On the question of increased diabetes risk with statin use, I think we need to put the various study results into perspective. I asked ChatGPT the following:
“How is incidence of NOMD calculated in this context? Is it a question of blood glucose and insulin levels crossing a threshold, or is there permanent damage to the pancreas necessitating exogenous insulin?”
The bot’s answer below. My take on this is that what we’re really talking about is a worsening of insulin resistance, which in those with existing IR can push them over the clinical definition of diabetes, and which should still be reversible with medications and lifestyle interventions.
ChatGPT:
Good question — and this is where “new-onset diabetes (NODM)” in statin studies often gets misunderstood.
## 1. What “NODM” actually means in these studies
In virtually all statin and cardiovascular outcome studies, NODM does not mean pancreatic failure or insulin dependence.
It is usually defined using standard diagnostic thresholds, such as:
-
Fasting plasma glucose ≥ 7.0 mmol/L (126 mg/dL), or
-
HbA1c ≥ 6.5%, or
-
2-hour OGTT glucose ≥ 11.1 mmol/L, or
- A clinical diagnosis recorded in medical records, often including:
- starting diabetes medication (metformin, etc.)
So it is fundamentally a biochemical/diagnostic threshold event, not a structural disease marker.
## 2. What it does not mean
NODM in these datasets does NOT imply:
- permanent pancreatic “damage”
- loss of insulin production requiring insulin therapy
- beta-cell destruction (like type 1 diabetes)
- irreversible metabolic failure at onset
Most people classified as NODM:
- are initially treated with lifestyle ± metformin
- often have mild or early type 2 diabetes
- retain substantial endogenous insulin production
## 3. What is actually happening biologically
In the context of statins (including pitavastatin, atorvastatin, etc.), the mechanism is believed to be:
- mild worsening of insulin sensitivity (muscle and liver)
- slightly increased hepatic glucose output
- small changes in insulin secretion dynamics in susceptible individuals
So statin-associated NODM is usually:
a shift across a diagnostic threshold in predisposed individuals, not a sudden biological switch to insulin dependence
## 4. Why this matters for interpreting studies like pitavastatin papers
When you see:
- “new-onset diabetes increased by X%”
that typically means:
more people crossed a lab or diagnostic cutoff during follow-up
It does not mean:
- more people developed clinically severe diabetes
- more people required insulin
- more people had pancreatic dysfunction
That distinction is why:
- absolute risk increases are small
- and many cases are reversible or mild early T2D
## 5. Why statins look like they “cause diabetes”
A key nuance:
Many people classified as NODM in statin trials were likely:
- already in pre-diabetic range at baseline
- progressing slowly anyway
Statins may:
- slightly accelerate the timing of crossing the threshold
- not necessarily create diabetes de novo
This is supported by analyses showing the strongest risk in people with:
- higher baseline fasting glucose
- higher BMI
- metabolic syndrome features
## 6. Does any of this involve insulin dependence?
Very rarely in these datasets.
In real-world terms:
- statin-associated diabetes is overwhelmingly type 2 diabetes phenotype
- insulin use at diagnosis is uncommon
- pancreatic “failure” is not the mechanism
If insulin is eventually needed, it is because:
- the person has progressive type 2 diabetes over time, not because of acute beta-cell loss from the statin
## 7. Bringing it back to pitavastatin
So when studies report:
- lower NODM risk with pitavastatin
what they really mean is:
fewer people crossed the biochemical diagnostic threshold for type 2 diabetes during follow-up
not:
pitavastatin protects pancreatic beta cells from destruction
Bottom line
- NODM = diagnostic threshold crossing (glucose/HbA1c/recorded diagnosis)
- NOT pancreatic failure
- NOT insulin dependence
- Usually represents early or mild type 2 diabetes in predisposed individuals
- Statin effect = small metabolic shift + threshold acceleration in some people
