Nightly CPAP, Wild Diabetes Dip

Diabetes management tools and supplies on a white surface
Photo: MStock00 / Shutterstock

A machine millions of Americans already use to stop snoring might also help stop diabetes before it starts. New research on continuous positive airway pressure, known as CPAP, found that regular use cut diabetes diagnoses by more than half in people with sleep apnea.

Quick Take

  • A cohort study of over 1,200 sleep apnea patients found regular CPAP use tied to a 52.8% drop in new diabetes cases.
  • Diabetes incidence fell from 3.41 to 1.61 cases per 100 person-years among consistent CPAP users.
  • Separate research from the University of Chicago found eight hours of nightly CPAP use improved blood sugar in people with prediabetes.
  • Other studies, including some randomized trials, found little or no effect on long-term blood sugar control, showing the science is still developing.

What The New Sleep Apnea Study Actually Found

Researchers tracked more than 1,200 people diagnosed with obstructive sleep apnea and compared those who stuck with CPAP therapy against those who didn’t. After adjusting for weight changes and existing metabolic risk factors, the team found a 52.8% reduction in new diabetes cases among steady users. That’s a striking number for a condition that affects roughly 38 million Americans, according to the Centers for Disease Control and Prevention.

Sleep apnea causes breathing to stop and start repeatedly through the night. Each pause starves the body of oxygen and jolts stress hormones like cortisol and adrenaline. Over years, that repeated stress response can wear down how the body handles insulin, the hormone that controls blood sugar. CPAP machines keep airways open with steady air pressure, letting the body breathe normally and sleep more deeply.

Why Blood Sugar Improves When Breathing Stays Steady

A University of Chicago study backs up the connection with a different angle. Researchers found that people with prediabetes who used a CPAP machine for eight hours a night improved their blood sugar control and may have lowered their risk of progressing to full diabetes. Eight hours is a high bar; many CPAP users struggle to wear the mask that long. But the study suggests the dose of treatment matters as much as the diagnosis.

Other research adds detail. One review found CPAP therapy meaningfully lowered a key blood sugar marker called hemoglobin A1c in a significant number of diabetic patients, suggesting the benefit isn’t just theoretical. For people managing both conditions at once, that kind of improvement could mean fewer medications or better long-term control.

Not Every Study Agrees, And That Matters

Some research pushes back on the size of the benefit. A study published in Diabetes Care followed patients with sleep apnea and existing heart disease and found no evidence that CPAP improved blood sugar control or diabetes risk over standard care. Other reviews of randomized trials found CPAP made no real difference in hemoglobin A1c after several months of use.

The disagreement likely comes down to study design. Observational studies, like the 1,200-patient cohort, follow real people over time and can be swayed by lifestyle factors that are hard to fully separate out. Randomized trials control those factors better but often run shorter and use smaller groups. Neither type gives a perfect answer alone, which is why sleep researchers keep pushing for larger, longer-term trials.

For now, the practical takeaway holds up regardless of which study wins the debate: treating sleep apnea consistently brings real health benefits, and blood sugar control appears to be one of them for many patients. Doctors already recommend anyone with type 2 diabetes or high blood pressure get screened for sleep apnea, since the two conditions so often overlap. Given the size of the potential upside and the low risk of the treatment itself, patients diagnosed with sleep apnea have good reason to stick with their CPAP machine every night, not just on the nights they remember.

Sources:

pmc.ncbi.nlm.nih.gov, hcplive.com, publications.ersnet.org, jamanetwork.com