If lowering blood glucose were the cure for type 2 diabetes, the patients in the trial that lowered it the most would have lived the longest. The trial was stopped because the opposite happened.
In 2008, a research team published the results of one of the largest, most carefully designed, and most expensive randomized controlled trials in the history of metabolic medicine. It was called ACCORD — the Action to Control Cardiovascular Risk in Diabetes trial. Its purpose was to settle, once and for all, whether aggressively lowering blood glucose in type 2 diabetics with medication would reduce cardiovascular events and extend life.
The trial enrolled 10,251 patients with established type 2 diabetes and a high cardiovascular risk profile. Half were assigned to standard glucose-lowering therapy targeting an HbA1c between 7.0% and 7.9%. The other half were assigned to intensive therapy targeting an HbA1c below 6.0% — essentially driving their glucose into the normal-non-diabetic range using whatever combination of pharmaceuticals it took.
The trial was supposed to run for five and a half years. It was stopped early — after three and a half years — because the intensive treatment arm had a 22% increase in all-cause mortality and a 35% increase in cardiovascular mortality compared to the standard treatment arm.
Read that again. The patients who lowered their glucose harder, with more medication, achieved better-looking blood sugar numbers — and died at significantly higher rates.
This is one of the most important trials in modern endocrinology. It is also one of the least-discussed. The published literature has spent the last seventeen years trying to explain away the result — proposing that the intensive patients were hypoglycemic too often, that they were on too many medications, that they were too sick at baseline, that the result was somehow a statistical anomaly. The simpler explanation — that the glucose number is not the same thing as the disease, and that lowering it with medication does not fix the underlying pathology — has been largely buried.
This post is about why. We’re going to look at what the ACCORD trial actually showed, what it implies about the entire framework of pharmaceutical T2D management, what the most popular drug classes are actually accomplishing in patients’ bodies, and why the cultural moment around Ozempic and the other GLP-1 agonists deserves a more honest conversation than it’s currently getting.
The drugs are not the cure. They were never going to be. And the data has been clear about this for almost two decades.
What ACCORD actually showed
The intensive treatment arm of ACCORD achieved its target. Median HbA1c dropped from a baseline of around 8.1% to 6.4% — well within the “normal” range that mainstream guidelines push as the goal of T2D management. The standard treatment arm achieved a median HbA1c of 7.5%, in line with conservative guideline targets.
By every measure of glycemic control, the intensive arm “succeeded.” Their glucose numbers were better. Their HbA1c was better. Their fasting glucose was better. If glucose lowering were the actual treatment for the disease, this group should have done dramatically better than the controls.
Instead, after a median follow-up of 3.4 years, the intensive group had hazard ratios for both cardiovascular and all-cause mortality of 1.35 and 1.22, respectively, compared to the standard treatment group. The Data Safety Monitoring Board halted the intensive arm. The patients were transitioned to standard therapy. The trial finished prematurely.
The interpretations that followed were elaborate. The researchers proposed that the excess mortality might have been driven by severe hypoglycemia events — the medications dropping glucose too low, especially in patients already vulnerable to cardiac arrhythmia. They proposed that the rapid pace of HbA1c reduction might have stressed the cardiovascular system. They proposed that weight gain from the intensive medication regimens (insulin, sulfonylureas) might have contributed. They proposed that the burden of multiple medications and the polypharmacy interactions might have been the issue.
All of these explanations may be partially correct. None of them resolve the deeper question, which is this: if lowering blood glucose with medication does not extend life — and may actually shorten it — then blood glucose is not the disease. The number is a downstream marker. The disease is upstream.
This is the conceptual rupture ACCORD created. For thirty years before the trial, the entire pharmaceutical treatment paradigm for T2D had been built on the assumption that high glucose was the harmful thing and that lowering it was the therapeutic goal. ACCORD showed that you can lower the number, achieve the guideline target, and still die at higher rates. The implication is that the number is not, by itself, what’s killing patients with type 2 diabetes. Something else is. And the drugs being used to lower the number are not addressing that something else.
What is killing them is the underlying metabolic dysfunction — the ectopic fat in the liver and pancreas, the chronic hyperinsulinemia, the inflammation, the cardiovascular damage that’s been accumulating for two decades, the brain insulin resistance that’s been driving cognitive decline. The glucose number is the symptom that finally got the patient diagnosed. The disease is everything else, and the medications are designed almost entirely to manage the symptom.
ACCORD didn’t kill patients. The pharmaceutical management of glucose — divorced from the metabolic mechanism underneath — didn’t help them survive their underlying disease.
