Concept
mHealth
11 studies
mHealth uses mobile apps and phones to deliver assessment, coaching, or treatment support.
App claims are loud; trial and field evidence here separates usable tools from marketing.
Studies
11
Findings
9
11 supporting · 0 challenging · 3 qualifying citations
Open tensions
2
Currently
What we know
- Multiple studies in this library examine mhealth with empirical patient or population outcomes rather than opinion alone.
- Past-month use occasions, AUDIT-C, DAST-10, PHQ-8, and GAD-7 improved (all P<.05). In-app craving ratings fell roughly by half by weeks 8–9. Engagement averaged about 15.7 use days over 8 weeks. Authors call the approach feasible and acceptable but note the need for controlled efficacy trials.
- 3070 people were randomized. The intervention × time interaction was highly significant: adjusted WEMWBS differences favored MoodGYM by about 2.5 points at 6 weeks and 2.9 at 12 weeks (≈Cohen d 0.34). Attrition was much higher in the intervention arm (73.5%) than control (26.9%).
- Audio coaching increased fruit intake by about three pieces per week versus text and control; vegetable intake had no significant main effect of condition (literacy interactions noted).
- PeR improved self-efficacy and quality of life and eased symptoms; health improvement was 2.24× higher with high vs low Ex-SRES; dropout was 11.3%.
Largest unresolved question
Effect sizes and settings differ across mhealth studies — digital vs clinic, trial vs observational — so results should not be pooled casually.
Common misconceptions
mHealth findings always generalise to every clinic.
These studies are context-bound; designs, populations, and endpoints limit transfer.
Related
Claim ledger
What the evidence shows
Drawn from 11 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
Multiple studies in this library examine mhealth with empirical patient or population outcomes rather than opinion alone.
Study Role Design N Population Outcome Woebot chatbot for substance use Supports Human experimentSingle-group 8-week pre–post usability/feasibility evaluation (no control) N=101 · 101 enrolled; 51 completed the post-treatment survey Adults initiating a therapeutic chatbot for problematic substance use Past-month use occasions, AUDIT-C/DAST-10, cravings, PHQ-8, GAD-7 over 8 weeks MoodGYM improves population wellbeing Supports RCTFully automated MoodGYM vs waiting-list; ITT mixed models N=3070 · Randomized NHS Choices users; high differential attrition NHS Choices website users randomized to MoodGYM or wait-list WEMWBS well-being at 6 and 12 weeks Audio vs text produce-app RCT Supports RCTThree-arm RCT: text-tailored vs audio-tailored vs questionnaire control N=146 · Final analytic sample Adults using a fruit-and-vegetable mobile app intervention Fruit and vegetable intake at 6 months Past-month use occasions, AUDIT-C, DAST-10, PHQ-8, and GAD-7 improved (all P<.05). In-app craving ratings fell roughly by half by weeks 8–9. Engagement averaged about 15.7 use days over 8 weeks. Authors call the approach feasible and acceptable but note the need for controlled efficacy trials.
3070 people were randomized. The intervention × time interaction was highly significant: adjusted WEMWBS differences favored MoodGYM by about 2.5 points at 6 weeks and 2.9 at 12 weeks (≈Cohen d 0.34). Attrition was much higher in the intervention arm (73.5%) than control (26.9%).
Audio coaching increased fruit intake by about three pieces per week versus text and control; vegetable intake had no significant main effect of condition (literacy interactions noted).
PeR improved self-efficacy and quality of life and eased symptoms; health improvement was 2.24× higher with high vs low Ex-SRES; dropout was 11.3%.
Intervention participants lost more weight than comparison participants (about 1.97 kg more; adjusted losses ~2.88 kg vs 0.91 kg).
A 7-week college stress app can beat waitlist on perceived stress, with many first-time help-seekers.
In 150 stressed college students, StudiCare Stress reduced perceived stress versus waitlist (d=0.69); 69% vs 35% were stress responders (NNT=2.89). Anxiety d=0.76 and depression d=0.63; 76.8% were first-time help-seekers. Effects held at 3 months, with higher intervention dropout.
- MoodGYM improves population wellbeing— large UK adult MoodGYM well-being RCT with high attrition — not a college-stress waitlist trial
Study Role Design N Population Outcome Can an internet app cut college-student stress? Supports RCTRCT of internet- and app-based StudiCare Stress with feedback-on-demand vs waitlist control (both with treatment as usual) in German college students N=150 · 150 students with PSS-4 ≥8 randomised 75:75; posttest at 7 weeks, follow-up at 3 months College students with elevated perceived stress (PSS-4 ≥8) Perceived stress (PSS-4) at 7 weeks; secondary mental-health, academic, and help-seeking outcomes MoodGYM improves population wellbeing Qualifieslarge UK adult MoodGYM well-being RCT with high attrition — not a college-stress waitlist trial RCTFully automated MoodGYM vs waiting-list; ITT mixed models N=3070 · Randomized NHS Choices users; high differential attrition NHS Choices website users randomized to MoodGYM or wait-list WEMWBS well-being at 6 and 12 weeks Phone sensors can track daily movement and usage that correlate with PHQ-9 severity.
