A Systematic Review of Ecological Momentary Assessment for Obsessive‐Compulsive Disorder

ABSTRACT

Obsessive‐compulsive disorder (OCD) is characterized by its symptomatology heterogeneity, which poses challenges for its assessment and treatment. Ecological momentary assessment (EMA) has been proposed as a promising method for gathering information and may offer advantages over retrospective methods. The present study aims to conduct a systematic review of the use of EMA in OCD, examining the main variables assessed, methodological considerations, and the advantages of EMA compared with retrospective measures to assess OCD. The systematic search was conducted using databases including Scopus, Pubmed, Web of Science, PsycNET, and Cochrane Library. Of the identified studies, 28 met the inclusion criteria. These studies focused on the assessment of obsessions, compulsions, experiential avoidance, and sleep, with compulsions being the most studied variable. The studies reveal diverse EMA methods, with traditional paper recordings and smartphones differing in frequency and duration of the records. Some studies reported that EMA may serve as a valid and sensitive tool for detecting changes in OCD and overcome some limitations of retrospective assessments. Despite the potential of EMA, their validity is still questionable, and their application during intervention processes is limited, particularly given the heterogeneity of the available evidence and the limited application of EMA during intervention processes. Further research is essential to enhance their integration in clinical settings.

1. Introduction

Obsessive–compulsive disorder (OCD) is a debilitating mental health condition that significantly impairs the lives of affected individuals and their families (Murthy et al. 2022). It is classified as a major disabling illness, with a prevalence of 2%–3% (Hollander et al. 2010). The assessment and treatment of OCD present several challenges, largely due to its heterogeneity, both in the content of obsessions and in its variability in symptom trajectories (McKay et al. 2015).

Theoretical frameworks have classified obsessions into four main categories: contamination, symmetry/ordering, taboo thoughts, and responsibility/fear (Stein et al. 2016). Literature suggests that it is common to observe individuals with obsessions and compulsions from different categories and who may transition from one category to another over time (Mataix‐Cols et al. 2005). Moreover, OCD is often comorbid with other disorders (Gillan et al. 2017; Hollander 1998), which may further complicate its diagnosis (Roessner et al. 2022). Some authors (Belloch et al. 2016; Gillan et al. 2017; Pascual‐Vera et al. 2017) have explored the transdiagnostic nature of mental intrusions, showing that OCD symptomatology can also manifest in other mental disorders, including body dysmorphic disorder, hypochondriasis, and eating disorders.

Additionally, studies have found a mean delay of approximately 7 years between symptom onset and diagnosis, particularly pronounced in early‐onset cases (Dell'Osso et al. 2019; Hezel et al. 2022). This delay is partly due to early, nonspecific symptoms being minimized or misinterpreted by both individuals and family members, leading to postponement in seeking treatment (Stengler et al. 2013).

Several self‐report scales are commonly used for the OCD assessment, such as the Yale‐Brown Obsessive‐Compulsive Scale (Y‐BOCS; Goodman et al. 1989), the Obsessive‐Compulsive Inventory‐Revised (Foa et al. 2002), the Vancouver Obsessional Compulsive Inventory (Thordarson et al. 2004), and the Padua Inventory‐Palatine Revision (Gönner et al. 2010). However, since all these scales are based on retrospective recall of self‐reported symptoms, such information may be influenced by selective recall biases (Clark 1988; Stone and Shiffman 2002; Stone et al. 2004). As noted in the review by Hezel and McNally (2016), individuals with OCD, especially those patients exhibiting checking rituals, show biases in memory and attentional processing, which appear to be linked to metacognitive beliefs about their ability to accurately recall and attend to relevant information. Furthermore, research indicates that individuals with OCD may have impaired access to their internal states, such as emotions and thoughts (Dar et al. 2016), which could reduce the reliability of retrospective self‐reports. This limitation may be even more pronounced in specific populations, such as children, who present even lower levels of insight into their obsessions and compulsions compared with adults (Geller et al. 2001).

These limitations in the OCD assessment often contribute to prolonged durations of illness without intervention (Dell'Osso et al. 2019), leading to worse predictions in prognosis (Elsouri et al. 2024). Therefore, it is essential to adopt a more naturalistic and longitudinal approach to better capture the complexity of OCD symptomatology (Roessner et al. 2022). In contrast to cross‐sectional methods, longitudinal approaches allow the continuous tracking of symptom dynamics over time and across contexts. This is crucial for OCD, where symptom expression can vary not only between individuals but also within the same individual across different time points and situations.

One longitudinal promising approach is Ecological Momentary Assessment (EMA) (Stone and Shiffman 1994), also known as Experience Sampling Methods (ESM) (Csikszentmihalyi and Larson 1987). This method involves repeatedly collecting real‐time data on participants’ experiences in their natural environments (Stone et al. 2007). By conducting EMA, memory distortions, such as recall biases, are minimized, and the ecological validity of the data is increased (Ebner‐Priemer and Trull 2009; Shiffman et al. 2008). In addition, using EMA during the intervention may be an effective method to detect changes and identify new compulsions and obsessions over time (Trull and Ebner‐Priemer 2009). Furthermore, Geiger et al. (2024) emphasize the importance of including multiple measurement points in study designs, as using a single outcome point may overestimate remission rates.

EMA is increasingly recognized as a valuable tool for addressing the heterogeneity of OCD and promoting deeper understanding of the disorder. However, to the best of our knowledge, no systematic review has specifically synthesized the literature on EMA in OCD so far. While previous systematic reviews have focused on anxiety disorders more broadly rather than focusing exclusively on OCD (e.g., Walz et al. 2014), other reviews, such as Hall et al. (2021) did not include OCD in their review due to the reorganization in the DSM‐5, which removed it from the “anxiety disorders” category (American Psychiatric Association, DSM‐5 Task Force 2013). Examining the use of EMA in OCD assessment is essential for understanding the current state of the field and identifying the key directions for advancing our knowledge of the lived experiences associated with OCD. Synthesizing the literature on the applicability of EMA, such as its frequency, timing, and modality, would help establish practical guidelines for its effective use in OCD research and clinical practice (Jaeger et al. 2024).

Given the heterogeneity in the manifestations of OCD and the benefits that the use of EMA could have in its assessment and treatment, this study primarily aimed to synthesize the current literature on the use of EMA on OCD to address the following research questions: (1) Which variables are commonly included using EMA in individuals with OCD? (2) How is EMA implemented in OCD research in terms of modality, frequency, duration, and time of day of assessments? And finally, (3) does the use of EMA offer advantages over retrospective measures assessing OCD‐associated symptomatology? As a secondary objective, this review also seeks to examine the methodological quality and risk of bias in the included studies.

