Photo by Beelith USA on Unsplash
What Happened
48.9%. That is the share of GLP-1 drug users — people taking medications like Ozempic or Zepbound — who also carry a diagnosis of depression or anxiety, according to a study published January 15, 2026, from the All of Us Research Program, which analyzed 633,534 participants. Compare that to roughly 25% in the general population and you have a signal worth pausing on: the drugs originally approved to manage blood sugar have found their largest real-world patient base among people managing conditions far outside the original FDA label.
As of June 30, 2026, according to HealthExec coverage reported through Google News, two complementary datasets now frame this shift. The All of Us study found obstructive sleep apnea (OSA) among GLP-1 users at 42.6% — against a 13.2% general-population baseline — and chronic kidney disease (CKD) at 23.1% versus 9.0% in the broader population. A separate Trillian Health national all-payer claims analysis covering GLP-1 patients from 2018 to 2024 placed OSA as the most prevalent emerging indication at 25.5%, followed by major depressive disorder at 18.2% and CKD at 10.6%. The direction of both datasets is consistent: GLP-1 users cluster heavily around three condition categories the drugs were not originally approved to treat.
Regulatory approval has begun catching up to clinical reality. In December 2024, Zepbound (tirzepatide) became the first FDA-approved medication specifically for sleep apnea, following phase 3 SURMOUNT-OSA trial data showing significant improvements in moderate-to-severe cases. Wegovy (semaglutide) subsequently gained FDA approval for MASH — metabolic-associated steatohepatitis, a liver disease tied to metabolic dysfunction — and for cardiovascular risk reduction. A drug class that spent two decades as a niche diabetes therapy now spans a rapidly widening therapeutic map.
What the Evidence Actually Shows — and Doesn't
The All of Us and Trillian Health analyses are observational in design, not randomized controlled trials (RCTs — the gold standard where patients are randomly assigned to drug or placebo so researchers can draw cause-and-effect conclusions). Observational data can show who takes GLP-1 drugs and what conditions those patients carry. It cannot, on its own, prove that GLP-1 drugs are causing improvement in those conditions.
That distinction is clinically important. People with obesity — the primary driver of off-label prescribing, with obesity-driven first-time GLP-1 starts up 21.7% as of March 2026 while antidiabetic starts declined 9.8% — are statistically more likely to develop sleep apnea and kidney disease as downstream consequences of excess weight. The elevated condition rates may partly reflect that underlying overlap, not independent therapeutic benefit from the drug itself.
Where the evidence stands firmer is in sleep apnea specifically. SURMOUNT-OSA was a full RCT, and its results were robust enough to earn a dedicated FDA indication. For depression and CKD, the clinical case is being built, not concluded. As of June 30, 2026, Morgan Stanley analysts noted that "initial clinical data over the next two years could open entirely new treatment categories and support the next phase of market expansion" — analyst optimism grounded in pipeline potential, not completed trials. The broader prescribing pattern adds useful context: fewer than 9% of patients who started a GLP-1 between 2018 and 2023 received a diabetes or prediabetes diagnosis within 30 days of starting, confirming that off-label prescribing was already widespread long before formal approvals began arriving.
Photo by Vadim Bogulov on Unsplash
Why the Market Is Taking Notice
For anyone building or reassessing an investment portfolio with pharma sector exposure, the scale of this market evolution is difficult to overstate. As of June 30, 2026, the global GLP-1 receptor agonist market was estimated at $82.0 billion — up from $66.4 billion in 2025 — and projected to reach $185.3 billion by 2033, a compound annual growth rate (CAGR — the consistent year-over-year rate that would produce that ending value) of 12.4%. GLP-1 prescription volume surged 635.6% between 2018 and 2024, and as of March 2026, nearly 8 out of every 100 prescriptions filled were for a GLP-1 receptor agonist.
Chart: Prevalence of three conditions among GLP-1 drug users versus the general population, based on the All of Us Research Program study (633,534 participants, published January 15, 2026). Sources: All of Us Research Program; Trillian Health national claims analysis.
For the 2030 horizon, analyst estimates diverge: Goldman Sachs projects a GLP-1 market of $130–$150 billion, Morgan Stanley at $144 billion, and J.P. Morgan's high-end scenario reaches $200 billion contingent on oral GLP-1 formulations achieving mass adoption. Eli Lilly launched its oral GLP-1 weight-loss drug, Foundayo, commercially on April 6, 2026 — an early-mover advantage in the oral format race. Bernstein analysts raised their price target on Eli Lilly stock to $1,300 from $1,100 following strong GLP-1 results, with one analyst stating that "given Eli Lilly's lineup and pipeline, it is clearly winning the GLP-1 race, for now." Among established GLP-1 patients, 51.3% carry at least one diagnosis associated with an emerging or newly approved indication beyond diabetes — meaning more than half the current user base already occupies therapeutic territory that formal FDA approvals are only beginning to cover.
AI Is Accelerating the Next Wave
The GLP-1 pipeline has expanded at unusual speed, and artificial intelligence is one reason why. As of June 30, 2026, ImmunoPrecise Antibodies had launched AI-designed GLP-1 receptor agonist peptide sequences in June 2025, reporting in vitro results showing comparable or superior receptor activation relative to semaglutide. If those results hold in clinical trials, it would be a concrete demonstration that machine learning can meaningfully compress the traditional drug-discovery timeline for this class.