In 28 adults carrying a sensor phone for 2 weeks, less regular GPS movement and more phone use tracked higher PHQ-9 (circadian movement r=−.63; usage duration r=.54). Normalized entropy classified PHQ-9 ≥5 vs <5 at 86.5% accuracy.
- Can an internet app cut college-student stress?— StudiCare is a treatment RCT on perceived stress, not a 2-week GPS/PHQ-9 sensor correlation
Study Role Design N Population Outcome Can phone sensors flag how depressed someone is? Supports Cross-sectional2-week exploratory sensor study: community adults carried Purple Robot; PHQ-9 at baseline correlated with GPS/usage features N=28 · 40 recruited; 28 with sufficient sensor data (18 GPS, 21 phone usage); 12 excluded for <50% data coverage US community adults (19–58 years; 20 women, 8 men in the analyzed set) recruited via Craigslist Correlation of GPS/phone-usage features with PHQ-9 severity; classification of PHQ-9 ≥5 vs <5 Can an internet app cut college-student stress? QualifiesStudiCare is a treatment RCT on perceived stress, not a 2-week GPS/PHQ-9 sensor correlation RCTRCT of internet- and app-based StudiCare Stress with feedback-on-demand vs waitlist control (both with treatment as usual) in German college students N=150 · 150 students with PSS-4 ≥8 randomised 75:75; posttest at 7 weeks, follow-up at 3 months College students with elevated perceived stress (PSS-4 ≥8) Perceived stress (PSS-4) at 7 weeks; secondary mental-health, academic, and help-seeking outcomes A 6-month shared-decision-making app raised activation in schizophrenia care, not symptoms or admissions.
In 194 Danish outpatients, 6 months of an SDM app plus usual care improved patient activation (mean difference 4.39; d=0.33), communication confidence, and decision preparedness, with no effects on symptoms, hope, or admissions.
- Can phone sensors flag how depressed someone is?— passive GPS/usage correlates of PHQ-9 ≠ an SDM activation RCT in schizophrenia
Study Role Design N Population Outcome Can a phone app help people with schizophrenia join treatment decisions? Supports RCTAssessor-blinded 2-arm parallel multicenter RCT (Momentum): 6-month smartphone SDM app plus TAU vs TAU; 1:1 block randomization N=194 · 194 participants with schizophrenia-spectrum disorders from 9 outpatient sites Outpatients with schizophrenia, schizotypal, or delusional disorder in the Capital Region of Denmark Primary: self-reported patient activation post-intervention; secondary: communication confidence, decision preparedness, symptoms, functioning, admissions Can phone sensors flag how depressed someone is? Qualifiespassive GPS/usage correlates of PHQ-9 ≠ an SDM activation RCT in schizophrenia Cross-sectional2-week exploratory sensor study: community adults carried Purple Robot; PHQ-9 at baseline correlated with GPS/usage features N=28 · 40 recruited; 28 with sufficient sensor data (18 GPS, 21 phone usage); 12 excluded for <50% data coverage US community adults (19–58 years; 20 women, 8 men in the analyzed set) recruited via Craigslist Correlation of GPS/phone-usage features with PHQ-9 severity; classification of PHQ-9 ≥5 vs <5
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
Effect sizes and settings differ across mhealth studies — digital vs clinic, trial vs observational — so results should not be pooled casually.
Effect sizes and settings differ across mhealth studies — digital vs clinic, trial vs observational — so results should not be pooled casually.
Study Role Design N Population Outcome Woebot chatbot for substance use Supports Human experimentSingle-group 8-week pre–post usability/feasibility evaluation (no control) N=101 · 101 enrolled; 51 completed the post-treatment survey Adults initiating a therapeutic chatbot for problematic substance use Past-month use occasions, AUDIT-C/DAST-10, cravings, PHQ-8, GAD-7 over 8 weeks MoodGYM improves population wellbeing Supports RCTFully automated MoodGYM vs waiting-list; ITT mixed models N=3070 · Randomized NHS Choices users; high differential attrition NHS Choices website users randomized to MoodGYM or wait-list WEMWBS well-being at 6 and 12 weeks Audio vs text produce-app RCT Supports RCTThree-arm RCT: text-tailored vs audio-tailored vs questionnaire control N=146 · Final analytic sample Adults using a fruit-and-vegetable mobile app intervention Fruit and vegetable intake at 6 months
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
StudiCare’s college waitlist stress effects, MoodGYM’s adult well-being RCT, a 2-week GPS/PHQ-9 sensor study, and a schizophrenia SDM activation trial are all digital mental-health papers, not one pooled app effect.