2. Methods

2.1. Search Strategy

The systematic literature search was performed independently by two reviewers (M.M. and I.J.) on February 1, 2024, using the PubMed, Web of Science, Scopus, Cochrane Library, and PsycNET databases. A search update was conducted on September 23, 2025, to ensure the inclusion of the most recent literature; the data presented in this review are based on the results of this updated search. The keyword string used was as follows: (“ecological momentary assessment” OR “experience sampling” OR “diary” OR “electronic diaries” OR “EMA” OR “ESM”) AND (“obsessive‐compulsive” OR “OCD”). The specific search strategies used in each database can be found in Supporting Information S1:File 1. In the case of articles whose abstracts or full text were not available, they were requested from the corresponding authors. This systematic review followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA; Page et al. 2021). The registration of this study was carried out on the Open Science Framework (OSF) platform, accessible through the link: https://osf.io/bxng3/overview?view_only=409248d6af6342ec8983070abfb2a3a0.

2.2. Inclusion Criteria

The titles and abstracts of the articles were reviewed to decide whether they would be considered in the full‐text reading phase after eliminating duplicate records. After performing the PICO (Participants, Intervention, Comparison, and Outcome) analysis, the following inclusion criteria were defined: (1) participants with a diagnosis of OCD; (2) adult population, adolescents, and children; (3) use of EMA, which was defined based on relevant literature (Stone and Shiffman 1994; Stone et al. 2007). EMA refers to a methodology involving multiple assessments conducted in real‐world environments, in real time or at selected moments, instead of relying on participants’ retrospective recall of their thoughts or feelings over extended periods. In the literature, these methodologies are sometimes also referred to as Experience Sampling Methods (ESM) or diary methods. Articles were excluded if: (1) included subclinical or healthy populations, to avoid variability that is not directly relevant to understanding symptom patterns, treatment sensitivity, or methodological considerations specific to OCD; (2) they were not original articles or gray literature, including reviews, meta‐analyses, protocols, or conferences; (3) they did not use EMA; (4) they were not written in English or Spanish.

2.3. Identification and Selection of Studies

Two reviewers (M.M. and I.J.) independently screened the articles. Initially, all titles and abstracts were reviewed to identify potential inclusions. Articles that did not meet the inclusion or exclusion criteria were eliminated. In the next phase, the full texts were examined to refine the selection further. During this in‐depth review, any articles that met exclusion criteria were removed. In cases of disagreement between the two reviewers, a third researcher (R.L.‐C.) resolved the discrepancies through discussion and consensus.

For the articles included, data were systematically collected across various categories: author(s), year and country of publication, total sample size and sample size specific to individuals with OCD, age group (adults, adolescents, or children), study design, format of EMA, variables assessed, duration of evaluation, and key findings.

2.4. Risk of Bias

The risk of bias in the selected studies was assessed using the National Heart Lung and Blood Institute (NHLBI) quality assessment tool, which is available on the NHLBI website (https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools). Depending on the designs used in the studies, specific criteria are established for their evaluation. Total quality scores range from 9 to 14 points, depending on the study design. This allows investigators to provide an overall rating of “good,” “fair,” or “poor” to each study. Two reviewers (M.M. and I.J.) evaluated all studies independently. Discrepancies were addressed by consensus with a third reviewer (R.L.‐C.).

3. Results

3.1. Selection and Inclusion of the Studies

As shown in Figure 1, the search in the five databases generated a total of 373 studies (PubMed = 71; Web of Science = 77; Scopus = 156; PsycNet = 20; Cochrane Library = 49). After removing duplicates (n = 175), a total of 198 studies remained and were independently reviewed independently by the two reviewers based on title and abstract. During this process, one corresponding author was contacted for a full article that was not available. This article was received and reviewed along with the others. After eliminating articles by title and abstract, 44 articles remained for full‐text reading. The authors carefully read the selected articles, and 14 articles were excluded according to the exclusion criteria: not having a diagnosis of OCD (n = 5), articles published in German (n = 2), being a protocol, thesis, or review (n = 5), and not using EMA, ESM, or daily diaries (n = 2). Therefore, a total of 30 articles remained for analysis. However, because the two publications by Rupp et al. (2019, 2020) and the two publications by Bischof et al. (2024, 2025) reported results from the same study using the same participant sample, the final synthesis comprised 28 studies. Discrepancies between authors during the article review process were resolved through discussion.

Figure 1.

Flow diagram for systematic reviews.

3.2. Characteristics of the Studies

The characteristics of the different studies included in this systematic review are shown in Table 1. The country where most research on EMA has been conducted in the OCD population is the United States (n = 11), followed by Canada (n = 7) and Germany (n = 5). England, Australia, France, Italy, and the Netherlands are also represented, with one study each.

Table 1.

Characteristics of the included studies.

