OBJECTIVES: This study examined the cross-sectional association between sleep duration, prediabetes, and type 2 diabetes, and its independence from the traditional lifestyle risk factors diet, physical activity, smoking behavior, and alcohol consumption.

METHODS: Cross-sectional data from 5561 people aged 40-75 years recruited into The Maastricht Study between 2010 and 2018 were used (1:1 female:male and mean age: 60.1 years [standard deviation: 8.6]). Sleep duration was operationalized as in-bed time, algorithmically derived from activPAL3 accelerometer data (median 7 nights, IQR 1). Glucose metabolism status was determined with an oral glucose tolerance test. Multinomial logistic regression was used to assess the association of sleep duration as restricted cubic spline with prediabetes and type 2 diabetes. We adjusted for sex, age, educational level, the use of sleep medication or antidepressants, and the following lifestyle risk factors: diet quality, physical activity, smoking behavior, and alcohol consumption.

RESULTS: A U-shaped association between sleep duration and type 2 diabetes was found. Compared to those with a sleep duration of 8 hours, participants with a sleep duration of 5 and 12 hours had higher odds of type 2 diabetes (OR: 2.9 [95% CI 1.9 to 4.4] and OR 3.2 [2.0 to 5.2], respectively). This association remained after further adjustment for the lifestyle risk factors (OR: 2.6 [1.7 to 4.1] and OR 1.8 [1.1 to 3.1]). No such association was observed between sleep duration and prediabetes.

CONCLUSIONS: Both short and long sleep durations are associated positively and independently of lifestyle and cardiovascular risk factors with type 2 diabetes, but not with prediabetes.

Original languageEnglish
Pages (from-to)733-741
Number of pages9
JournalSleep Health
Issue number5
Publication statusPublished - Oct 2023


Dive into the research topics of 'Both short and long sleep durations are associated with type 2 diabetes, independent from traditional lifestyle risk factors-The Maastricht Study'. Together they form a unique fingerprint.

Cite this