Yes, on the evidence we have. A structured programme puts a clinician, regular review and a record around weight-loss medication.
In other long-term conditions, text reminders roughly doubled the odds of people taking their medicine as prescribed across 16 randomised trials. [1] The National Institute for Health and Care Excellence (NICE) tells clinicians to reinforce adherence through regular review of weight-management medicines. [2]
Two UK digital programmes have reported figures, and most people on them stayed on treatment. One is our own group’s programme (disclosed below): 91.4% retention at 6 months among 8,127 women with no comparison group. [3] The other reported its own conference data, not yet peer-reviewed: 65% to 84% of 180 people still collecting their medicine at 48 weeks. [4]
Adherence means taking a medicine as prescribed for as long as you and your clinician decide. Persistence means still being on it at a given point in time. A randomised trial assigns people to groups by chance, which is the fairest test of whether something works.
We found no study that compared a structured programme with care that lacks one for weight-loss medication. This article sets out the three strands that do exist.
How many people stop weight-loss medication in routine care?
Most people in a large 2025 US study of routine health records had stopped within a year. Of 125,474 adults with overweight or obesity starting one of these medicines, 64.8% of those without type 2 diabetes had stopped by one year. [5] Stopping meant 60 days or more without medicine on hand. [5]
Statistics describe populations, not any one person’s results.
In the US, persistence has risen since 2021. Among insured adults without diabetes, it was 33.2% for people starting in 2021 and 60.9% for people starting in the first half of 2024. [6] The authors suggest possible reasons: the end of supply shortages, better dose and side-effect management and lifestyle programmes. [6]
Why do people stop?
In one US health system, the commonest recorded reason for stopping within a year was cost or insurance. It applied to 47.6% of a random sample of 288 people. [7]
Side effects accounted for 14.6% and supply shortages for 11.8%. [7] Unsatisfactory weight loss was the reason for 1.7%. [7]
Of those reasons, side effects are the one a clinical review can act on. NICE asks clinicians to discuss them with you and to consider adjusting the dose. [9]
Side effects also show up in the 125,474-adult US study. Moderate or severe gastrointestinal side effects were linked with a higher rate of stopping. [5] Among people without type 2 diabetes, each 1% of weight lost was linked with a 3.3% lower rate of stopping. [5]
These are associations, not causes. A study that follows people over time can show what goes together. It cannot show what causes what.
Statistics describe populations, not any one person’s results.
How are treatment length and weight loss linked?
The link between treatment length and weight loss runs both ways. People who lose more weight are more likely to carry on. [5] People who carry on longer have lost more by the time they are measured. [8]
These are associations, not causes.
What does a structured programme do?
A structured programme puts a clinician, a review and a record around the medicine. NICE’s guidance for weight-management medicines asks clinicians to monitor their effect and “reinforce behavioural advice and adherence through regular review”. [2]
Its medicines-adherence guideline fills in the detail. NICE writes clinical guidance for the NHS. It says non-adherence is common and most people are non-adherent sometimes, and asks clinicians to assess it without judgement. [9]
The guideline lists practical supports for an identified need. “Suggesting that patients record their medicine-taking” and “encouraging patients to monitor their condition” are the first two. [9] Where side effects are a problem, it tells clinicians to discuss how you want to deal with them and to consider adjusting the dose. [9]
The guideline adds that no single intervention suits everyone, and that support should match the difficulty you actually have. [9] It also says adherence support needs should be reviewed regularly, because people use medicines long term. [9]
What did the trials at lowest risk of bias have in common?
Ongoing support from a health professional. A Cochrane review of 2014 (an independent systematic review of trials) covered 182 randomised trials of adherence support. [10]
Trials at lowest risk of bias are the ones designed so their results are least likely to mislead. Those trials generally used complex interventions with tailored, ongoing support from professionals such as pharmacists. [10] That support often meant intense education, counselling or daily treatment support. [10]
The same review is candid about the limits. People prescribed self-administered medicines typically take only about half their doses, and effects across the trials were inconsistent. [10] The trials at lowest risk of bias tested support that was tailored and kept going rather than one-off, and even those had mixed results. [10]
What is the evidence that a structured programme helps?
Three strands point the same way: randomised trials in other long-term conditions, UK and international guidance and the figures from UK programmes. Each has limits, stated as it goes.
Do reminders and check-ins work in trials?