StudiCare’s college waitlist stress effects, MoodGYM’s adult well-being RCT, a 2-week GPS/PHQ-9 sensor study, and a schizophrenia SDM activation trial are all digital mental-health papers, not one pooled app effect.
- Can an internet app cut college-student stress?
- MoodGYM improves population wellbeing
- Can phone sensors flag how depressed someone is?
- Can a phone app help people with schizophrenia join treatment decisions?
Study Role Design N Population Outcome Can an internet app cut college-student stress? Supports RCTRCT of internet- and app-based StudiCare Stress with feedback-on-demand vs waitlist control (both with treatment as usual) in German college students N=150 · 150 students with PSS-4 ≥8 randomised 75:75; posttest at 7 weeks, follow-up at 3 months College students with elevated perceived stress (PSS-4 ≥8) Perceived stress (PSS-4) at 7 weeks; secondary mental-health, academic, and help-seeking outcomes MoodGYM improves population wellbeing Supports RCTFully automated MoodGYM vs waiting-list; ITT mixed models N=3070 · Randomized NHS Choices users; high differential attrition NHS Choices website users randomized to MoodGYM or wait-list WEMWBS well-being at 6 and 12 weeks Can phone sensors flag how depressed someone is? Supports Cross-sectional2-week exploratory sensor study: community adults carried Purple Robot; PHQ-9 at baseline correlated with GPS/usage features N=28 · 40 recruited; 28 with sufficient sensor data (18 GPS, 21 phone usage); 12 excluded for <50% data coverage US community adults (19–58 years; 20 women, 8 men in the analyzed set) recruited via Craigslist Correlation of GPS/phone-usage features with PHQ-9 severity; classification of PHQ-9 ≥5 vs <5 Can a phone app help people with schizophrenia join treatment decisions? Supports RCTAssessor-blinded 2-arm parallel multicenter RCT (Momentum): 6-month smartphone SDM app plus TAU vs TAU; 1:1 block randomization N=194 · 194 participants with schizophrenia-spectrum disorders from 9 outpatient sites Outpatients with schizophrenia, schizotypal, or delusional disorder in the Capital Region of Denmark Primary: self-reported patient activation post-intervention; secondary: communication confidence, decision preparedness, symptoms, functioning, admissions
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
Concept page published
mHealth
- Added a scope qualifier
MoodGYM improves population wellbeing
- Added a supporting study
Can an internet app cut college-student stress?
- Added a supporting study
MoodGYM improves population wellbeing
- Added a supporting study
Can an internet app cut college-student stress?
New claim
In 150 stressed college students, StudiCare Stress reduced perceived stress versus waitlist (d=0.69); 69% vs 35% were stress responders (NNT=2.89). Anxiety d=0.76 and depression d=0.63; 76.8% were first-time help-seekers. Effects held at 3 months, with higher intervention dropout.
New tension
StudiCare’s college waitlist stress effects and MoodGYM’s adult well-being RCT differ in population, endpoint, and attrition; both are digital mental-health trials, not one pooled app effect.
- Added a scope qualifier
Can an internet app cut college-student stress?
- Added a scope qualifier
Can phone sensors flag how depressed someone is?
- Added a supporting study
Can phone sensors flag how depressed someone is?
- Added a supporting study
Can a phone app help people with schizophrenia join treatment decisions?
- Added a supporting study
Can a phone app help people with schizophrenia join treatment decisions?
- Added a supporting study
Can phone sensors flag how depressed someone is?
New claim
In 28 adults carrying a sensor phone for 2 weeks, less regular GPS movement and more phone use tracked higher PHQ-9 (circadian movement r=−.63; usage duration r=.54). Normalized entropy classified PHQ-9 ≥5 vs <5 at 86.5% accuracy.
New claim
In 194 Danish outpatients, 6 months of an SDM app plus usual care improved patient activation (mean difference 4.39; d=0.33), communication confidence, and decision preparedness, with no effects on symptoms, hope, or admissions.
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Coverage edits are distinct from reviewed research changes and corrections. Dates record PaperFren updates; study publication dates appear in the timeline.
In 28 adults carrying a sensor phone for 2 weeks, less regular GPS movement and more phone use tracked higher PHQ-9 (circadian movement r=−.63; usage duration r=.54). Normalized entropy classified PHQ-9 ≥5 vs <5 at 86.5% accuracy.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Can phone sensors flag how depressed someone is?
