitly built‐in, some participants noted that completing EMA prompts served as a reminder to engage in adaptive behaviors, and for those who had previously accessed treatment, reinforced their progress and recovery. Though a balance must be struck between offering support and attempting to not influence data collected, recommendations for embedded support are consistent with previous research examining EMA study experience (Acorda et al. 2021; Gromatsky et al. 2022). Taken together, findings suggest an interest in ecological momentary interventions and just‐in‐time adaptive interventions (Juarascio et al. 2018) in individuals with EDs.

Strengths of the present study include asking a heterogeneous sample of individuals with EDs to describe their experience of participating in an EMA study, providing the vital perspective of those with lived experience on study design. Results should be interpreted cautiously given several limitations, including lack of gender and ethnic diversity, and possible retrospective bias (for some, EMA study participation occurred over a year prior). Further, results may not generalize to EMA studies using different methodologies (e.g., more intensive sampling, physiological assessments). Additionally, the present study omitted the perspectives of those who withdrew or who were discontinued from the study due to failure to complete the minimum number of prompts; these viewpoints may be especially valuable in providing insight into participation barriers.

Taken together, participants offered important suggestions regarding ways to improve EMA study design to increase potential benefits while reducing burden and mitigating potential harm. Results suggest that EMA research is generally well‐tolerated and can even have therapeutic effects but can also have negative consequences that participants should be made aware of during informed consent. Careful consideration of each element of study design when developing EMA protocols may also result in improved quantity and quality of data collected. Integration of these suggestions into research design may respond to the need to improve rates of participation and retention in ED research (Ortiz et al. 2020). The inclusion of individuals with lived experience into study design is invaluable in developing studies that not only measure the most relevant constructs but that also do so in a way that is acceptable to participants.

Author Contributions

Samantha Wilson: conceptualization, formal analysis, investigation, project administration, writing – original draft. Laura Lapadat: conceptualization, investigation, methodology, project administration, writing – review and editing. Lisa Y. Zhu: conceptualization, investigation, project administration, writing – review and editing. Sarah E. Racine: conceptualization, funding acquisition, resources, supervision, writing – review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

File S1. Sociodemographic characteristics of participants in the initial ecological momentary assessment study and the subsequent qualitative study.

File S2. Study experience and personalized feedback qualitative interview.

Wilson, S. , Lapadat L., Zhu L. Y., and Racine S. E.. 2025. “Ecological Momentary Assessment in Eating Disorders Research: A Qualitative Examination of Participant Experience and Recommendations for Future Studies.” International Journal of Eating Disorders 58, no. 9: 1710–1722. 10.1002/eat.24473.

Data Availability Statement

The data are not publicly available due to ethical restrictions.