What metformin actually does
Of all the drugs in the modern T2D arsenal, metformin is the closest to legitimate. It’s the most-prescribed first-line agent worldwide, it’s cheap, it’s been studied for decades, and its mechanism is actually metabolically reasonable. It is worth giving an honest accounting.
Metformin works primarily by reducing hepatic glucose production. It inhibits gluconeogenesis in the liver — the same overproduction of glucose by the fatty, insulin-resistant liver that the Twin Cycle Hypothesis identified as the central pathology. In that sense, metformin is actually targeting the mechanism, not just the downstream glucose number. Patients on metformin typically see modest reductions in HbA1c (around 1-1.5%), some weight neutrality or mild weight loss, improved hepatic insulin sensitivity, and a reduction in cardiovascular event rates in most trials.
It is also one of the better-tolerated drug classes. Most metformin side effects are gastrointestinal — nausea, loose stools, gas — and tend to resolve within a few weeks of starting or with the extended-release formulation. The dosing is straightforward. It’s been in clinical use long enough that we have good long-term safety data.
What metformin does not do is remove fat from the liver or pancreas. It does not address the underlying ectopic fat accumulation that’s driving the disease. It does not restore beta cell function. It does not reverse the personal fat threshold breach. It does not get the patient off the medication. It manages the disease.
There is one additional consideration: chronic metformin use depletes vitamin B12 over time. Up to 30% of long-term metformin users develop measurable B12 deficiency, which can produce neuropathy that resembles diabetic neuropathy and may be partially mistaken for it. Any patient on metformin for more than three to five years should have B12 levels checked regularly and almost certainly needs supplementation.
The honest summary of metformin: it is a reasonable bridge medication. It buys time. It is appropriate for patients who need glucose control while they undertake the lifestyle work that will actually address the underlying disease. It is not appropriate as a permanent solution, because the underlying mechanism continues to operate while the patient takes it.
The GLP-1 phenomenon
The cultural moment around GLP-1 receptor agonists — Ozempic, Wegovy, Mounjaro, Zepbound — deserves an honest conversation that the popular discussion has so far failed to deliver.
These drugs work. They produce some of the most dramatic weight loss results ever seen in pharmaceutical history. The STEP trials demonstrated average weight loss of approximately 15% on semaglutide. The SURMOUNT trials demonstrated even greater weight loss on tirzepatide — averaging 20% or more in some patient populations. Patients who have struggled with obesity and metabolic disease for decades are seeing transformations that no prior pharmaceutical intervention has come close to producing.
The mechanism is elegant. GLP-1 (glucagon-like peptide-1) is a naturally occurring gut hormone that stimulates insulin release, suppresses glucagon, slows gastric emptying, and acts on satiety centers in the brain to reduce appetite. The GLP-1 receptor agonists are synthetic versions that bind to the same receptors and amplify the same effects, with much longer half-lives than the natural hormone. Patients on these drugs eat less, feel fuller faster, have better post-meal glucose control, and lose substantial weight as a downstream consequence.
For a meaningful subset of patients — those with advanced obesity, established cardiovascular disease, prior heart attack, or T2D that has progressed beyond the realistic remission window — these drugs are genuinely beneficial. The cardiovascular outcomes data is favorable. The weight loss reduces ectopic fat burden. The improved glycemic control reduces complication risk. For the right patient, GLP-1 agonists are one of the better tools in modern endocrinology.
But.
There are three problems with the way these drugs are currently being prescribed and discussed, and your audience deserves the honest version.
The first problem is what happens when you stop. A 2021 randomized trial published in JAMA showed that participants who stopped semaglutide regained roughly two-thirds of their lost weight within one year, while those who continued maintained or lost additional weight. A 2026 University of Cambridge meta-analysis found that a year after stopping the medication, people regain on average 60% of their lost weight. The mechanism is not mysterious — when the medication stops, appetite returns, the artificially suppressed hunger signals resume, the slowed gastric emptying normalizes, and the underlying metabolic dysfunction (which the medication was managing but not curing) reasserts itself.
This means GLP-1 agonists, as currently prescribed, are functionally permanent medications. Patients who stop them regain most of what they lost. Patients who stay on them indefinitely commit to a pharmaceutical regimen for the rest of their lives, with all the associated costs, side effects, and dependencies that implies. Few prescribers are explicit about this with patients at the start of treatment.