A 2016 pooled analysis of 16 randomised trials in other long-term conditions found that text-message reminders roughly doubled the odds of adherence. [1]
Assume that half of people take their medicine as prescribed to begin with. On that basis the authors put the rate with text messages at 67.8%, a rise of 17.8 percentage points. [1]
The trials were short, with a median length of 12 weeks. [1] Self-report was the most common way of measuring adherence. [1]
Apps built to support medicine-taking show a similar effect. A 2020 analysis pooled 9 randomised trials (988 people analysed). [11] App users had about twice the odds of reporting that they took their medicine. [11]
The conditions studied included cardiovascular disease, depression and Parkinson’s disease. [11] The evidence was of moderate quality and the longest follow-up was 16 weeks. [11] Average trial results are not a prediction of individual results.
What does the guidance say?
NICE builds regular review into the use of weight-management medicines. [2] Its adherence guideline lists a record of medicine-taking and monitoring your condition among the supports. [9]
The World Health Organization’s guideline of December 2025 covers the long-term use of these weight-loss medicines. [12] It recommends offering intensive behavioural support alongside the medicine, on low-certainty evidence. [12]
Both of its recommendations are graded conditional. [12] A conditional grade means the guideline group was less certain the recommendation suits everyone.
Its summary cites limited long-term data, cost, health-system readiness and equity. [12] It also says medication alone cannot solve the obesity burden. [12]
What did the UK study with weekly follow-up report?
Two UK digital programmes have reported how many people stayed on treatment, each on its own data. The first is a study led by University College London and published in Nature Medicine in October 2026. [3] It followed 8,127 women on a programme run jointly by SheMed and eMed. [3]
HeMed is a trading name of SheMed Limited, and eMed provides HeMed’s clinical services. The paper’s authors include employees of both companies. [3] So this is a study of our own group’s programme, and HeMed has published no retention figure of its own.
The programme combined compulsory blood-marker monitoring, weekly video-supported follow-up and dose changes guided by a clinician. [3] Retention was 91.4% at 6 months, in an observational study with no comparison group. [3] The abstract does not say how retention was defined.
The people studied paid for their own treatment, and follow-up was 6 months. [3] An observational study records what happened to one group. It cannot show whether the programme or any part of it was what kept people on treatment.
What did the second UK programme report?
The second programme reported its own conference data, covered by the trade press in October 2026 and not yet peer-reviewed. [4] It followed 180 people on NHS-funded and self-pay pathways for 48 weeks. [4] At 48 weeks, 84% of the NHS-funded group and 65% of the self-pay group were continuing with their medication. [4]
The programme defined continuing as a recorded dispensing within 8 weeks of the 48-week point. It says itself that this does not confirm that every dose was taken.
The two groups were not randomised and differed at the start. Neither programme was compared with care outside a programme.
Statistics describe populations, not any one person’s results.
Does weighing yourself at home help?
Within a programme, it may. A 2015 review of randomised trials of behavioural weight-loss programmes pooled the four that added self-weighing and other self-monitoring to a programme. [13]
Those four found 1.7 kg more weight loss on average than the same programme without it. [13] Average trial results are not a prediction of individual results.
Self-weighing on its own showed no evidence of effect in the one trial that tested it without a programme. [13] Programmes where people knew a therapist or researcher was checking that they weighed themselves did better, although only two trials lacked that check. [13] There is some evidence that the measurement worked better when someone else was watching for it. [13]
Where does the evidence stop?
The evidence stops at the comparison. We found no study that compared a structured programme with care that lacks one for weight-loss medication, so the size of any benefit is unknown.
The trials of reminders and apps come from other conditions and were short. [1] [11] The text-message trials ran a median of 12 weeks, and no app trial followed people beyond 16 weeks. [1] [11]
The persistence figures are from the US. [5] The UK programme figures come from a women-only study of our own group’s programme and from one programme’s conference data. [3] [4]
We found no independent UK figure for persistence on private treatment. The two UK programme figures above are each the programme’s own.
NICE has set up evidence generation in the NHS. It allows a set of digital weight-management programmes to be used while more evidence is generated over 4 years. [14] It lists adherence and completion rates among the evidence still needed, including reasons for stopping. [14]
How does HeMed fit in?
HeMed starts with a blood test. We measure 10 metabolic markers across four systems from a needle-free blood test at home.
