Most Routine Blood Tests Barely Moved With or Without Fasting

CATEGORY: DISEASE PREVENTION

If you’ve ever sat in a lab’s waiting room at 8 a.m., queasy and coffee-less, this one’s for you. Researchers who combed through 9.7 million blood test results found that only 10 of 121 common tests meaningfully shifted with how long people had fasted.¹ And fasting past 12 hours sharpened nothing. Not one test got more accurate for the extra hunger.

The Ten That Moved, and the 111 That Didn’t

The tests that did change are mostly the ones you’d guess. Blood sugar ran 10.8% higher in people who’d fasted under eight hours, and triglycerides, a type of fat the blood carries after meals, ran 20.1% higher. A few digestive enzymes rose too, including lipase, up 36.4%. Two readings actually ran higher in people who fasted longest: bilirubin and vitamin D. Total cholesterol wasn’t on the changed list. Neither were kidney markers, blood counts, or the great bulk of what a routine panel measures. In the sharpest comparison, people who’d fasted 8 to 12 hours matched their own over-12-hour results on every single test.

Three Years of One Hospital’s Blood Draws

Nobody was assigned a breakfast here. A university hospital in Seoul looked back at its outpatient records from 2021 through 2023, covering 101,148 adults, and sorted each result by the fasting time patients reported at the draw. A between-group comparison like that can mislead, since early-morning faster and late-morning eater may differ in other ways. So the team leaned on a stricter check: the same person’s results across repeat visits, fasted on one occasion and not on another. Only differences that held up within individuals made the list of ten. It’s still a look back at records rather than an experiment, at a single hospital, with fasting times nobody verified and patients skewing older. The authors call for other centers to confirm it.

Blood Sugar Is the Real Holdout

Glucose was the one test that kept drifting hour by hour. In people within four hours of a meal, it fell about 4.3 mg/dL for each additional hour without food. That’s why the authors conclude a true fast of eight hours or more is still needed when the goal is a fasting glucose reading or diabetes screening, even as they argue that rigid all-tests fasting rules could give way to test-specific ones.¹ The lipid finding echoes something European laboratory societies concluded back in 2016: routine cholesterol panels don’t require fasting.² One open question the study flags itself is how often a fasting difference would actually flip a result across a diagnostic line.

Key Takeaways

• Fasting longer than 12 hours made nothing more accurate; 8 to 12 hours matched it on all 121 tests examined.
• Only 10 of 121 routine blood tests changed meaningfully with fasting time, led by blood sugar and triglycerides.
• This was a records study from a single Korean hospital with self-reported fasting times, not an experiment, and glucose testing by the authors’ own account still calls for a real fast of 8 hours or more.

References

1. Shin S, Kim S, Yu S, et al. Fasting Duration and Differences Across Routine Laboratory Tests. JAMA Internal Medicine. 2026; DOI 10.1001/jamainternmed.2026.4718. link
2. Nordestgaard BG, Langsted A, Mora S, et al. Fasting Is Not Routinely Required for Determination of a Lipid Profile: Joint Consensus Statement From the European Atherosclerosis Society and European Federation of Clinical Chemistry and Laboratory Medicine. European Heart Journal. 2016;37(25):1944–1958. link