The second problem is lean mass loss. This is the issue that should matter most to anyone in the strength, fitness, or physique world. Multiple studies have now documented that a significant portion of the weight lost on GLP-1 agonists is not fat — it is lean body mass, including skeletal muscle. The mechanism is straightforward: the medications produce such effective appetite suppression that patients often enter severe caloric and protein deficits. Muscle is metabolically expensive tissue. When the body is in chronic deficit and protein intake is inadequate, muscle is broken down for amino acids.
Recent estimates suggest 25 to 40% of total weight lost on GLP-1 agonists is lean mass. For a patient losing 40 pounds on Ozempic, that’s potentially 10 to 16 pounds of muscle. For an older patient, who is already losing muscle through age-related sarcopenia, this represents an acceleration of frailty. For an active patient or a patient who wants to remain functionally strong, this is a dramatic trade-off.
There is a way to mitigate this. Patients on GLP-1 agonists who maintain high protein intake (1 gram per pound of target body weight, minimum) and prioritize resistance training throughout the medication period preserve substantially more lean mass. But this is not the default protocol. Most patients are simply prescribed the medication, told to “eat less,” and not coached on the resistance training and protein intake required to protect muscle. The result is rapid weight loss accompanied by significant body composition deterioration — and a body that, after the medication is stopped, has less muscle to support metabolic health than it had before treatment.
The third problem is what the medication doesn’t address. A patient on a GLP-1 agonist is, mechanically, having their appetite externally suppressed and their gastric emptying externally slowed. The medication does not remove fat from the liver. It does not remove fat from the pancreas. It does not heal beta cells. It does not reverse the underlying insulin resistance that’s been operating in the patient’s body for two decades. It can, indirectly, contribute to fat loss through its appetite effects. But the dietary and metabolic environment that produced the disease in the first place is not changed by the medication. When the medication is stopped, that environment is still operating, and the disease resumes.
GLP-1 agonists are a tool. They are an extraordinarily effective tool for certain patients. They are not a cure for type 2 diabetes, and any prescribing conversation that frames them that way is misleading the patient about what they’re actually committing to.
What the drugs aren’t doing
Step back from the specific medications and look at the system.
Metformin manages liver glucose output. GLP-1 agonists manage appetite. SGLT2 inhibitors (Jardiance, Farxiga) cause the kidneys to excrete more glucose in urine. Sulfonylureas force the pancreas to produce more insulin. Insulin itself replaces the failing endogenous supply.
Every single one of these mechanisms targets a downstream symptom of the underlying metabolic disease. None of them addresses the ectopic fat in the liver and pancreas. None of them restores beta cell function once it’s been lost. None of them — by themselves — gets the patient to a state where the disease has resolved and the medication is no longer needed.
This is the structural problem with pharmaceutical T2D management. The framework treats the disease as a chronic condition requiring lifelong symptom management. The pharmaceuticals are designed to manage the symptoms for life. The patient is told this is the standard of care. And the underlying disease continues to operate, accumulating organ damage, cardiovascular risk, and complication burden over the years and decades the patient remains on medication.
If type 2 diabetes were actually chronic and progressive — if there were no way to reverse the underlying pathology — this framework would be defensible. Patients would still suffer the consequences, but the medical system would be doing the best it could with the tools available.
The problem is that T2D is not, in fact, chronic and progressive for the majority of patients. It is reversible. The trial data is now overwhelming. And the medical system has not yet built the patient conversation around that fact.
That’s the post that closes this series.
What’s coming next
Post 4 is the resolution. The DiRECT trial. The ReTUNE trial. The Virta Health 5-year data. The mechanism of remission — what actually happens to liver fat, pancreas fat, and beta cell function when the dietary input that’s been driving the disease is removed. The 6-year window after diagnosis where remission rates are highest. The role of exercise specifically — why resistance training is the most under-prescribed therapy in modern T2D care. And the framework you can take to your doctor or coach if you want to actually resolve this disease rather than manage it for life.
This is the post the medical establishment knows the data for and hasn’t, yet, built the conversation around. The conversation has to start somewhere.
We finish this series next.
This post is part of a four-part series on type 2 diabetes — what causes it, what your doctor isn’t telling you, what the drugs can and can’t do, and what actually reverses the disease.
- Pre-Diabetes Is Not the Early Stage of Type 2 Diabetes. It’s the Late Stage of Insulin Resistance.
- The Twin Cycle: What Actually Causes Type 2 Diabetes.
- The ACCORD Trial and the GLP-1 Trap: Why the Drugs Aren’t the Answer. (this post)
- The Cure: How Type 2 Diabetes Is Reversed
One BadAss BITCH 