Study	Country	OCD sample	Age	Design	EMA Measure	Intervention	Frequency	Duration	Outcomes	Results
Béland and O'Connor (2014)	Canada	8	Adults	Case series	Paper diary	Inference‐based therapy	1 time/day	20 weeks	Time spent on obsessions and compulsions, time spent neutralizing, emotions identification, level of credibility of the consequences, and degree of resistance to rituals.	Anxiety plays a major role in the persistence of primary doubt. The results also confirmed the effectiveness of Doubt Inhibition Therapy.
Bischof et al. (2024, 2025)	Germany	72 71	Adults	Prospective observational	Smartphone	No intervention	6 times/day	6 days	Momentary affect, reasons for non‐regulation, emotion regulation strategies, perceived effectiveness of emotion regulation strategies, OC symptoms and insight.	a Positive affect differed between OCD and healthy participants after controlling baseline depression, with the OCD group showing more avoidance strategies and lower perceived regulation effectiveness. b Insight levels fluctuated notably over the assessment period. Higher insight was partly linked to greater perceived emotion regulation effectiveness and emotional clarity. Additionally, symptom occurrence was consistently associated with lower momentary insight.
Bond and Guastello (2013)	USA	17 (of 33)	Adults	Case–control study	Paper diary	No intervention	Each activity lasting more than 15 min	1week	Daily activities and sleep; rituals during these activities; family members’ reaction to the rituals.	Family reactions and emotional responses do little to eliminate the rituals; instead, they may strengthen the dynamics.
Bouvard et al. (2020)	France	32	Adults	Prospective observational	Paper diary	No intervention	3 times/day minimum or each compulsion	3 days	Discomfort from the obsession, terrible consequences of not performing the compulsion, need for checking; duration and repetitions of the compulsion; general relief, relief from guilt, relief from responsibility, and reduction of discomfort.	The need to ritualize could explain the maintenance (or post‐treatment recurrence) of OCD in many patients.
Brown et al. (2020)	USA	2 (of 10)	Adults	Prospective observational	Online diary survey	No intervention	1‐4 times/day	20 weeks	Emotional distress‐anxiety; Distress‐depression; Sleep problems; Social isolation; Overall physical health; Physical functionality; Ability to manage emotions.	Some participants prefer an email or call over a text message as a reminder and that the repetition of questions could influence their response pattern, and sometimes they didn't have time to answer one question before the next one appeared.
Bucarelli and Purdon (2015)	Canada	22	Adults	Prospective observational	Paper diary	No intervention	3 times/day minimum or each compulsion	3 days	Compulsions. Obsession that triggered it; how much distress the obsession caused, and how terrible they believed the consequences of not performing the compulsion would be. General relief, specific relief from guilt, responsibility, and distress.	The number of repetitions of a compulsion correlated more with the feeling of uncertainty than with the distress from the obsession.
Cavallot ti et al., 2016	Italy	18 (of 36)	Adults	Case–control study	Paper diary	No intervention	1 time/day	1week	Narrative description of dreams.	The amount of obsession and compulsion themes in dreams is equal in both groups, the OCD group and the healthy group.
Coles et al. (2020)	USA	24 (of 39)	Adults	Case–control study	Sleep paper diary	No intervention	1 time/day	1week	Sleep‐wake schedules and timing of saliva collection to assess the onset of melatonin under dim light conditions.	42% of OCD patients met criteria for Delayed Sleep‐Wake Phase Disorder (DSWPD) compared to 0% in the control sample. The onset time of melatonin (DLMO) was significantly later in individuals with OCD compared to controls.
Cox and Olatunji (2022)	USA	20 (of 80)	Adults	Case–control study	Online diary survey	No intervention	1 time/day	1week	Subjective sleep duration and objective sleep duration.	Higher insomnia symptoms in those with OCD compared to healthy controls.
Cox et al. (2023)	USA	20 (of 74)	Adults	Prospective observational (Psychometric validation)	Online diary survey	No intervention	3 times/day	9 days	Checking, contamination, ordering, and obsessing subscales.	The d‐OCS demonstrated good internal consistency, adequate test‐retest reliability, and good convergent validity
Dean and Purdon (2021)	Canada	36	Adults	Prospective observational	Tablet	No intervention	Each compulsion	6 days	Compulsion time, descriptions of obsessive worry and compulsion, distress, satisfaction with the outcome of the episode, how harmful the consequences of not performing the compulsion correctly would be; general relief, relief of responsibility, and relief of guilt after completing the compulsion.	No relationships were found with the distress caused by obsession and belief in the obsession, nor was belief related to repetitions and duration.
Gloster et al. (2008)	USA	35	Adults	Prospective observational	PDA	No intervention	4 times/day	1 week	Compulsion, neutralization; duration of sleep the previous night, location, whether the participant was alone, mood, level of stress, anxiety, loneliness, and whether they experienced distress after an interpersonal conflict.	Participants tended to underestimate the frequency of their behaviors and overestimate the duration.
Goodman et al. (2023)	USA	4 (of 364)	Adults	Single‐arm longitudinal intervention study	Paper diary	Behavioral‐cognitive therapy and Dialectical Therapy	1 time/day	1 week (the first week of the intervention)	Disengagement (distraction, expressive suppression), perseveration (rumination, experiential avoidance), and engagement strategies (cognitive reappraisal, acceptance).	The day‐to‐day trajectories showed that the decrease in the use of experiential avoidance predicted changes the following day in the use of distraction and suppression.
Grenier et al. (2008)	Canada	8	Adults	Case series	Paper diary	Behavioral‐cognitive therapy	1 time/day	24 weeks	Intensity of the main obsessive doubt, intensity of belief in the consequences of doubt, and duration of compulsive rituals.	A decrease in belief in secondary consequences is accompanied by a decrease in the intensity of obsessions, but the opposite is not true.
Hohensee et al. (2025)	Germany	72	Adults	Prospective observational	Smartphone	No intervention	6 times/day	6 days	Momentary affect, emotion regulation strategies, perceived effectiveness of emotion regulation strategies, OCD symptoms	Higher self‐perceived ER effectiveness was associated with more positive affect, less negative affect, and fewer OC symptoms, while avoidance strategies showed partly negative effects. Engagement strategies were mostly non‐significant, except from a negative association with subsequent OCD symptoms.
Jacobson et al. (2022)	USA	3 (of 84)	Adults	Prospective observational	Online diary survey	No intervention	1 time/day	1week	Identify the best and worst time of the day, evaluate the positive and negative aspects of the three categories (morality, communion, responsibility).	The results of the multilevel modeling showed that controlling for depression and worry, OCD symptoms uniquely predicted higher daily distress.
Jaspers‐Fayer et al. (2018)	UK	25 (of 51)	Children and adolescents	Case–control Study	Paper diary	No intervention	1 time/day	1week	Bedtimes, wake times, estimated sleep latency and sleep quality.	72% of OCD patients had sleep problems compared to 15% of the control group.
Knox et al. (1996)	USA	4	Children	Case series	Paper diary	EPR therapy	Each compulsion	4weeks	Obsession related to compulsion, triggering location, day, time, and associated anxiety.	The results demonstrate that ERP involving parents is beneficial.
Landmann et al. (2019)	Germany	49	Adults	Prospective observational	Smartphone	No intervention	10 times/day	6 days	Presence of OCD symptoms, insight, worry, distraction, self‐punishment, social control, mindfulness, and compulsions.	Mindfulness predicts OCD insight over and above OCD symptoms
Landmann et al. (2020)	Germany	38	Adults	RCT	Smartphone	MBCT	6 times/day	6 days	Presence of OCD symptoms, insight, distress, positive and negative affect, and acceptance of emotions.	MBCT did not lead to significant changes in positive affect, negative affect, acceptance of momentary emotions, or distress. However, the presence of momentary OCD symptoms at post‐assessment predicted increased insight.
MacLaren et al. (2019)	USA	21	Adults	Prospective observational	Smartphone	No intervention	4 times/day	6‐10 days	Level of distress/anxiety and interference due to obsessions.	Participants overestimate the symptoms or their severity at the time of evaluation.
Perreault and O'Connor (2014)	Canada	4	Adults	Case series	Paper diary	Intervention/Inference Based Therapy	Each compulsion	20 weeks	Degree of belief in the primary obsessive doubt, perceived ability to resist compulsion in response to the primary obsessive belief, and duration of compulsion.	Decrease in obsessive doubt, increase in resistance to compulsion after intervention and decrease in daily duration of compulsion.
Polman et al.(2011)	Netherlands	7	Adults	Case series	Paper diary	Intervention/Behavioral‐cognitive therapy	1 time/day	19 weeks	Idiosyncratic thoughts, triggering situations, level of anxiety or stress, credibility level of thoughts, time consumed by idiosyncratic thinking.	Decrease in the credibility level of thoughts, association of anxiety with dysfunctional thoughts, association with compulsions and anxiety.
Puccinelli et al. (2025)	Canada	50	Adults	Prospective observational	Online diary survey	No intervention	4 times/day	5 days	Obsession triggers, links to incompleteness and harm‐avoidance motivations (adapted from OC‐CDI), interpretation of obsessions, and OCD behaviors (compulsions, avoidance, thought suppression, reassurance seeking)	Harm‐avoidance was associated with future harm beliefs, avoidance, reassurance seeking, and thought suppression, while incompleteness was associated with reduced negative self‐beliefs, compulsions, and lower likelihood of inaction. Both motivations shape OCD experiences via distinct cognitive and behavioral pathways.
Purdon et al. (2007)	Canada	37	Adults	Prospective observational	Paper diary	No intervention	3 times/day	3 days	Proactive or reactive avoidance, level of distress, trigger, possible outcome, strategies, duration of avoidance, outcome, emotional impact, and the ability to plan activities.	The results indicated that individuals with OCD engage in frequent, intense, time‐consuming efforts that ultimately are unsuccessful in controlling their thoughts.
Rupp et al. (2019) a 2020 b	Germany	40	Adults	RCT	Smartphone	Evaluation method pre‐post treatment (CR and DM)	10 times/day	4 days (pre‐post‐treatment)	Level of distress and relaxation; obsessions, perceived burden, anxiety, shame, sadness, guilt, hopelessness, disgust, tension, frustration, uncertainty, compulsive neutralization, compulsion, distance from obsession, obsession responsibility, risk, avoidance, frequency, difficulty, and relief.	a There was a reduction of avoidance, behavior, obsessions, and burden due to obsessions from pre to post‐treatment b There were no differences between CR and DM behaviors; that is, both interventions equally well promoted a seemingly general treatment effect.
Schubert et al. (2020)	USA	26 (of 44)	Adults	Case–control study	Paper diary	No intervention	1 time/day	1week	Sleep diary: sleep duration; Symptoms diary: frequency of obsessions/compulsions, level of distress from obsessions/compulsions, and perceived ability to control.	The results showed that bedtime (later bedtime schedules) predicted prospective increases in both obsessions and compulsions in individuals with OCD, but not in the other two groups.
Tilley and Rees (2014)	Australia	3	Children and adults	Case series	Smartphone and audio records	No intervention	10 times/day	1 day	Unwanted thoughts, type, trigger, and response to neutralize them.	Fewer symptoms of obsessions and compulsions were reported by the Y‐BOCS, but new obsessions and compulsions not captured by the Y‐BOCS were discovered.

