CoChat
Sign Up
Cited by 168Digital Mental Health InterventionsMental Health Research TopicsAnxiety, Depression, Psychometrics, Treatment, Cognitive Processes

Ecological momentary interventions for depression and anxiety

Open access
Schueller, Stephen M., Aguilera, Adrian, Mohr, David C.·2017·Depression and Anxiety
DOI 10.1002/da.22649
Open this paper in CoChat — free accountOpen PDF

Abstract

Ecological momentary interventions (EMIs) are becoming more popular and more powerful resources for the treatment and prevention of depression and anxiety due to advances in technological capacity and analytic sophistication. Previous work has demonstrated that EMIs can be effective at reducing symptoms of depression and anxiety as well as related outcomes of stress and at increasing positive psychological functioning. In this review, we highlight the differences between EMIs and other forms of treatment due to the nature of EMIs to be deeply integrated into the fabric of people's day-to-day lives. EMIs require unique considerations in their design, deployment, and evaluation. Furthermore, given that EMIs have been advanced by changes in technologies and that the use of behavioral intervention technologies for mental health has been increasing, we discuss how technologies and analytics might usher in a new era of EMIs. Future EMIs might reduce user burden and increase intervention personalization and sophistication by leveraging digital sensors and advances in natural language processing and machine learning. Thus, although current EMIs are effective, the EMIs of the future might be more engaging, responsive, and adaptable to different people and different contexts.

AI snapshot

Written from the paper’s full text.

TL;DR

Ecological momentary interventions (EMIs) can reduce depression and anxiety symptoms, with small-to-medium effects, and are distinguished from traditional treatments by being embedded in daily life and governed by decision rules. The review argues that future EMIs may become more engaging, responsive, and personalized through sensors, natural language processing, and machine learning.

Key findings

  • In a meta-analysis of 33 EMI studies, 17 included depression outcomes and showed a small-to-medium within-person pre-post effect of Hedges’ g = .48 (95% CI 0.34–0.61). Supported EMIs had larger effects (g = .73) than stand-alone EMIs (g = .45) or stand-alone EMIs plus care as usual (g = .38).
  • For anxiety, 15 studies showed a within-person pre-post effect of g = .47 (95% CI 0.32–0.63), nearly identical to depression. Another review of 7 studies found a small pooled effect of d = .32 (95% CI 0.12–0.53) favoring EMIs over comparison/control conditions; stress outcomes were significantly superior to comparison conditions.
  • EMIs differ from face-to-face and other digital treatments in design and evaluation; they combine intervention options with decision rules, and just-in-time adaptive interventions (JITAIs) use algorithms to optimize individual interventions over time.
  • Current EMIs are often limited: many require user initiation, pull from prepopulated options, and only two anxiety EMI studies used automated sensors, despite technical advances enabling passive detection.

Methods

  • Narrative review of EMIs for depression and anxiety, covering design, prior evidence, evaluation methods, and future directions.
  • Evidence is drawn from prior systematic reviews and meta-analyses, especially Versluis et al. (2016; 33 studies, 17 with depression outcomes, 15 with anxiety outcomes) and Gee et al. (2016; 7 studies on anxiety/stress).
  • Discusses evaluation approaches including microrandomized trials (MRTs), n-of-1 trials, and machine-learning/bandit algorithms for personalization.

Limitations

  • The paper states that the overall quality rating of studies in the meta-analysis was low, averaging 2.29 on a six-point scale, and that EMI methodology is still in its infancy, so results should be interpreted with caution.
  • The paper notes that some comparisons are meta-analytic rather than experimental, and that stand-alone EMIs plus care as usual were not more effective than stand-alone EMIs; only two anxiety EMI studies used automated sensors.
  • Evident from the text: this is a narrative review without a reported systematic search protocol or primary data collection, so its conclusions depend on the cited reviews and studies.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1002/da.22649

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 41 cited sources checked · 0 supported · 0 flagged

  • Premiseown data

    Additional support was the one significant moderator of EMI effects on depression.

    “Analysis revealed one significant moderator, which was additional support, which is consistent with findings in other areas of BITs in which supported interventions tend to be more efficacious than unsupported interventions.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • Premisecitation

    Early EMIs used platforms such as palmtop computers or mobile phones that lacked both the penetration and features of current devices.

    “Early EMIs used platforms such as palmtop computers (Newman, Przeworski, Consoli, & Taylor, 2014) or mobile phones (Burns et al., 2011) that lacked both the penetration and features of current devices.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Newman, Przeworski, Consoli, & Taylor, 2014 source

      Not checked in the free run.

    • cites Burns et al., 2011 source

      Not checked in the free run.

