Short and long sleep durations associated with higher diabetes incidence, study finds

By Vijay Kumar MalesuReviewed by Lily Ramsey, LLMMay 1 2024

In a recent study published in BMC Medicine, a group of researchers identified plasma proteomic patterns associated with sleep duration and evaluated their impact on the risk of developing diabetes mellitus (DM) and coronary heart disease (CHD).

Study: Very short sleep duration reveals a proteomic fingerprint that is selectively associated with incident diabetes mellitus but not with incident coronary heart disease: a cohort study. Image Credit: Lysenko Andrii/Shutterstock.com

Background 

Research has established links between varying sleep durations and diseases like DM and CHD across diverse studies.

Despite known associations, the biological mechanisms remain unclear, particularly the role of inflammatory markers like C-reactive protein (CRP) and interleukin 6 (IL-6), which are linked to cardiometabolic risks but provide limited insights due to non-specificity and potential reverse causality.

Advances in proteomic assays now allow exploration of the biological pathways that connect sleep duration to cardiometabolic outcomes.

Further research is needed to precisely identify and understand the biological pathways linking sleep duration to specific health outcomes, thereby improving prevention and treatment strategies.

About the study 

The Malmö Diet and Cancer (MDC) study, a population-based, prospective cohort study, was conducted in Malmö, Sweden, and enrolled men and women aged 45-73 years between 1991 and 1996.

Participants underwent baseline examinations where they provided anthropometric data blood samples, and completed detailed lifestyle questionnaires covering a range of topics from heredity to medication use.

A subset of these participants was included in the MDC Cardiovascular Cohort (MDC-CC) to study carotid artery disease specifically. This subgroup underwent additional examinations, including carotid ultrasonography, and provided plasma for novel proteomic marker analysis.

For the present study, participants with pre-existing conditions like DM or CHD were excluded to minimize biases in the analysis of incident cases.

The study focused on identifying proteomic markers linked to different sleep durations, measured through detailed questions about weekday and weekend sleep patterns. These markers were analyzed using a high-specificity immunoassay to determine their association with new-onset DM and CHD.

The final sample consisted of 3,336 participants, followed until the end of 2018. Statistical analyses, including logistic regression and Cox proportional hazards regression, explored the associations between sleep duration, proteomic markers, and health outcomes. 

Study results 

In the present detailed study, researchers analyzed baseline characteristics and proteomic markers across different groups of habitual sleep duration among participants initially free from these conditions.

The group with the median sleep duration served as the reference in survival analyses, characterized by the youngest average age, lowest insomnia symptoms, and smallest waist circumference.

These participants also had the highest education levels beyond elementary school and varied significantly in terms of alcohol consumption and physical activity.

The study identified 16 unique proteomic markers significantly associated with varying sleep durations, reflecting predominantly inflammation and apoptosis processes.

These markers were differentially associated with shorter and longer sleep durations than the median group.

Advanced regression techniques highlighted that combinations of these proteomic markers effectively predicted habitual sleep duration, particularly at the extremes of the shortest and longest sleep durations.

D distinct patterns emerged over the follow-up period averaging 21.8 years for DM and 22.4 years for CHD. When compared with the median, all sleep duration groups showed a positive and significant relationship with incident DM.

However, only the shortest sleep duration was associated with a higher risk of CHD. Incorporating proteomic scores, particularly for the shortest sleep duration, modified these relationships, significantly attenuating the association with DM.

The study's findings suggest that specific proteomic profiles can mediate the link between sleep duration and health outcomes.

For instance, the proteomic risk score for the shortest sleep duration mediated a substantial proportion of the association with DM during later follow-up years. This suggests a potential biological pathway through which sleep duration may influence the development of DM and possibly CHD.

However, the associations with CHD were less pronounced. While initial models showed a significant association between shorter sleep durations and CHD, adding proteomic scores mitigated these links.

While short sleep duration remained a predictor of CHD, its direct relationship could be somewhat obscured or confounded by underlying proteomic factors not fully captured in the models.

Conclusions 

In the study, sleep duration emerged as a crucial factor linked to the incidence of DM and CHD. Sixteen unique proteomic markers effectively predicted variations in sleep duration and suggested distinct biological pathways influencing these cardiometabolic diseases.

A specific proteomic score significantly mediated the association between shorter sleep durations and increased DM incidence, particularly in later years.