Abbreviations: CR, cognitive restructuring; DM, detached mindfulness; ERP, exposure and response prevention; EMA, Ecological Momentary Assessment; MBCT, Mindfulness Cognitive Behavioral Therapy; OCD, Obsessive Compulsive Disorder; RCT, randomized control trial.

Regarding the age of the participants, most of the included studies (n = 25) were conducted with adult participants, except for two studies conducted with a sample of children and one study conducted with a sample composed of children and adolescents. In addition, one of the studies included a 5‐year‐old participant along with other adults (Tilley and Rees 2014). Regarding the design of the studies included in this systematic review, most were observational prospective studies (n = 13) in which EMA was used to study the association between OCD‐related variables. One prospective observational study was a psychometric validation of an EMA scale (Cox et al. 2023). A large number of articles were case–control studies (n = 6) or case series studies (n = 6). Two randomized controlled studies used EMA during the delivery of a psychological intervention. Specifically, one study used the EMA as a pre‐and post‐treatment assessment instrument (Rupp et al. 2019, 2020), while another study used the EMA to record patients’ progress during the application of the intervention (Landmann et al. 2020). Finally, one single‐arm longitudinal intervention study was found. It used EMA during the first week of the intervention to study short‐term emotion regulation changes through psychotherapy (Goodman et al. 2023).

3.3. Risk of Bias Assessment

As summarized in Table 2, prospective observational studies received scores ranging from 7 to 9 out of 9, indicating good quality. The main risk of bias was related to the lack of sample size justification and a loss to follow‐up greater than 20% or not reported.

Table 2.

Assessment for prospective observational studies.

	Bischof et al. (2024, 2025)	Bouvard et al. (2020)	Brown et al. (2020)	Bucarelli and Purdon (2015)	Cox et al. (2023)	Dean and Purdon (2021)	Gloster et al. (2008)	Hohensee et al. (2025)	Jacobson et al. (2022)	Landmann et al. (2019)	MacLaren et al. (2019)	Puccinelli et al. (2025)	Purdon et al. (2007)
1. Was the research question or objective in this paper clearly stated?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
2. Was the study population clearly specified and defined?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
3. Was the participation rate of eligible people at least 50%?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
4. Were all the subjects selected or recruited from the same or similar populations (including the same time period)? Were inclusion and exclusion criteria for being in the study prespecified and applied uniformly to all participants?	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes
5. Was sample size justification, power description, or variance and effect estimates provided?	Yes	No	No	No	No	No	No	Yes	Yes	Yes	Yes	No	No
6. Were the exposure measures (independent variables) clearly defined, valid, reliable, and implemented consistently across all study participants?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
7. Was the exposure(s) assessed more than once over time?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
8. Were the outcome measures (dependent variables) clearly defined, valid, reliable, and implemented consistently across all study participants?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
9. Was loss to follow‐up after baseline 20% or less?	Yes	NR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No
Total score	9	7	7	8	8	8	8	9	9	9	9	7	7

Note: Some items from the NIH tool for assessing observational cohort studies were removed to adapt it for assessing prospective observational studies.

Abbreviation: NR, not reported.

Also, case‐series studies are of good quality (Table 3), with all the studies obtaining scores from 6 to 9 out of 9. Most studies did not use a consecutive case design, which can introduce significant selection bias. Additionally, two studies did not provide sufficient descriptions of the included participants or information regarding the follow‐up.

Table 3.

Assessment of case series studies.

	Béland and O'Connor (2014)	Grenier et al. (2008)	Knox et al. (1996)	Perreault and O'Connor (2014)	Polman et al. (2011)	Tilley and Rees (2014)
1. Was the study question or objective clearly stated?	Yes	Yes	Yes	Yes	Yes	Yes
2. Was the study population clearly and fully described, including a case definition?	No	Yes	Yes	No	Yes	Yes
3. Were the cases consecutive?	No	No	NR	No	Yes	No
4. Were the subjects comparable?	Yes	Yes	Yes	Yes	Yes	Yes
5. Was the intervention clearly described?	Yes	Yes	Yes	Yes	Yes	Yes
6. Were the outcome measures clearly defined, valid, reliable, and implemented consistently across all study participants?	Yes	Yes	Yes	Yes	Yes	Yes
7. Was the length of follow‐up adequate?	No	Yes	Yes	Yes	Yes	NA
8. Were the statistical methods well‐described?	Yes	Yes	Yes	Yes	Yes	NA
9. Were the results well‐described?	Yes	Yes	Yes	Yes	Yes	Yes
Total score	6	8	8	7	9	6

Abbreviations: NA, not applicable; NR, not reported.

The category of pre‐post treatment studies without a control group (Table 4) included one study (Goodman et al. 2023), which was rated as good quality with 9 out of 12 points. Although the study conducted by Goodman et al. (2023) was classified by the authors of this systematic review as a single‐arm longitudinal intervention study design. However, it was evaluated using the tool for pre‐post studies due to the lack of a specific NHLBI tool for this design and the similarity between both designs in terms of the assessment of risk of bias. Only one item related to the use of an interrupted time‐series design was scored as not applicable (NA) since the aim of the Goodman et al. (2023) study was to assess changes during the first week of the intervention using repeated daily assessments rather than a traditional pre‐post comparison.