  • Premisecitation

    More than 90% of smartphone owners report having their phones charged, turned on, and constantly within arm's reach.

    “More than 90% of owners report having their phones charged, turned on, and constantly within arm's reach (Rainie & Zickuhr, 2015).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Rainie & Zickuhr, 2015

      Not checked in the free run.

  • Premiseown data

    The overall quality rating for the studies analyzed was low, with an average of 2.29 on a six-point scale.

    “Importantly, the overall quality rating for the studies analyzed was low, with an average of 2.29 on a sixpoint scale.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • Premisecitation

    EMIs intended to reduce depression do so through interventions aimed at various proximal outcomes such as engagement in pleasurable activities, increasing positive emotions, or other pathways.

    “EMIs intended to reduce depression do so through interventions aimed at various proximal outcomes such as engagement in pleasurable activities (Ly et al., 2014), increasing positive emotions (Tugade & du Pont, 2014), or other pathways.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Ly et al., 2014

      Not checked in the free run.

    • cites Tugade & du Pont, 2014

      Not checked in the free run.

  • Classified as probably not a substantial claim. 18
  • May combine two claims. 17

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisereasoning

    The development of EMIs has been increasing in recent years due to the expansion of technologies such as smartphones and wearables.

    “The development of EMIs has been increasing in recent years due to the expansion of technologies such as smartphones and wearables.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Smartphone membership in the United States increased from 35% in 2011 to 72% in 2016, and is higher in those under age 35 at 92%.

    “Now, however, smartphones are much more pervasive, with membership in the United States increasing from 35% in 2011 to 72% in 2016 (and higher in those under age 35 at 92%) (Poushter, 2016).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Poushter, 2016

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisenone given

    Current smartphones contain an increasingly large complement of embedded sensors and can connect to various external devices that can power deeply personalized EMIs.

    “Furthermore, they contain an increasingly large complement of embedded sensors and can connect to various external devices that can power deeply personalized EMIs.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    EMIs for depression have used diverse conceptual treatment strategies including acceptance and commitment therapy and interpersonal therapy, but the majority have been cognitive-behavioral in focus.

    “EMIs for depression have also made use of diverse conceptual treatment strategies including acceptance and commitment therapy (Ahtinen et al., 2013;Lappalainen et al., 2013;Ly, Dahl, Carlbring, & Andersson, 2012) and interpersonal therapy (Dagoo et al., 2014), but the majority have been cognitive-behavioral in focus”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Ahtinen et al., 2013 source

      Not checked in the free run.

    • cites Dagoo et al., 2014 source

      Not checked in the free run.

    • cites Burns et al., 2011 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    A recent meta-analysis identified 33 EMI studies, 17 of which included a measure of depression as an outcome with levels ranging from mild symptoms to those meeting criteria for major depressive disorder.

    “This analysis identified 33 studies, 17 of which included a measure of depression as an outcome with levels ranging from mild symptoms to those meeting criteria for major depressive disorder.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • May combine two claims.

    Premiseown data

    Supported EMIs had larger effect sizes (g = .73) than stand-alone EMIs (g = .45) and stand-alone EMIs provided in combination with care as usual (g = .38).

    “Supported EMIs had larger effect sizes (g = .73),suggesting that larger or more consistent results can be obtained with some amount of human support compared with stand-alone EMIs (g = .45)and stand-alone EMIs provided in combination with care as usual (g = .38).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    Early EMI research provided proof-of-concept that EMIs could work, but powered by current devices, more recent research is demonstrating increased potential of such interventions.

    “Early EMI research provided proof-of-concept that EMIs could work, but powered by current devices, more recent research is demonstrating increased potential of such interventions.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    EMIs can be effective at reducing symptoms of depression and anxiety as well as related outcomes of stress and at increasing positive psychological functioning.

    “Previous work has demonstrated that EMIs can be effective at reducing symptoms of depression and anxiety as well as related outcomes of stress and at increasing positive psychological functioning.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

    • cites Gee et al., 2016 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisenone given

    EMIs differ from traditional face-to-face treatments and other digital treatments such as Internet treatments in the design of such treatments and the evaluation of their efficacy.

    “EMIs differ from traditional face-to-face treatments, as well as other digital treatments such as Internet treatments, in several ways including the design of such treatments and the evaluation of their efficacy.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The computing power of a current smartphone is several times more powerful than Cray-2 supercomputer, the fastest computer in the world in the 1980s.

    “In fact, the computing power of a current smartphone is several times more powerful than Cray-2 supercomputer, which was the fastest computer in the world in the 1980s (Processing power compared, 2017).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Processing power compared, 2017

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    The fact that the sum of these aspects may evolve over the course of a trial means that traditional RCTs may be a poor fit for early stages of EMI development.

