Running a GLP-1 Programme? 12 Ways to Prove It Works to Payers
Sooner or later, whoever pays for your GLP-1 programme will ask whether it is working and whether it is worth the cost. The answer depends less on the medicine and more on the data you collect from the first consultation onwards.
Suhina Singh
Founder & CEO, Jonda Health
October 24, 2026
·
11 min read
GLP-1 (glucagon-like peptide-1) receptor agonists such as semaglutide, and dual agonists such as tirzepatide, have moved from specialist clinics into telehealth services, employer benefits, insurer programmes and pharmacy-led weight management in a very short time. Demand has grown faster than the evidence most programmes hold about their own patients.
The evidence usually exists, but it is scattered. Baseline blood tests arrive as PDFs from outside laboratories, weights come from home scales and phone photos, side effects sit in chat transcripts, and refills live in a pharmacy system. The twelve points below set out what to capture, and when, so that your programme can answer the payer's question with confidence.
Who asks for proof, and what they want to see
The questions differ a little by programme type, but the underlying data problem is the same.
Telehealth weight management
Asked by
Patients, regulators and partner insurers, who want safe prescribing and sustained results.
Employer and insurer programmes
Asked by
Benefits leaders and actuaries, who want outcomes per member, adherence and cost offsets.
Endocrinology and hospital clinics
Asked by
Hospital management and public payers, who want outcomes against national guidance.
Pharmacy-led programmes
Asked by
Chains, regulators and payers, who want safe supply, monitoring and timely referral.
The twelve points at a glance
01Capture a complete baseline
02Apply the right BMI threshold
03Define a response and measure it on time
04Track the outcomes payers value
05Keep an exact dose and product history
06Measure persistence honestly
07Record why people stop
08Follow people after they stop
09Look beyond the scale
10Make home readings count
11Capture safety events as data
12Package it for each payer, and protect it
Each number matches a point below, so you can jump to the stage your programme finds hardest.
1. Capture a complete baseline before the first dose
Every outcome you report is a change from baseline, so a weak baseline weakens everything that follows. At a minimum, record weight, height, body mass index (BMI), waist circumference, blood pressure, glycated haemoglobin (HbA1c), lipids, kidney and liver function, current medications and the person's recent weight history.
Payers check this too. The United States Medicare GLP-1 Bridge pilot, which opened on 1 July 2026, judges eligibility on BMI at the start of GLP-1 therapy, so a programme must be able to produce that historical figure even if the person has since lost weight.1
Pitfall: Much of the baseline arrives as laboratory PDFs, photographed reports and referral letters in different formats and languages. If it stays as attachments, nobody can calculate a change from it later.
2. Apply the right BMI threshold for the people you serve
BMI thresholds are not the same everywhere, and payers notice when a programme applies the wrong one. Singapore treats a BMI of 27.5 or above as a high-risk category for Asian populations.2 In England, the National Institute for Health and Care Excellence (NICE) usually lowers thresholds by 2.5 for people from South Asian, Chinese, other Asian, Middle Eastern, Black African or African-Caribbean backgrounds.3 In Australia, the Pharmaceutical Benefits Advisory Committee (PBAC) proposed 32.5 rather than 35 for people of Asian, Aboriginal or Torres Strait Islander background.4
Simplified from the sources cited in this point and in point 1. NICE's standard route also covers BMI 30 to 34.9 for people who meet specialist referral criteria.
Tip: Record which threshold was applied to each person, and under which guideline or payer rule. That single field makes an audit or payer review far simpler.
3. Define what a response looks like, and measure it on time
A programme that cannot say who is responding cannot show value. NICE advises considering stopping semaglutide if less than 5% of initial weight has been lost after 6 months of treatment.5 Whatever rule your market or payer uses, write it down and measure percentage weight change from baseline at the same points for everyone.
An example schedule. Use the response rule that applies to your market and payer.
4. Track the outcomes payers value, not only weight
Weight is the headline, but much of the clinical value sits in what changes alongside it. Track HbA1c, blood pressure, lipids and the status of weight-related conditions over time. In one large real-world cohort, people with prediabetes who stayed on therapy were twice as likely to reach normal glucose levels at one year as those who stopped early.6
Adults with obesity without type 2 diabetes starting semaglutide or tirzepatide, Cleveland Clinic cohort.6
5. Keep an exact dose and product history
Outcomes cannot be attributed to treatment if you do not know exactly what the person took and when. Record each dose step, any pause, switches between medicines, and whether a product was branded or generic. This matters more each year: generic semaglutide launched in India in March 2026 after the patent expired there, with dozens of brands entering a largely self-pay market.7
Illustrative. Without the pause and the switch on record, a plateau in weight loss looks like a treatment failure.
6. Measure persistence honestly, with a stated definition
Persistence is often the first figure a payer asks for, and real-world rates are far lower than trial rates. In a United States study of more than 125,000 adults, most people without type 2 diabetes had stopped within a year.8 A meta-analysis of real-world studies put the overall figure at nearly half.9
US cohort of 125,474 adults8 and a meta-analysis of 12 real-world studies.9
Pitfall: Studies define stopping differently, often as a gap of 60 or 90 days in supply. A persistence rate with no definition next to it cannot be compared with anything.