Table 4.

Assessment of before‐after (pre‐post) studies with no control group (including single‐arm longitudinal intervention studies).

	Goodman et al. (2023)
1. Was the study question or objective clearly stated?	Yes
2. Were eligibility/selection criteria for the study population prespecified and clearly described?	Yes
3. Were the participants in the study representative of those who would be eligible for the test/service/intervention in the general or clinical population of interest?	Yes
4. Were all eligible participants that met the prespecified entry criteria enrolled?	Yes
5. Was the sample size sufficiently large to provide confidence in the findings?	Yes
6. Was the test/service/intervention clearly described and delivered consistently across the study population?	Yes
7. Were the outcome measures prespecified, clearly defined, valid, reliable, and assessed consistently across all study participants?	Yes
8. Were the people assessing the outcomes blinded to the participants’ exposures/interventions?	NR
9. Was the loss to follow‐up after baseline 20% or less? Were those lost to follow‐up accounted for in the analysis?	No
10. Did the statistical methods examine changes in outcome measures from before to after the intervention? Were statistical tests done that provided p values for the pre‐to‐post changes?	Yes
11. Were outcome measures of interest taken multiple times before the intervention and multiple times after the intervention (i.e., did they use an interrupted time‐series design)?	NA
12. If the intervention was conducted at a group level (e.g., a whole hospital, a community, etc.) did the statistical analysis take into account the use of individual‐level data to determine effects at the group level?	Yes
Total score	9

Abbreviations: NR, not reported; NA, not applicable.

The randomized control study category (Table 5) included two studies. One study, with results reported in two published articles (Rupp et al. 2019, 2020), obtained 11 points out of 14, indicating good quality. The other study received 9 points out of 14, indicating fair quality. The main limitations of these studies were the lack of reporting on whether the treatment assignment was concealed and whether the participants and researchers were unaware of the treatment group assignment.

Table 5.

Assessment of controlled intervention studies.

	Landmann et al. (2020)	Rupp et al. (2019, 2020)
1. Was the study described as randomized, a randomized trial, a randomized clinical trial, or an RCT?	Yes	Yes
2. Was the method of randomization adequate (i.e., use of randomly generated assignment)?	Yes	Yes
3. Was the treatment allocation concealed (so that assignments could not be predicted)?	NR	NR
4. Were study participants and providers blinded to treatment group assignment?	NR	NR
5. Were the people assessing the outcomes blinded to the participants’ group assignments?	NR	NR
6. Were the groups similar at baseline on important characteristics that could affect outcomes (e.g., demographics, risk factors, co‐morbid conditions)?	Yes	Yes
7. Was the overall drop‐out rate from the study at endpoint 20% or lower of the number allocated to treatment?	Yes	Yes
8. Was the differential drop‐out rate (between treatment groups) at endpoint 15 percentage points or lower?	Yes	Yes
9. Was there high adherence to the intervention protocols for each treatment group?	Yes	Yes
10. Were other interventions avoided or similar in the groups (e.g., similar background treatments)?	Yes	Yes
11. Were outcomes assessed using valid and reliable measures, implemented consistently across all study participants?	Yes	Yes
12. Did the authors report that the sample size was sufficiently large to be able to detect a difference in the main outcome between groups with at least 80% power?	No	Yes
13. Were outcomes reported or subgroups analyzed prespecified (i.e., identified before analyses were conducted)?	Yes	Yes
14. Were all randomized participants analyzed in the group to which they were originally assigned, i.e., did they use an intention‐to‐treat analysis?	No	Yes
Total score	9	11

Abbreviation: NR, not reported.

Finally, the case–control studies (Table 6) showed scores between fair and good quality, ranging between 8 and 9 out of 12. The main limitation of these studies was the lack of justification for the sample size. In addition, studies usually showed a risk of bias related to the control group. Specifically, some studies recruited control samples with comparability limitations with the experimental group or did not use concurrent controls.

Table 6.

Assessment of case–control studies.

	Bond and Guastello (2013)	Cavallotti et al. (2016)	Coles et al. (2020)	Cox and Olatunji (2022)	Jaspers‐Fayer et al. (2018)	Schubert et al. (2020)
1. Was the research question or objective in this paper clearly stated and appropriate?	Yes	Yes	Yes	Yes	Yes	Yes
2. Was the study population clearly specified and defined?	Yes	Yes	Yes	Yes	Yes	Yes
3. Did the authors include a sample size justification?	No	No	No	No	No	No
4. Were controls selected or recruited from the same or similar population that gave rise to the cases (including the same timeframe)?	No	No	Yes	Yes	Yes	No
5. Were the definitions, inclusion and exclusion criteria, algorithms or processes used to identify or select cases and controls valid, reliable, and implemented consistently across all study participants?	Yes	Yes	Yes	Yes	Yes	Yes
6. Were the cases clearly defined and differentiated from controls?	Yes	Yes	Yes	Yes	Yes	Yes
7. If less than 100 percent of eligible cases and/or controls were selected for the study, were the cases and/or controls randomly selected from those eligible?	Yes	Yes	No	Yes	Yes	Yes
8. Was there use of concurrent controls?	No	No	No	No	No	No
9. Were the investigators able to confirm that the exposure/risk occurred prior to the development of the condition or event that defined a participant as a case	Yes	Yes	Yes	Yes	Yes	Yes
10. Were the measures of exposure/risk clearly defined, valid, reliable, and implemented consistently (including the same time period) across all study participants?	Yes	Yes	Yes	Yes	Yes	Yes
11. Were the assessors of exposure/risk blinded to the case or control status of participants?	NR	NR	NR	NR	NR	NR
12. Were key potential confounding variables measured and adjusted statistically in the analyses? If matching was used, did the investigators account for matching during study analysis?	Yes	Yes	Yes	Yes	Yes	Yes
Total score	8	8	8	9	9	8

Abbreviation: NR, not reported.

3.4. Methodological Considerations in the Application of Ecological Momentary Assessment

3.4.1. Modality, Frequency and Duration of Assessment

The reviewed studies reveal great heterogeneity in the assessment designs used to monitor OCD, highlighting the diversity in the frequency of assessments and the recording format used. Traditional paper‐based assessments (n = 15) were characterized by their extended duration, ranging from 1 week to 24 weeks. The frequency of assessment is usually one to two times per day, with the records at the beginning or end of the day being the most common. These daily assessments, with low recording frequency, allow for long‐term analysis of OCD‐associated behaviors.

On the other hand, electronic device‐based assessments (n = 14) included the use of smartphones (n = 8), online surveys (n = 4), personal digital assistants (PDAs; n = 1), or tablets (n = 1), which introduce a more immediate and reactive approach. These devices use different methods, such as auditory signals, text messages, or vibrations, to remind participants of the completion of the assessments. The frequency of entries using these devices ranged from 1 time/day to 10 times/day with varied duration ranging from 1 day to 20 weeks. The most common frequency and duration was once per day for 1 week. Some studies indicated that a limited number of days may be insufficient to capture OCD symptomatology. For example, Bouvard et al. (2020) concluded that using three entries per day for 3 consecutive days was not enough to identify individual's compulsions. They recommend a 1‐week period to better reflect compulsions occurring in daily life. Similarly, Cox et al. (2023) indicated in their study limitations that periods longer than 1 week would be beneficial for capturing OCD symptomatology, although they did not recommend an exact duration.