    “The fact that sum of these aspects may evolve over the course of a trial (e.g., the decision rules may be updated based on new information gained) means that traditional RCTs may be a poor fit for early stages of EMI development.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    EMIs for anxiety share similar conceptual diversity and similar efficacy to those for depression.

    “EMIs for anxiety share a lot of similarity to those for depression.They have similar conceptual diversity and similar efficacy.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    Current EMI interventions can be somewhat clunky, usually pulling from a prepopulated option created without knowledge of each user and each context.

    “interventions can be somewhat clunky usually pulling from a prepopulated option created without knowledge of each user and each context.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Levels of anxiety in EMI studies ranged from mild to clinical ranges.

    “Levels of anxiety again ranged from mild to clinical ranges of anxiety.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    Group-level models can work from the start for a new user but may ultimately be less accurate.

    “Grouplevel models can work from the start for a new user but may ultimately be less accurate.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    Individual-level models require experience with a specific user before predictions can be made but are more accurate once data are gathered.

    “For example, individuallevel models require experience with a specific user before predictions can be made but are more accurate once data are gathered.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    When EMIs include learning models, no single treatment is given to all individuals; treatments represent a collection of intervention interactions aimed to optimize outcome for each given individual.

    “When EMIs include learning models, then no single treatment is given to all individuals.Instead, treatments represent a collection of intervention interactions aimed to optimize outcome for each given individual.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    Many current EMIs are somewhat limited.

    “However, many current EMIs are somewhat limited.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisereasoning

    This is different than RCTs that make inferences on means (i.e., group-level analyses) first before conducting within-person or moderator analyses.

    “This is different than RCTs that make inferences on means (i.e., group-level analyses) first before conducting within-person or moderator analyses that might attempt to make recommendations to individuals.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Mixed results were found for other anxiety disorders, but only three studies investigated these (two for panic disorder and one for social anxiety).

    “Mixed results were found for the other anxiety disorders, but only three studies investigated these (two for panic disorder and one for social anxiety).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Gee et al., 2016 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Interpretationown data

    The low quality of EMI studies suggests that results might be interpreted with caution.

    “This is perhaps unsurprising, given that EMI methodology is still in its infancy; however, it suggests that results might be interpreted with caution.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisenone given

    On a group level, what is being evaluated for an EMI is the sum of the interventions, tailoring variables, decision variables, and decision rules.

    “It is worth noting that on a group level what is being evaluated for an EMI is the sum of the interventions, tailoring variables, decision variables, and decision rules.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisereasoning

    Simple EMIs use static decision rules to provide interventions to people, but more complex EMIs often use algorithms to optimize and personalize systems over time.

    “Simple EMIs use static decision rules to provide interventions to people, but more complex EMIs often use algorithms to optimize and personalize systems overtime.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A bandit algorithm changes the likelihood of presentation of a particular intervention over time based on the impact of that intervention in the past on proximal outcomes.

    “One example, is a bandit algorithm that changes the likelihood of the presentation of a particular intervention overtime based on the impact of that intervention in the past on proximal outcomes (Rabbi et al., 2016).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Rabbi et al., 2016

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Significant technical work has advanced the use of sensors to detect anxious states, opening the possibility of triggering and tailoring interventions using passive data collection, which reduces burden on the user.

    “This is despite significant technical work advancing the use of sensors to detect anxious states, which opens the possibility of triggering and tailoring interventions using passive data collection, which reduces the burden on the user (Huang et al., 2016;Miranda, Calderon, & Favela, 2014;Rennert & Karapanos, 2013).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Huang et al., 2016 source

      Not checked in the free run.

    • cites Miranda, Calderon, & Favela, 2014

      Not checked in the free run.

    • cites Rennert & Karapanos, 2013

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    Current evidence supports that EMIs can be beneficial to reduce depression and anxiety and associated aspects of mental health such as stress, acceptance, and quality of life.

    “Current evidence supports that EMIs can be beneficial to reduce depression and anxiety and associated aspects of mental health such as stress, acceptance, and quality of life (Gee et al., 2016;Versluis et al., 2016).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Gee et al., 2016 source

      Not checked in the free run.

    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    In a previous review, 15 studies examined anxiety symptoms as an outcome and found a within-person pre-post effect size of g = .47 (95% CI 0.32-0.63), nearly identical to that found for depression.

    “In the previous review discussed, 15 studies examined anxiety symptoms as an outcome and found a within-person pre-post effect size of g = .47(95% CI 0.32-0.63),which was nearly identical to that found for studies with depression as an outcome (Versluis et al., 2016).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Versluis et al., 2016 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Advances in natural language processing and machine learning are increasingly being used in mental health to create treatments that can interact and adapt to individual people.