More broadly, AI tools are being applied across chemical libraries, preclinical datasets, and trial records to identify molecular candidates and predict which GLP-1 variants might perform best against specific conditions — including depression and CKD, where the clinical evidence is still accumulating. Industry analysts note that AI-assisted candidate screening is one reason the GLP-1 pipeline has been able to expand in breadth and pace simultaneously: more conditions, more molecular variants, and more parallel trials than traditional bench chemistry alone could sustain. For investors building positions at the AI-pharma intersection as part of their financial planning strategy, the GLP-1 class represents one of the higher-stakes and most actively developing spaces in the current landscape.
What Patients and Investors Should Consider
The evidence is uneven across conditions. Sleep apnea is the strongest case — tirzepatide has an FDA indication backed by RCT data as of December 2024. Depression and CKD signals are early-stage and should inform a conversation with your physician, not a self-directed treatment decision. One important caution: as of early 2025, the FDA had logged 455+ adverse event reports tied to compounded semaglutide and 320+ for compounded tirzepatide, many involving dosing errors serious enough to require hospitalization. If GLP-1 therapy is appropriate for your situation, using an FDA-authorized branded product materially reduces that compounding-related risk.
The GLP-1 expansion is a multi-year structural theme, not a near-term trade. The Goldman Sachs–to–J.P. Morgan 2030 range of $130–$200 billion reflects genuine uncertainty around oral-formulation uptake, competitive dynamics between Eli Lilly and Novo Nordisk, and insurer reimbursement decisions. Foundayo's commercial launch performance and payer coverage decisions in the back half of 2026 will be meaningful data points to watch — oral accessibility is the variable with the most upside and the most uncertainty.
The All of Us and Trillian Health studies are large-scale and methodologically credible, but observational. Note that efficacy varies considerably even within the GLP-1 class: the systematic review data shows retatrutide (6.5 mg) produced 13.2 kg of weight reduction over 52 weeks, while liraglutide showed only 4.03 kg — a meaningful difference within a single drug category. Before depression or CKD become established GLP-1 indications, expect full RCT results and FDA review. Track pipeline milestones rather than reacting to single-study headlines.
Frequently Asked Questions
How does GLP-1 help sleep apnea — and what does the clinical evidence show?
GLP-1 receptor agonists appear to reduce sleep apnea severity primarily through weight loss: excess tissue around the neck and upper airway is a major driver of OSA, and GLP-1 drugs produce meaningful weight reduction that can relieve airway obstruction. The evidence for sleep apnea is the strongest of the emerging indications — Zepbound (tirzepatide) received a specific FDA approval for sleep apnea in December 2024 based on the phase 3 SURMOUNT-OSA randomized controlled trial, which demonstrated significant improvements in moderate-to-severe OSA. That is RCT-level evidence, not just observational correlation.
What conditions can GLP-1 drugs treat beyond type 2 diabetes as of mid-2026?
As of June 30, 2026, FDA-approved indications for GLP-1 drugs include type 2 diabetes, obesity, cardiovascular risk reduction, sleep apnea (tirzepatide/Zepbound), and MASH (for semaglutide/Wegovy). Depression and chronic kidney disease are among the most studied emerging targets, but do not yet carry FDA approval for GLP-1 therapy — those indications remain in clinical development. The All of Us Research Program study published January 2026 found 48.9% of GLP-1 users had depression or anxiety and 23.1% had CKD, reflecting widespread off-label use alongside approved indications, not formal therapeutic approval for those conditions.
What are the main side effects of GLP-1 medications like Ozempic and Zepbound?
The most commonly reported side effects are gastrointestinal: nausea, vomiting, diarrhea, and constipation, particularly during dose escalation. More serious but less common risks include pancreatitis, gallbladder disease, and potential thyroid C-cell concerns (observed in rodent studies, not confirmed in humans at clinical doses). The FDA's adverse event data through early 2025 — 455+ reports for compounded semaglutide, 320+ for compounded tirzepatide, many involving dosing errors requiring hospitalization — highlights that compounded versions carry additional risk beyond the standard branded product profile.
Are GLP-1 drugs like Ozempic safe for long-term use, and how long until results appear?
Long-term safety data is accumulating but not yet decades-long, since the obesity indication is relatively recent. Multi-year cardiovascular outcome trials — including LEADER for liraglutide and SUSTAIN-6 for semaglutide — generally demonstrated cardiovascular benefit over extended follow-up, which is an encouraging long-term signal. On timeline, most patients see meaningful weight and metabolic changes within 12–16 weeks, with maximum effect typically requiring 52 or more weeks of treatment. The SURMOUNT-OSA trial, for instance, used a 52-week endpoint to measure sleep apnea outcomes. Duration and appropriateness of treatment is ultimately a decision to make with your physician based on your individual health profile.
Bottom line: In my analysis, the most important signal in the All of Us and Trillian Health data is structural: a drug class approved for one disease has found its primary real-world user base among people with a cluster of metabolic, respiratory, and mood conditions. Whether that reflects shared underlying biology — obesity as a common root driving multiple downstream conditions — or genuine multi-condition therapeutic efficacy is a question ongoing RCTs will continue to resolve. What the market picture shows clearly is expansion on multiple fronts simultaneously: more conditions under study, oral formulations broadening the addressable population, AI tools compressing discovery timelines, and analyst consensus projecting a market north of $130 billion by 2030. The risk that deserves watching is not the science — it is long-term safety data and insurer reimbursement policy. Those two variables will define the ceiling on this market more than the pipeline alone.
Disclaimer: This article is for informational and educational purposes only and does not constitute financial or medical advice. Consult a qualified financial advisor before making investment decisions and a licensed healthcare professional before making any medical decisions. Research based on publicly available sources current as of June 30, 2026.