The timing of assessments also varied across studies. When studies included more than one prompt per day, prompts were typically delivered at random intervals throughout the participation period, with minimum gaps of 30 min (Landmann et al. 2019, 2020), 1 h (Bischof et al. 2024, 2025; Hohensee et al. 2025; Puccinelli et al. 2025) or 2 h (Tilley and Rees 2014) between each prompt.

3.4.2. Source of items

Some authors used validated measures to assess obsessions and compulsions (Gloster et al. 2008; Goodman et al. 2023; Landmann et al. 2019, 2020; Tilley and Rees 2014; MacLaren et al. 2019), such as the OCI‐R and the Y‐BOCS. Items from semi‐structured interviews like the Clinical Scales of O'Connor and Robillard (1996) were also used (Perreault and O'Connor 2014). Other researchers developed their own assessment of momentary OC symptoms and calculated the individual intraclass correlation coefficients (ICCs). They reported a moderate reliability across repeated measures (Bischof et al. 2024, 2025; Hohensee et al. 2025), with a certain degree of variability reflecting the dynamic and fluctuating nature of OCD symptoms in daily life.

The most used format to assess OCD symptoms was the Likert scale, usually ranging from 1 to 5 (Bischof et al. 2024, 2025; Hohensee et al. 2025; Landmann et al. 2020) or from 1 to 7 (Purdon et al. 2007; Rupp et al. 2019, 2020). This format was used to evaluate aspects such as intensity of OCD symptoms, insight, degree of belief or interference in daily life, and degree of probability of occurrence. Items using a VAS scale ranging from 0 to 100 were also developed to assess the degree of conviction in the probability of the doubt, degree of conviction in anticipated consequences, and conviction that feared belief is realistic or due to OCD (Grenier et al. 2008; Landmann et al. 2019).

Notably, two studies adapted items from existing scales to develop validated instruments for assessing daily OCD symptomatology. Specifically, Cox et al. (2023) examined the psychometric properties of an EMA scale (the d‐OCS), which integrates items from established OCD measures modified for daily use. The d‐OCS correlated significantly with widely used instruments such as the Y‐BOCS and the OCI‐R, demonstrating good convergent validity. This scale employed a 0–4 Likert format, with higher scores reflecting greater severity of daily OCD symptoms. Similarly, Puccinelli et al. (2025) adapted items from the Obsessive‐Compulsive Core Dimensions Interview (Summerfeldt 2004) to capture obsessional triggers and to evaluate the extent to which participants’ experiences were linked to motivations of incompleteness and harm avoidance. Responses were recorded either as multiple‐choice options or on a scale ranging from 0 (“not at all”) to 100 (“extremely”).

Regarding the contextualization of items, many studies began assessments with prompts referring to participants’ OCD symptoms since their last report. For example, Tilley and Rees (2014) asked “Since the last prompt…”, while Gloster et al. (2008) started with “In the last four hours…,”. Other studies asked participants about their state immediately before the prompt (e.g., “Immediately before the signal”) (Bischof et al. 2024, 2025; Hohensee et al. 2025; Landmann et al. 2019, 2020). In most studies, however, the contextualization of items was not clearly specified. Sometimes this was due to the use of paper‐based diaries, in which participants were instructed to complete a set of questionnaires at a specific time of day (e.g., in the evening) or whenever possible in real time. Therefore, the assessment is completed based on what occurred on the last day or during the most recent compulsive episode.

Other researchers have used diaries developed in previous studies that combine closed‐ and open‐ended questions to evaluate OCD symptoms. These diaries are generally not validated paper‐based measures, as they are primarily designed to collect narrative or qualitative information about symptom occurrence and characteristics. For example, Dean and Purdon (2021) utilized the repeated actions diary (RAD) by Purdon and Bucarelli (2010) to assess the characteristics of compulsive episodes. Additionally, Béland and O'Connor (2014) used log tables based on the guideline from Freeston and O'Connor (1997) for OCD self‐monitoring. Another format was seen in Polman et al. (2011), where a diary was collaboratively developed with the patient to collect the most important symptoms during the intervention.

Similarly, in studies including sleep assessment, participants were asked to record sleep‐wake schedules (Coles et al. 2020; Cox and Olatunji 2022; Jaspers‐Fayer et al. 2018; Schubert et al. 2020). Also, the study of Cavallotti et al. (2016) used narrative descriptions of dreams to identify whether they included themes related to typical obsessions/compulsions in OCD. In some cases, sleep diaries were complemented with actigraphy data (Coles et al. 2020; Cox and Olatunji 2022; Jaspers‐Fayer et al. 2018; Schubert et al. 2020), although findings regarding the consistency between self‐reported sleep and actigraphy have been mixed. Specifically, Schubert et al. (2020) reported good validity of the sleep diary, observing a high correlation between daily sleep records and corresponding actigraphy data, whereas Jaspers‐Fayer et al. (2018) found discrepancies between self‐reported bedtimes and actigraphy recordings.

3.4.3. Clinical Applications of EMA

Another aspect to consider in the studies is whether the EMA has been integrated into an intervention process. On the one hand, Rupp et al. (2019, 2020) used EMA as a pre‐ and post‐treatment assessment method. Specifically, they compared changes in OCD‐related emotions assessed with EMA after the use of interventions based on cognitive restructuring or detached mindfulness. They found that the EMA method is sensitive to treatment effects in OCD, showing a reduction of obsessions, compulsions, and avoidance from pre‐treatment to post‐treatment. However, no control groups or participants with other psychiatric conditions were included, so it was not possible to assess whether the observed differences might have been influenced by reactivity to the EMA itself.

On the other hand, some studies incorporated the use of EMA during an intervention process. Specifically, two of the studies used EMA together with the delivery of cognitive‐behavioral treatment (Grenier et al. 2008; Polman et al. 2011), one together with cognitive‐behavioral therapy with Dialectical Therapy components (Goodman et al. 2023), two together with Inference‐Based Therapy (Béland and O'Connor 2014; Perreault and O'Connor 2014), one with Exposure with Response Prevention (Knox et al. 1996), and one in Mindfulness‐based on cognitive therapy (Landmann et al. 2020).