A clinician reviews your blood test first. If they approve treatment, we deliver it the next working day, discreetly packaged, with delivery included in your plan.
A weekly in-app check-in, reviewed by clinicians, so your plan can be adjusted when it needs to be. A retest every 6 months shows what’s actually changing.
You reach the clinical team through the app. To report side effects, open the help menu (the question mark icon), choose Report side effects and tick the box to be contacted. The clinical team then gets in touch.
You can start a live check-in with the team in the app, 24 hours a day, 7 days a week. If you feel seriously unwell or it is an emergency, call 999.
Treatment is prescription-only and is prescribed only where a clinician decides it is suitable. Whether it’s right for you is a clinical decision, made from your health information and your baseline blood test.
What should you track between check-ins?
Track five things: each dose, the dose level, side effects, a weekly weight and your questions for the next check-in. Record your weight once a week and the rest as they happen. Keep the record short enough that you keep making it.
| What | How often | Why it helps |
|---|---|---|
| Day and time of each dose | Each dose | The record NICE lists as an adherence support [9] |
| Dose level | Each dose, and whenever it changes | NICE asks clinicians to monitor the effect of the medicine at review [2] |
| Side effects, with the date | When they happen | A leading reason people stop; a dated note informs the review [7] [9] |
| Weight | Once a week, same scales, same time of day | No evidence that daily beat weekly in trials of behavioural programmes [13] |
| Questions for your next check-in | As they come up | Regular review is where NICE says adherence is reinforced [2] |
Use whatever you will keep using: your phone’s notes, a calendar reminder or a paper diary. The rule for a late or missed dose depends on the medicine and on how late you are. Check the patient information leaflet for your medicine or ask your clinician before you take a late dose.
Frequently asked questions
Does a structured programme help people take weight-loss medication as prescribed?
Yes, on the evidence we have. In other long-term conditions, reminders and adherence apps roughly doubled the odds of taking medicine as prescribed. NICE builds regular review into the use of weight-management medicines.
Two UK digital programmes reported that most people stayed on treatment, neither with a comparison group. Our own group’s programme reported 91.4% retention at 6 months; a second programme’s conference data put 65% to 84% still collecting medicine at 48 weeks. Statistics describe populations, not any one person’s results.
How many people stop weight-loss medication within a year?
In a US study of 125,474 adults in routine care, 64.8% of those without type 2 diabetes had stopped by one year. Among US insured adults without diabetes, one-year persistence has risen. It was 33.2% for people starting in 2021 and 60.9% for people starting in the first half of 2024.
In one US health system, the commonest recorded reasons for stopping were cost or insurance, side effects and supply shortages. Statistics describe populations, not any one person’s results.
What does regular review of weight-loss medication involve?
NICE asks clinicians to monitor the effect of weight-management medicines and to reinforce behavioural advice and adherence at regular review. Its adherence guideline adds a non-judgemental check on missed doses and a discussion of side effects, including whether to adjust the dose.
A record of your medicine-taking, side effects and weight gives that review something to work with.
How often should I weigh myself on weight-loss medication?
Once a week is the frequency this article suggests. The trials of behavioural programmes found no evidence that daily weighing beat weekly weighing. In those trials, self-weighing worked better where someone checked the weighing, though only two trials lacked that check.
Use the same scales at the same time of day. Those trials were of behavioural programmes, not of medication.
How does HeMed’s weekly check-in work?
A weekly in-app check-in, reviewed by clinicians, so your plan can be adjusted when it needs to be. You can start a live check-in with the team in the app, 24 hours a day, 7 days a week.
To report side effects, open the help menu (the question mark icon), choose Report side effects and tick the box to be contacted. The clinical team then gets in touch.
Is there UK evidence that digital weight-management programmes help?
NICE allows a set of digital weight-management programmes in the NHS while evidence is generated over 4 years. It lists adherence and completion rates among the evidence still needed.
One UK programme run jointly by SheMed and eMed reported 91.4% retention at 6 months in 8,127 self-paying women, with no comparison group. HeMed is a trading name of SheMed Limited, so that is a study of our group’s programme rather than of HeMed members. Statistics describe populations, not any one person’s results.
HeMed provides clinician-led weight management. Treatment is prescription-only and is prescribed only where a clinician decides it is suitable. This page is general health information, not medical advice. If you are worried about symptoms, speak to your GP, or use NHS 111.
Next step: see what the HeMed blood test measures.