- Coverage updated · Added a scope qualifier: Can an internet app cut college-student stress?
In 194 Danish outpatients, 6 months of an SDM app plus usual care improved patient activation (mean difference 4.39; d=0.33), communication confidence, and decision preparedness, with no effects on symptoms, hope, or admissions.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Can a phone app help people with schizophrenia join treatment decisions?
- Coverage updated · Added a scope qualifier: Can phone sensors flag how depressed someone is?
StudiCare’s college waitlist stress effects, MoodGYM’s adult well-being RCT, a 2-week GPS/PHQ-9 sensor study, and a schizophrenia SDM activation trial are all digital mental-health papers, not one pooled app effect.
- Coverage updated · Added a supporting study: Can a phone app help people with schizophrenia join treatment decisions?
- Coverage updated · Added a supporting study: Can phone sensors flag how depressed someone is?
In 150 stressed college students, StudiCare Stress reduced perceived stress versus waitlist (d=0.69); 69% vs 35% were stress responders (NNT=2.89). Anxiety d=0.76 and depression d=0.63; 76.8% were first-time help-seekers. Effects held at 3 months, with higher intervention dropout.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Can an internet app cut college-student stress?
- Coverage updated · Added a scope qualifier: MoodGYM improves population wellbeing
StudiCare’s college waitlist stress effects and MoodGYM’s adult well-being RCT differ in population, endpoint, and attrition; both are digital mental-health trials, not one pooled app effect.
- Coverage updated · New tension
- Coverage updated · Added a supporting study: MoodGYM improves population wellbeing
- Coverage updated · Added a supporting study: Can an internet app cut college-student stress?
- Coverage updated · Concept page published
Papers
11 studies in this library bear on mHealth, ordered by citations.
- SMS coaching for weight loss
Daily tailored text/MMS coaching produced about 2 kg more weight loss than print materials over 4 months in overweight adults.
- Can phone sensors flag how depressed someone is?
In 28 adults carrying a sensor phone for 2 weeks, less regular GPS movement and more phone use tracked higher PHQ-9 scores; normalized entropy classified any vs no depressive symptoms at 86.5% accuracy.
- Can an internet app cut college-student stress?
In 150 stressed college students, a 7-week internet- and app-based program (StudiCare Stress) reduced perceived stress versus waitlist (d=0.69), with 69% vs 35% stress responders (NNT=2.89).
- Woebot chatbot for substance use
An 8-week fully automated Woebot SUD program showed pre–post drops in substance use, cravings, depression, and anxiety in a single-group study.
- MoodGYM improves population wellbeing
In a large UK web RCT, automated MoodGYM raised mental well-being versus wait-list control despite high intervention attrition.
- Fitbit wear during an activity trial
Survivors wore Fitbits most days and roughly doubled ActiGraph MVPA over 12 weeks in the intervention arm.
- WeChat rehab helped COPD self-efficacy
A Chinese RCT used WeChat-supported personalized remote pulmonary rehab (PeR) to improve COPD patients’ self-efficacy, quality of life, and symptoms.
- Mobile Immediate Mood Scaler validation
A 22-item mobile Immediate Mood Scaler correlated strongly with standard PHQ-9 and GAD-7 scores in 110 participants.
Show 3 more studiesShow fewer studies
- Tech parenting program cut postpartum distress
A Singapore RCT found a technology-based supportive educational parenting program improved bonding and satisfaction while reducing postnatal depression and anxiety.
- Audio vs text produce-app RCT
A tailored mobile app’s audio messages raised fruit intake more than text or control, while vegetable intake showed no overall main effect.
- Can a phone app help people with schizophrenia join treatment decisions?
In 194 Danish outpatients, 6 months of a shared-decision-making app plus usual care improved patient activation (d = 0.33), communication confidence, and feeling prepared to decide, but not symptoms, hope, or admissions.
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
Effect sizes and settings differ across mhealth studies — digital vs clinic, trial vs observational — so results should not be pooled casually.
StudiCare’s college waitlist stress effects, MoodGYM’s adult well-being RCT, a 2-week GPS/PHQ-9 sensor study, and a schizophrenia SDM activation trial are all digital mental-health papers, not one pooled app effect.
Study this conceptflashcards and short-answer questions
What is mHealth and why do health students study it?
mHealth uses mobile apps and phones to deliver assessment, coaching, or treatment support. App claims are loud; trial and field evidence here separates usable tools from marketing.
Name one limit of the evidence base for mHealth in this library.
Single-setting trials, observational designs, or digital-only samples limit causal and external claims.
Flashcards
0 of 8 answers reviewed