If you need a lie-in at weekends to make up for lack of sleep in the week, you may be at risk of obesity and type 2 diabetes, a study suggests.
The sleeping habits of 522 people found those losing sleep on weekdays were more likely to develop the conditions.
The findings, shown at the Endocrine Society's annual meeting, suggested increasing sleep could help patients.
Experts said the findings were interesting and called for the idea to be tested in large trials.
The action of throwing the body clock out of sync is thought to disrupt the natural rhythm of hormones in the body, leading to a host of health problems.
But the pressures of work and social lives mean many people cut their sleep during the week and catch up at the weekend. Researchers are investigating whether there is a health impact.
The study, by a team at the University of Bristol in the UK and Weill Cornell Medical College in Qatar, assessed "sleep debt" - a measure of the difference in the nightly hours asleep on weekdays and at the weekend.
"We found that as little as 30 minutes a day sleep debt can have significant effects on obesity and insulin resistance," said Prof Shahrad Taheri from Weill Cornell.
He added: "Sleep loss is widespread in modern society, but only in the last decade have we realised its metabolic consequences.
"Our findings suggest that avoiding sleep debt could have positive benefits for waistlines and metabolism and that incorporating sleep into lifestyle interventions for weight loss and diabetes might improve their success."
The study was funded by the UK's Department of Health, where 10% of healthcare budgets are already spent on treating diabetes.
The disease can lead to blindness, increase the risk of heart attacks and strokes, as well as damaging nerves and blood vessels - dramatically increasing the risk of a foot needing to be amputated.
What the researchers do not know is the impact of improving people's sleep so they get more on a weeknight and do not need a weekend lie-in.
Dr Denise Robertson, a senior lecturer from the University of Surrey, commented: "This work is interesting and consistent with prospective data found in healthy individuals without type 2 diabetes.
"However, before this association between sleep length, obesity and metabolic status can be used in terms of public health we need the next tier of evidence.
"To date there have been no randomised controlled trials where sleep debt is addressed and a metabolic benefit is observed. However, the potential for such interventions to impact on health is great."