Explore this article's sources with AI
Follow PeptideMethods on Google
Digital Support With GLP-1 Drugs
A retrospective program study linked app use with higher odds of weight-loss milestones, but self-selection prevents a causal conclusion.
Why we wrote this. Digital program results can sound causal when engaged users select the support. This readout separates adjusted associations from a randomized app effect.
In this article (5 sections)
A retrospective Australian study found that adults using a smartphone support app alongside semaglutide or tirzepatide had higher odds of reaching several weight-loss thresholds than adults in the same direct-to-consumer program who used medication without the app. The finding is an association, not evidence that the app caused the difference[1]. People chose whether to use the app, so motivation and engagement could explain part of the result.
The analysis included 6,113 adults who self-enrolled in the Mosh direct-to-consumer weight-loss program between July 2024 and August 2025. Participants received semaglutide or tirzepatide and had optional access to a smartphone app plus clinician support. Eligibility for this analysis required a baseline weight and at least one later recorded weight[1]. That requirement can select for people who remained engaged enough to report follow-up data.
What the researchers compared
The main models compared the odds of losing at least 5%, 10%, or 15% of baseline weight around 60, 120, and 180 days. The researchers adjusted for medication type and several recorded participant characteristics, including age, gender, ethnicity, and baseline BMI[1]. Adjustment can reduce measured differences between groups, but it cannot balance unmeasured motivation, income, health literacy, or prior experience with digital tools.
A second set of mixed-effects models examined weight trajectories and particular app tools over time. These models included linear and curved time terms to allow the rate of weight change to vary. The study therefore asked two related questions: whether app users crossed clinically familiar thresholds more often, and whether use of specific tools tracked with faster change.
This was not a randomized app trial. Participants self-selected into a commercial program and then chose whether to use optional digital support. The medication-only group may have differed at baseline or during follow-up in ways the available variables did not capture. Reverse causation is also possible if people who were already losing weight became more willing to log progress.
The threshold results
Across the reported timepoints, app users had adjusted odds ratios from 1.38 to 2.22 for reaching at least 5% weight loss compared with medication-only participants. The confidence intervals ranged from 1.09 to 1.75 at one end and from 1.40 to 3.45 at the other[1]. Odds ratios describe relative odds, not the percentage of all users who reached a target.
For at least 10% weight loss, adjusted odds ratios ranged from 1.67 at day 120 to 1.97 at day 180. The corresponding 95% confidence intervals were 1.33 to 2.10 and 1.35 to 2.90[1]. At 180 days, the adjusted odds ratio for losing at least 15% was 1.99, with a 95% confidence interval from 1.26 to 3.29[1].
Those relative estimates sound large, but the authors described differences in overall weight trajectories as modest. This is not contradictory. A small average shift can move more people across a fixed threshold when many participants cluster near that cutoff. Threshold results can therefore look stronger than the average trajectory difference.
The abstract does not provide the unadjusted proportion reaching every threshold in each group. Without those absolute numbers, readers cannot calculate a straightforward absolute benefit or number needed to engage with the app. The odds ratio is informative, but it should not be translated into claims that app users lost twice as much weight.
Which app tool stood out
Progress logging was associated with an additional 0.26 percentage points of weight loss per 30 days, with a 95% confidence interval from 0.14 to 0.38 percentage points. The model's p value was below 0.001[1]. No other app tools were consistently associated with changes in weight-loss trajectories.
Logging may help some users notice patterns and maintain attention. It can also be a marker of persistence rather than the active ingredient. A participant who opens an app regularly, reports weight, attends follow-up, and continues medication may differ from someone who does none of those things. The model cannot fully separate the act of logging from the person likely to log.
The study combined people receiving semaglutide with those receiving tirzepatide. Medication type was included in adjustment, but this was not a head-to-head medicine trial and does not establish that digital support works equally with each drug. It also does not compare different app designs or levels of clinician contact.
How this differs from drug trials
Randomized drug trials answer a different question. SURMOUNT-1 assigned 2,539 adults to tirzepatide or placebo for 72 weeks, with mean weight changes from 15.0% to 20.9% downward across tirzepatide groups and 3.1% downward with placebo[2]. That design estimates the medicine's effect under trial conditions. It does not isolate the added effect of a commercial smartphone program.
SURMOUNT-1 also recorded adverse events and discontinuation prospectively. Gastrointestinal events were most common, and discontinuation because of adverse events ranged from 4.3% to 7.1% across tirzepatide groups versus 2.6% with placebo[2]. The retrospective digital study focused on recorded weight outcomes and cannot substitute for randomized safety evidence.
The direct-to-consumer setting creates further limits. People willing and able to self-enroll in a paid or commercial digital program may differ from patients in public clinics, primary care, or specialist obesity services. Australia-specific access and prescribing arrangements may also limit transfer to other health systems.
What we do not know yet
The study does not show whether app engagement caused sustained weight differences after 180 days. It does not establish which behavioral content, clinician interaction, reminders, or self-monitoring routines would remain effective in a randomized comparison. Attrition and missing weights may also change estimates if follow-up differed between groups.
A useful next trial would randomly assign comparable medication users to defined digital support or usual follow-up, measure engagement before outcomes occur, and report absolute threshold rates alongside mean change. It should also track quality of life, adverse events, medication persistence, and outcomes after people stop using the app.
The practical conclusion is modest. App use, especially progress logging, marked a more successful group within this program, but the analysis cannot tell us how much of that difference the technology produced. Digital support may be useful, yet it should not be marketed as a proven multiplier of semaglutide or tirzepatide efficacy until randomized evidence isolates its contribution[1].
Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.
Frequently asked
Did the smartphone app cause more weight loss?
The study cannot establish that. App use was optional in a retrospective program, so motivation, engagement, and unmeasured differences could explain part of the association.
Which app tool was linked to weight loss?
Progress logging was associated with 0.26 additional percentage points of weight loss per 30 days. Other tools were not consistently associated with trajectory changes.
Were both semaglutide and tirzepatide included?
Yes. Medication type was included in statistical adjustment, but the study was not designed to compare the medicines or prove equal app effects with each one.
Sources
- [1]Digital Augmentation of Glucagon-Like Peptide-1 for Obesity Management: Retrospective Study. Journal of Medical Internet Research. 2026. PMID 42976342.Tier 1 · primary↩
- [2]A Study of Tirzepatide (LY3298176) in Participants With Obesity or Overweight. ClinicalTrials.gov. NCT04184622.Tier 1 · primary↩
No revisions yet. First published .