3.5. Main Outcomes Evaluated With Ecological Momentary Assessments

The most frequently studied domains in the studies included in this review were obsessions, compulsions, avoidance, and sleep. Obsessions were assessed in 20 studies. Specifically, this variable was analyzed in terms of perceived burden, the intensity of obsessive doubts, and various associated emotions, such as anxiety, shame, sadness, and guilt, among others. Compulsions were the most frequently studied variable, with 23 studies with detailed descriptions of their frequency, duration, and triggers, as well as the emotional consequences of their performance, such as relief or reduced distress. Participants assessed the discomfort linked to the obsessions, the severe consequences they perceived for not performing the compulsions, and the need for verification. In addition, the studies investigated general and specific relief from guilt, responsibility, and distress derived from the compulsions. Anxiety was the most repeated emotion associated with obsessions and compulsions. In one of the studies, the positive relationship between anxiety and dysfunctional thoughts, as well as between anxiety and compulsions, was shown. Specifically, the lower the anxiety, the fewer the dysfunctional thoughts and compulsions (Polman et al. 2011). Another study suggested that anxiety played a key role in maintaining primary doubt, decreasing both during the course of therapy (Béland and O'Connor 2014).

A smaller number of studies assessed avoidance (n = 9), by examining both proactive (trying to prevent an obsession from entering the mind before it appears) and reactive (trying to eliminate an obsession that is already in the mind) strategies. In addition, they assessed the timing and duration of avoidance, along with its emotional impact and its effect on the ability to plan activities (Purdon et al. 2007). Bischof et al. (2024) examined the use of emotion regulation strategies and found that individuals with OCD tend to rely more on avoidance‐oriented strategies than healthy controls. Moreover, at the within‐subjects level, higher momentary negative affect was associated with increased use of avoidance strategies and with reduced perceived effectiveness and emotional engagement. Similarly, Hohensee et al. (2025) reported that engagement in avoidance‐oriented strategies was linked to lower positive affect, higher negative affect, and greater OCD symptom severity. Findings at the between‐person level were more robust than findings on the within‐person level, so that higher‐than‐average use of avoidance‐oriented ER strategies was associated with more OCD symptoms and negative affect. Also, one study showed that reduced avoidance at the same time that obsessions, the burden they carry, and compulsions decreased (Rupp et al. 2019).

Finally, a total of seven studies included in this systematic review included sleep, covering bedtime, wake time, sleep latency estimation, and sleep quality assessment. These studies demonstrated the relationship between OCD and sleep problems. For example, Jaspers‐Fayer et al. (2018) showed that 72% of OCD patients included in their study experienced sleep problems, compared with 15% of the control group. Specifically, insomnia symptoms were higher in people with OCD than in the healthy control group. Coles et al. (2020) showed that 42% of the sample of people with OCD met the criteria for delayed sleep‐wake phase disorder, compared with no cases among participants without OCD. It was also confirmed by the time of onset of salivary melatonin secretion, which was markedly later in people with OCD compared with the control group. Furthermore, one study identified the mediating role of insomnia between delayed circadian rhythms and OCD symptoms (Cox and Olatunji 2022). Finally, Schubert et al. (2020) indicated that people who go to bed late are more likely to have obsessions and compulsions.

3.6. Comparing Ecological Momentary Assessment and Retrospective Assessments

Although EMA protocols vary in design and the items they include, some studies included in this systematic review showed the validity and reliability of their EMA for monitoring OCD symptoms. For example, EMA measures such as the d‐OCS (Cox et al. 2023) have been shown to have good internal consistency, showing stability across three daily time points in samples of individuals with OCD, psychiatric controls, and healthy controls. The d‐OCS also showed good test–retest reliability, comparable to that of traditional assessments (i.e., Y‐BOCS and OCIR). In addition, it demonstrated good convergent validity, as evidenced by significant correlations with established measures of OCD symptoms, including the Y‐BOCS and the OCI‐R.

Gloster et al. (2008) failed to detect consistent differences between OCD symptoms retrospectively recalled and those recorded through EMA, supporting its convergent validity with traditional measures. Moreover, the same study conducted within‐subjects analyses across three time points and showed that EMA did not provoke reactivity; in other words, the act of self‐recording symptoms via EMA over time did not alter the frequency or intensity of the OCD symptoms. Although the study lasted only 1 week, this suggests that EMA can reliably capture momentary symptom patterns without significantly influencing them. In contrast, MacLaren et al. (2019) observed that participants overestimated their OCD symptoms during the retrospective assessment, rating significantly higher than when utilizing Experience Sampling Methods. Specifically, this study found reactivity for obsession distress, but no reactivity for other variables such as interference, indicating that some OCD symptom domains may be more sensitive to repeated self‐monitoring than others.

On the other hand, EMA has been shown to capture clinically relevant information that may be missed by standard retrospective measures. Tilley and Rees (2014) found that EMA identified new obsessions and compulsions not captured by the Y‐BOCS, suggesting that EMA can provide additional clinically relevant information beyond standard measures.

4. Discussion

Research on OCD has advanced considerably, using different methodologies to address the complexity and diversity of this mental disorder. This study aimed to explore the application of EMA as an innovative tool in assessing OCD, focusing on which variables are measured, how EMA is implemented, and how it is compared with traditional retrospective assessments.

Regarding the first aim of this study, which was to examine the variables assessed through EMA in individuals with OCD, most studies have focused on obsessions and compulsions as primary outcomes, which correspond to two fundamental components of Cognitive‐Behavioral Therapy (CBT). Several EMA studies reported findings that are broadly consistent with central maintenance theories of OCD, such as Salkovskis’ cognitive model (Salkovskis 1999). Overall, the available studies suggested that a greater number of compulsions were performed when there was less certainty regarding the obsession, when the obsession is of an unattainable nature, or when individuals report a heightened need for security (Bucarelli and Purdon 2015; Bouvard et al. 2020; Dean and Purdon 2021). Importantly, EMA data also show that compulsions are followed by increased relief, including relief from responsibility, alongside decreased discomfort (Bouvard et al. 2020). These results are consistent with the idea that compulsions temporarily reduce distress and reinforce inflated responsibility beliefs, thereby perpetuating the cycle of OCD symptoms, as proposed in cognitive‐behavioral models.

Another critical factor included in EMA studies is avoidance, which acts as a negative reinforcement mechanism for obsessions (Goodman et al. 2023; Purdon et al. 2007; Rupp et al. 2019, 2020). The studies often analyzed this strategy alongside other avoidance‐oriented emotion regulation strategies, such as distraction (e.g., Bischof et al. 2024), in order to examine the dynamics of daily emotion regulation in individuals with OCD. Importantly, these avoidance‐oriented strategies were analyzed separately from engagement‐oriented strategies, such as problem solving or cognitive reappraisal. The available findings showed that individuals with OCD chose avoidance strategies more frequently than healthy controls (Bischof et al. 2024). In addition, individuals who used a higher average number of avoidance strategies reported more intense OCD symptoms (Hohensee et al. 2025). These findings are consistent with literature using traditional measures, which showed that OCD patients who relied on avoidance presented greater obsessive–compulsive severity compared with those who did not (McGuire et al. 2012). They also align with Salkovskis’ model (Salkovskis 1999), which emphasizes the role of neutralizing strategies, such as avoidance, in maintaining OCD symptoms.

