Why more sleep helped but less sleep did not hurt
The clearest experimental comparison comes from a randomized crossover study in 36 adolescents with obesity and other risk factors for type 2 diabetes [2]. Participants first slept their usual amount, then spent a week with time in bed increased by 1 hour 30 minutes per night and another week with time in bed decreased by the same amount. During the usual week they slept about 7 hours 31 minutes per night; the extension week added about 1 hour 2 minutes, and the restriction week cut about 1 hour 19 minutes [2].
After the extension week, insulin sensitivity—measured with a two-hour oral glucose tolerance test—was 20% higher than after either the usual or the restricted week [2]. The restriction week, however, did not produce a significant drop compared with usual sleep [2]. That asymmetry is the key point: in this group, gaining sleep improved insulin sensitivity, but losing about an hour did not measurably worsen it.
In adults with prediabetes or type 2 diabetes, sleep duration alone did not track with insulin sensitivity
A cross-sectional analysis of 221 adults with prediabetes or recently diagnosed, untreated type 2 diabetes found no association between actigraphy-measured sleep duration and insulin sensitivity, fasting glucose, or two-hour glucose during an oral glucose tolerance test [1]. Sleep duration was also not linked to β-cell responses. The average sleep duration in this group was 6.6 hours per night, and most participants had obstructive sleep apnea, a condition that can fragment sleep [1].
This does not mean sleep is irrelevant. It means that in this adult sample, the natural variation in sleep duration—measured over one week—did not line up with insulin sensitivity or post-meal glucose. The study was not an experiment that extended or restricted sleep, so it cannot show whether changing sleep would change those outcomes [1].
Short-term restriction can reduce insulin sensitivity, but the effect depends on the group
A broader review of experimental, physiological, and epidemiological evidence reports that short-term sleep restriction can rapidly reduce whole-body insulin sensitivity and impair β-cell responsiveness, even without weight gain [3]. That is consistent with the idea that sleep loss can be metabolically harmful. But the review also describes sleep-focused interventions, including sleep extension, as producing modest but clinically meaningful improvements in insulin sensitivity and glycemic outcomes [3].
The apparent conflict with the adolescent trial—where restriction did not significantly lower insulin sensitivity—may reflect differences in how long or how severely sleep was restricted, the age of the participants, or the fact that the adolescents were sleeping a relatively healthy 7.5 hours at baseline [2]. The review's summary is broad and does not establish that restriction and extension always move insulin sensitivity in opposite directions [3].
What about post-meal glucose specifically?
The adolescent trial measured insulin sensitivity from an oral glucose tolerance test but did not report a separate post-meal glucose result in the abstract [2]. The adult cross-sectional study found no association between sleep duration and two-hour glucose on an oral glucose tolerance test [1]. The review discusses glycemic outcomes more generally but does not provide a specific comparison of extension versus restriction on post-meal glucose in people with prediabetes or type 2 diabetes [3].
So the direct answer for post-meal glucose is less certain than for insulin sensitivity. The evidence supports the possibility that extending sleep improves insulin sensitivity in at-risk adolescents, but it does not yet show a clean opposite effect of restriction, and it does not establish the same pattern for post-meal glucose in adults with prediabetes or type 2 diabetes [1][2][3].
Sources used in this answer
Obstructive Sleep Apnea, Glucose Tolerance, and β-Cell Function in Adults With Prediabetes or Untreated Type 2 Diabetes in the Restoring Insulin Secretion (RISE) Study
In adults with prediabetes or recently diagnosed type 2 diabetes, sleep duration measured by actigraphy was not associated with insulin sensitivity, fasting glucose, two-hour glucose, or β-cell responses, although higher apnea-hypopnea index was associated with higher HbA1c.
Sleep and insulin sensitivity in adolescents at risk of type 2 diabetes: the Sleep Manipulation in Adolescents at Risk of Type 2 Diabetes randomized crossover study
In adolescents at risk for type 2 diabetes, one week of sleep extension increased insulin sensitivity by 20% compared with habitual and restricted sleep, while one week of sleep restriction did not significantly change insulin sensitivity compared with habitual sleep.
The Role of Sleep in Insulin Sensitivity and Type 2 Diabetes Risk
A review reports that short-term sleep restriction can rapidly reduce whole-body insulin sensitivity and impair β-cell responsiveness, and that sleep-focused interventions including sleep extension show modest but clinically meaningful improvements in insulin sensitivity and glycemic outcomes.
Time-restricted feeding improves blood glucose and insulin sensitivity in overweight patients with type 2 diabetes: a randomised controlled trial
In overweight adults with type 2 diabetes, 12 weeks of time-restricted feeding improved HbA1c, insulin resistance, and β-cell function compared with a control diet, but the study did not test sleep extension or restriction.
Sleep and insulin sensitivity in adolescents at risk for type 2 diabetes
In adolescent girls at risk for type 2 diabetes, most actigraphy-measured sleep characteristics were not significantly associated with insulin sensitivity after adjustment, with only trend-level associations for weekday sleep efficiency and weekend wake after sleep onset.