Sources (14)
- Thakkar J et al., “Mobile telephone text messaging for medication adherence in chronic disease: a meta-analysis”, JAMA Internal Medicine 2016;176(3):340 to 349 (abstract; full text paywalled). doi.org/10.1001/jamainternmed.2015.7667. Accessed 26 October 2026.
- NICE, “Overweight and obesity management”, guideline NG246, recommendation 1.17.6 (published 14 January 2025, last updated 8 January 2026). nice.org.uk/guidance/ng246. Accessed 26 October 2026.
- Deanfield J et al., Nature Medicine, published online 16 October 2026 (the title names a medicine class and is omitted here; abstract and competing-interests statement; full text paywalled). doi.org/10.1038/s41591-026-04687-x. Accessed 26 October 2026.
- Medtech Insight (Citeline), “Digital Health Roundup”, 20 October 2026, by Marion Webb, Brian Bossetta, Shubham Singh and George Underwood: the section reporting conference data from a UK digital weight-management programme presented at the 2026 UK Congress on Obesity (trade press; the programme’s own data; not peer-reviewed; the full headline names a medicine class and is omitted here). insights.citeline.com, Medtech Insight, Digital technologies section (not linked: the page address carries a medicine class name; the evidence register holds it). Accessed 26 October 2026.
- Rodriguez PJ et al., JAMA Network Open 2025;8(1):e2457349 (the title names a medicine class and is omitted here; US electronic health records, 125,474 adults). doi.org/10.1001/jamanetworkopen.2024.57349. Accessed 26 October 2026.
- Marshall LZ et al., Journal of Managed Care & Specialty Pharmacy 2026;32(3):281 to 291 (the title names a medicine class and is omitted here; US commercial claims, 33,607 adults; four of the five authors are employed by the study sponsor, a pharmacy benefits manager). doi.org/10.18553/jmcp.2026.32.3.281. Accessed 26 October 2026.
- Gasoyan H et al., Obesity 2025;33(12):2296 to 2303 (the title names medicines and is omitted here; Cleveland Clinic, 288 adults without type 2 diabetes who stopped within a year). doi.org/10.1002/oby.70058. Accessed 26 October 2026.
- Samuels JM et al., Diabetes, Obesity and Metabolism 2025;27(11):6200 to 6209 (the title names medicines and is omitted here; Vanderbilt University Medical Center, 2,306 adults; publicly funded, no conflicts declared). doi.org/10.1111/dom.70004. Accessed 26 October 2026.
- NICE, “Medicines adherence: involving patients in decisions about prescribed medicines and supporting adherence”, clinical guideline CG76, recommendations 1.2.1 to 1.2.9 and 1.3 (published 28 January 2009; 2019 surveillance decision: no update needed). nice.org.uk/guidance/cg76. Accessed 26 October 2026.
- Nieuwlaat R et al., “Interventions for enhancing medication adherence”, Cochrane Database of Systematic Reviews 2014, CD000011 (abstract). doi.org/10.1002/14651858.CD000011.pub4; pubmed.ncbi.nlm.nih.gov/25412402. Accessed 26 October 2026.
- Armitage LC, Kassavou A, Sutton S, “Do mobile device apps designed to support medication adherence demonstrate efficacy? A systematic review of randomised controlled trials, with meta-analysis”, BMJ Open 2020;10(1):e032045. doi.org/10.1136/bmjopen-2019-032045. Accessed 26 October 2026.
- World Health Organization guideline on the treatment of obesity in adults (the guideline’s title names a medicine class and is paraphrased here), published 1 December 2025; summary in JAMA 2026;335(5):434 to 438 (same note). doi.org/10.1001/jama.2025.24288. Accessed 26 October 2026.
- Madigan CD et al., “Is self-weighing an effective tool for weight loss: a systematic literature review and meta-analysis”, International Journal of Behavioral Nutrition and Physical Activity 2015;12:104. doi.org/10.1186/s12966-015-0267-4. Accessed 26 October 2026.
- NICE, “Digital technologies for delivering multidisciplinary weight-management services: early value assessment”, HealthTech guidance HTG700 (formerly HTE14), recommendations 1.1 to 1.8 (published 26 October 2023, last updated 6 March 2024). nice.org.uk/guidance/htg700. Accessed 26 October 2026.
Numbered markers in the page text link to these entries. Statistics describe populations, not any one person’s results.