7. Record why people stop, in a structured way
Studies in the same meta-analysis cited gastrointestinal side effects, cost, medication shortages and perceived lack of benefit as reasons people stopped.9 If those reasons sit only in chat messages and call notes, you cannot count them, and you cannot show a payer what your programme did about each one. Capture a reason for every stop, using a short fixed list.
GIGastrointestinal side effectsCOSTCost or coverageSUPSupply or shortageBENPerceived lack of benefitGOALReached goalCLINClinical decisionOTHOther, with a note
Tip: Keep the list short enough that staff use it every time. Seven well-used codes beat thirty that nobody picks.
8. Follow people after they stop
What happens after treatment ends is part of the evidence. Data from the STEP 4 and STEP 1 extension trials suggest people regain half to two-thirds of lost weight around a year after stopping semaglutide.10 A programme that keeps measuring weight and key results after discontinuation, and can show how many people restart, gives payers a far more credible picture.
Illustrative shape based on published summaries of STEP 1, its extension and STEP 4.10 Your own programme data will differ.
9. Look beyond the scale
Payers and clinicians increasingly ask what kind of weight is lost. In a body composition substudy of the STEP 1 trial, total lean body mass fell by 9.7%, yet lean mass made up a larger share of body weight at the end than at the start, because fat fell faster.11
Wilding et al., STEP 1 body composition analysis.11
Tip: Where you can, capture body composition scans or bioimpedance readings at baseline and follow-up, and store the device type with each result. Readings from different devices are not directly comparable.
10. Make home readings count as evidence
Much of a GLP-1 programme's data is now generated at home: smart scales, blood pressure cuffs, glucose meters, and photos of device screens sent through an app. These readings only count as evidence if each one carries the context that makes it trustworthy.
A weight recorded in pounds in one place and kilograms in another, with no date, will not survive a payer review.
11. Capture side effects and safety events as data
Safe prescribing is part of proving value, especially for telehealth and pharmacy-led programmes. Record side effects, dose reductions, emergency visits and hospital admissions as structured entries rather than free text, each with a date and a link to the dose the person was on at the time.
Pitfall: Hospital discharge summaries and specialist letters often hold the most important safety information, and they rarely arrive in a consistent format. If nobody structures them, the event never reaches your safety report.
12. Package the evidence for each payer, and protect it
An employer wants outcomes per member and cost offsets, an insurer wants eligibility and persistence, and a public payer wants outcomes against national guidance. Build your reporting so the same underlying data can be filtered for each audience. Share data in pseudonymised form wherever possible, record consent clearly, and check whether local rules require data to be processed in country.
Coverage rules differ widely by market, and every one of them rewards programmes that can show baseline eligibility, response and persistence.
Market
Where coverage stands
What to have ready
United States
Medicare GLP-1 Bridge pilot opened 1 July 2026 for people meeting BMI and condition criteria12
BMI at therapy start, qualifying conditions, prior authorisation records
United Kingdom
NICE recommends semaglutide and tirzepatide; tirzepatide is being phased into primary care with required wraparound support13
Comorbidities, 6-month response, evidence of wraparound support
Australia
PBAC recommended a restricted Wegovy listing for established cardiovascular disease with obesity; a revised proposal is due at the November 2026 PBAC meeting14
Documented cardiovascular history, BMI with the applicable threshold
Singapore and Southeast Asia
Largely self-funded or employer-funded, with Asian BMI cut-offs2
Outcomes per member for employers, records in many formats and languages
India
Generic semaglutide launched March 2026; not covered by government health schemes7
Product and brand history, self-pay persistence data
South Africa and wider Africa
Wegovy launched in South Africa in 2025; obesity is not a prescribed minimum benefit, so medical scheme cover is limited15
Comorbidity outcomes that make the case to medical schemes
Coverage snapshot as of October 2026. These rules are changing quickly, so check them before you rely on them.
How Jonda Health can help
Most of the twelve points come down to one problem: the evidence exists, but it is scattered across PDFs, photos, device screens and systems that do not speak to each other. We help programmes turn it into structured, standardised information that clinicians, payers and systems can actually use.
JondaX
Make the evidence usable
One API for laboratory reports, clinical documents, discharge summaries and photographed device readings, in 10+ languages.
Cleans the data, resolves the different ways the same test is captured, and maps it to medical coding standards or your own code sets.
Clinically trained reviewers check flagged results, with 99% field-level accuracy after expert review.
ISO 27001 certified, designed to comply with HIPAA, GDPR and PDPA, and deployable in country.
Jonda Health Services
Design the data side of your programme
What to capture at baseline and follow-up, and when.
Outcomes and persistence reporting shaped for each payer.
Consent, pseudonymisation and data flows that meet local rules.
Integration planning into your platform, electronic health record or clinic system.
See it on your own records. Send us a sample of real programme documents (up to 20 pages) and we will run them through JondaX free of charge, so you can see the structured output next to the originals. If you would rather talk it through first, book a free 30-minute data readiness conversation.