Although only a limited number of studies have examined sleep using daily diaries, the available evidence suggests an association between OCD and sleep disturbances, emphasizing the need to incorporate sleep‐related factors into clinical evaluations. Research included in this systematic review showed that individuals with OCD had a markedly higher prevalence of insomnia symptoms compared with healthy control groups (Cox and Olatunji 2022). These findings suggest that sleep dysfunction may represent an important clinical factor to consider during assessment and treatment of OCD, as improving sleep quality may enhance overall therapeutic outcomes and reduce symptom severity. It is important to note that while some studies consider the potential influence of comorbid disorders (e.g., Coles et al. 2020) or exclude participants with comorbid conditions from the OCD sample (e.g., Cavallotti et al. 2016; Jaspers‐Fayer et al. 2018;), other studies included broader samples with additional anxiety disorders or did not control for comorbidities or medications that might affect their results.

Regarding characteristics for EMA implementation, a remarkable difference was observed between studies utilizing paper‐based methods and those employing electronic devices. Studies using electronic devices often report a higher frequency of assessments, likely due to the convenience and use of automated prompts that remind participants to complete evaluations. Although the most common frequency and duration in studies was once per day during a week, some research acknowledges that this may be limited and recommends monitoring periods of at least 1 week to adequately capture OCD symptomatology (Bouvard et al. 2020; Cox et al. 2023). However, caution is warranted, since an increased frequency of assessments may influence participants’ response patterns and potentially introduce fatigue or reactivity, as shown by an EMA study on anxiety disorders (Brown et al. 2020). Previous research emphasizes the importance of maintaining a manageable assessment schedule to reduce participant burden and ensure data quality (Wrzus and Neubauer 2022). These findings underscore the need for further research comparing different EMA designs and establishing standardized guidelines for implementation. A lack of standardization may limit the comparability and generalizability of results, thereby hindering a consistent understanding of OCD in real‐world contexts.

In terms of the source of items, several studies reported using items from validated measures or indicated satisfactory psychometric properties of the items used. In this context, the study by Cox et al. (2023) is particularly noteworthy, as it validated a specific instrument, the d‐OCS, for assessing daily OCD symptoms through EMA. However, more than half of the studies did not report the origin or validation of their items. Future research is needed to further examine the validity and reliability of traditional measures in longitudinal contexts, with the aim of improving EMA assessments. Some instruments such as the Y‐BOCS and Dimensional Obsessive–Compulsive Scale or the d‐OCS have shown good psychometric properties for being used in a longitudinal way (Kuckertz et al. 2021), although additional validation across different EMA protocols is still needed. The use of these validated measures is recommended to enhance the comparability of EMA studies and make it possible to draw conclusions about symptom patterns and treatment effects. Additionally, combining EMA with subjective measures, an approach that has only been implemented in sleep studies with individuals with OCD, could provide richer information about individuals’ internal experiences in real time. Such integration would improve our understanding of symptom dynamics and the interactions between behavioral, cognitive, and physiological processes.

When comparing EMA with retrospective measures, some studies included in this systematic review indicated that EMA was associated with fewer recall biases compared with other conventional methods (e.g., Y‐BOCS) (MacLaren et al. 2019; Tilley and Rees 2014). In addition, EMA may identify symptoms that could not be identified with retrospective methods (Tilley and Rees 2014). However, these conclusions are based on a limited number of studies.

Despite the benefits EMA can bring to treatment, it is critical to acknowledge the concerns raised as potential limitations by various studies. For example, it is important to pay attention to the presence of social desirability, as participants may provide socially acceptable responses rather than accurately report their actual experiences or emotional states, potentially compromising the reliability and validity of EMA (Grenier et al. 2008; Rupp et al. 2019; Landmann et al. 2019). The subjectivity associated with factors such as mood and individual motivations introduces the possibility of variability in responses over time, which could affect data consistency and interpretation (Grenier et al. 2008). However, although such variability can cause challenges for internal validity by complicating the consistency of measurements, it may also represent a potential strength of EMA, as it enables the study of within‐person fluctuations and the dynamic interplay between emotions, symptoms, and contextual factors in real time. In addition, social desirability is not a challenge specific to EMA and may be even lower than in direct interviews (M. R 2007).

Regarding the risks of bias in the included studies, good quality was found overall. There is a need for justification of sample sizes and the improvement of the comparability of control groups with the OCD groups. Also, future studies should indicate whether strategies are taken for blinding the allocation of the participants.

This review provides a comprehensive examination of EMA in OCD, suggesting that EMA may be useful for capturing real‐time symptom dynamics, while also evaluating methodological considerations and identifying areas for future research. Despite these strengths, it is not without limitations. First, the studies obtained in this systematic review were very heterogeneous in terms of design and objectives, making it difficult to obtain robust findings. As noted earlier, the use of validated measures for this purpose would represent a significant contribution, as it would enhance the reliability of the data and allow for broader generalization of findings. In addition, the language restriction to Spanish and English may have been left out of the review of relevant studies published in other languages. Finally, this systematic review did not include studies with non‐patient samples, although it did include studies with OCD patients using EMA methods not specifically designed to monitor OCD symptoms (i.e., Goodman et al. 2023). Consequently, some studies that could have been of interest in terms of design or EMA implementation characteristics were excluded, while some studies included provided limited information relevant to this review.

5. Conclusion

In conclusion, this review highlights the potential of EMA as a valuable tool for the assessment of OCD. Although the number of articles found is limited, this is likely due to the relative newness of EMA research and their recent incorporation into the clinical setting, where EMA appears to be a promising method. The available evidence suggests that EMA may provide more ecologically valid assessments compared with retrospective methods, potentially reducing recall bias and enabling clinicians to capture symptoms as they occur in real‐world settings and support them in developing more personalized and targeted treatment strategies.

Smartphones appear to be a practical tool for administering EMA, as they allow participants to receive prompts, ensuring the completion of the assessments. However, careful consideration of assessment frequency is crucial, as overly frequent evaluations may affect participants’ response patterns and data quality.

Despite their benefits, EMA is not without limitations. While some studies offer encouraging findings, further research is needed to replicate these results and establish the sensitivity of EMA in detecting symptom changes. Additionally, more research is necessary to assess external validity, particularly examining factors such as social desirability, mood influences, and their impact on responses. Standardizing EMA protocols and addressing these limitations will be essential for maximizing the utility of EMA in both research and clinical practice.

Overall, although the available evidence is encouraging, further standardized and methodologically rigorous studies are needed before firm conclusions can be drawn regarding the validity, sensitivity, and clinical advantages of EMA in OCD assessment and treatment.

Author Contributions

Irene Jaén: conceptualization. Irene Jaén, Maria Mostazo‐Arolo and Rosa Lorente‐Català: literature search and data analysis. María Mostazo‐Arolo: writing the manuscript. Irene Jaén and Rosa Lorente Català: review and editing the manuscript.

Ethics Statement

This systematic review was conducted using data from previously published studies. No primary data involving human participants or animals were collected. Therefore, ethical approval and informed consent were not required. References marked with an asterisk indicate studies included in the systematic review.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File

Data Availability Statement

This review is based solely on previously published studies. All data underlying this article are already publicly available in the cited sources.