    “These techniques are increasingly being used in mental health to create treatments that can interact and adapt to individual people (Calvo, Milne, Hussain, & Christensen, 2017).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Calvo, Milne, Hussain, & Christensen, 2017

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Interpretationnone given

    Many current EMIs require some initiation by the user either in the form of a request or assessment.

    “Many require some initiation by the user either in the form of a request or assessment”
    Quote in the paperSubstantial claim
29 of 29

  • The paper relies on this without support.

    EMIs are more dynamic and complex than traditional treatments.

    “One reason they are overlooked is that it is difficult to translate clinical guidelines or empirical findings from traditional interventions into decision rules for EMIs because EMIs are more dynamic and complex than traditional treatments.”

  • The paper relies on this without support.

    EMIs are deployed in people's real lives and as such often require a great deal of data collection to inform their deployment and evaluation.

    “EMIs are deployed in people's real lives and as such often require a great deal of data collection to inform their deployment and evaluation.”

  • The paper relies on this without support.

    Multiple variables likely contribute to the impact of any given intervention.

    “One challenge with such approaches is that multiple variables (e.g., time of day, recent stressors, previous practice of skills) likely contribute to the impact of any given intervention.”
3 of 3

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Ecological momentary assessments have been shown to be a valuable mood-tracking tool in digital therapeutics aiming to reduce levels of anxiety and depression.

    “EMAs have been shown to be a valuable mood-tracking tool in the context of digital therapeutics aiming to reduce levels of anxiety and depression [46].”

    From Development and feasibility of a digital acceptance and commitment therapy⇓based intervention for generalized anxiety disorder: Pilot acceptability study (2021) · cites this paper as [46]

    The Introduction says EMA brings observation into daily life, and §3 (EMIs for Depression) reports that daily mood questions are useful proxies for a depression assessment. The paper also discusses digital interventions for depression and anxiety, but the cited mood-tracking evidence concerns depression; it does not establish EMA mood tracking as valuable for anxiety treatment. The copy provides no page numbers for these passages. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

  • Promising results have already been shown for the application of EMA and EMI to anxiety disorders and stress-related disorders, highlighting the potential of these tools to provide psychological support in daily life and to investigate symptom fluctuations across time.

    “Still, promising results have been already shown also regarding the application of EMA and EMI to anxiety disorders [93,94] and stress-related disorders [95,96], highlighting the potential of these tools to provide psychological support in daily life and to investigate symptom fluctuations across time.”

    From Current state and future directions of technology-based ecological momentary assessment and intervention for major depressive disorder: a systematic review (2019) · cites this paper as [94]

Cites

Flagged first, then by citations

Research works printed but not linked: 4

  1. Grassi, 2007, A relaxing journey: The use of mobile phones for well-being improvement, Annual Review of Cyber Therapy and Telemedicine, 5, 123
  2. Poushter, 2016, Smartphone ownership and Internet usage continues to climb in emerging economies
  3. Rainie, 2015, Americans' views on mobile etiquette
  4. Tugade, 2014, The effects of daily positive emotion reminders on emotional complexity, physical health, and cognitive reasoning

Cited by

Sorted by citations

Related

Authors
Schueller, S. M., Aguilera, A., & Mohr, D. C.
Affiliations
  • Northwestern University (US)
  • School of Social Welfare; University of California, Berkeley; Berkeley CA USA
Corresponding author
Schueller, Stephen M.
ORCID
Venue
Depression and Anxiety
Publisher
Wiley
Published
2017-05-11
Volume
34
Issue
6
Pages
540-545
ISSN
1091-4269 (ISSN-L)
DOI
10.1002/da.22649
OpenAlex
W2613845253
Cited by
168
References
36
Citations by year
2026: 13 · 2025: 24 · 2024: 12 · 2023: 13 · 2022: 19 · 2021: 25 · 2020: 41 · 2019: 10 · 2018: 10 · 2017: 1
FWCI
1.96
Type
journal article
Language
en
Open access
Licence
Locations
  • Depression and Anxiety (publishedVersion)
  • PubMed (publishedVersion)
  • eScholarship (California Digital Library) (submittedVersion)
Funders
National Institute of Mental Health (K08MH102336, K23MH094442, P20MH090318)
Topics
  • Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Mental Health Research Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
Keywords
ecological momentary intervention, depression, anxiety, digital mental health interventions, smartphone sensors, natural language processing, machine learning, user burden, stress, intervention design and evaluation
MeSH terms
Ecological Momentary Assessment, Anxiety, Depression, Humans, Therapy, Computer-Assisted

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review