Diabetes

Could less sleep increase teen diabetes risk?

“Teenagers who stay up all night playing video games could be putting themselves at increased risk of diabetes,” the Daily Mail has reported.

This story is based on a study that assessed sleep length and insulin resistance in American teenagers. Insulin resistance is a condition where the body’s cells are not able to respond normally to the hormone insulin by absorbing glucose, leading to higher levels of glucose left in the blood. People with insulin resistance are at increased risk of developing type 2 diabetes.

The study found that teens who slept less had higher levels of insulin resistance, but, based on only the evidence, it’s impossible to make a direct cause-and-effect link between sleep and insulin resistance. Other unmeasured factors such as genetics or diet may also be influencing the link.

Furthermore, as the study measured sleep and insulin resistance over the same time period, it is not possible to say which of these occurred first, and therefore whether lack of sleep could be causing insulin resistance or whether insulin resistance could be affecting sleep patterns.

The study did not assess why some teens got less sleep, so the press reports are wrong to single out video games as the cause. It could just as easily be explained by hard-working teenagers staying up to do their homework.

The researchers expressed surprise over how little sleep the teenagers in the study actually got – the average was about six-and-a-half hours a night (the recommended amount for teenagers in the US is nine hours).

This study alone cannot tell us whether sleep duration will affect teens’ risk of diabetes. Studies that follow teens over time will be needed to determine whether this is the case.

Where did the story come from?

The study was carried out by researchers from the University of Pittsburgh and the University of California and was funded by the National Institutes of Health. The study was published in the peer-reviewed journal Sleep.

Both the Daily Mail and Express covered this story, with both suggesting that the teens were up all night playing video games or listening to music. These activities were not highlighted in the American Academy of Sleep Medicine's press release on the study and so are likely to be an unwarranted editorial addition by the newspapers.

Neither newspaper discussed any limitations to the study, such as whether insulin resistance could be the cause of disturbed sleep rather than vice versa.

What kind of research was this?

This was a cross-sectional study looking at sleep and a particular metabolic condition called insulin resistance in healthy teenagers. Insulin resistance is a condition where the cells of the body are not able to respond normally to the hormone insulin by taking up glucose, leading to higher levels of glucose in the blood. People with insulin resistance are at increased risk of developing type 2 diabetes.

The researchers said that there is growing evidence that a lack of sleep is related to metabolic problems, including insulin resistance and diabetes. They said that teenagers may be particularly at risk of having less sleep as they may stay up late with activities such as homework, part-time jobs, socialising or using media (such as TV, video games or the internet), while still having to get up early for school.

Few studies have previously looked at this age group, but the writers of the report did mention a recent US survey that found that 87% of American teenagers do not get enough sleep.

A cross-sectional study measures all exposures and outcomes at the same time. This means that it can’t tell us which event occurs first, and therefore whether one event could potentially cause the other – that is, whether less sleep causes insulin resistance or whether insulin resistance affects sleeping patterns.

Another possibility is that the association is due to confounding from other unmeasured factors. For example, a poor diet may be associated with both poor sleep patterns and with type 2 diabetes risk.

What did the research involve?

The researchers recruited 250 teenagers aged between 14 and 19 years old from health and gym classes in a single school in the US. Fifty-six per cent of the sample examined was African American, an ethnic group known to be at increased risk of type 2 diabetes.

The teens wore a monitor which recorded their movement continuously during the day and night over a period of a week. They were assumed to be asleep when their movement went below a set threshold. The researchers also assessed whether the teens had fragmented sleep, where they were restless and moved about during a sleep period. The teens provided a sleep diary, which was also used to assess total sleep time. They provided fasting blood samples, which were used to measure their glucose and insulin levels. These were used to calculate their insulin resistance using a standard method. The teens also reported how many days they were physically active for at least an hour during the week.

The researchers then used statistical tests to determine whether teens who slept for shorter or longer periods were more likely to show insulin resistance.

They took into account a number of confounders that could influence the results, such as their:

  • age
  • sex
  • race
  • body mass index (BMI)
  • waist circumference

Five of the participants were excluded from the analyses because of either missing data or they had a BMI that was much higher than the average in the group.

What were the basic results?

The researchers found that the teens in the study slept for an average of 6.4 hours a night based on the activity monitor, ranging from 4.3 to 9.2 hours. Almost half of the teens were overweight or obese based on adult BMI thresholds.

Unsurprisingly, the amount of sleep dipped during school nights, as the participants had to wake early the next morning for school.

After adjusting for confounding factors, teens who slept for shorter periods were more likely to have insulin resistance. Teens who slept for long periods or who had fragmented sleep (whereby their sleep was more frequently interrupted by activity over the course of the night) were not more likely to have insulin resistance.

How did the researchers interpret the results?

The researchers concluded that reduced sleep duration is associated with insulin resistance in teenagers. They suggest that “interventions to extend sleep duration may reduce diabetes risk in youth”.

Conclusion

This relatively small study has found a link between sleep duration and insulin resistance in teenagers. The main limitation of this study is that as it assessed sleep duration and insulin resistance in the same time period, it is not possible to say whether reduced sleep might directly cause insulin resistance, or whether conversely insulin resistance may affect sleep patterns. There are other limitations:

  • Although the study took some factors into account that might influence results (such as BMI and waist circumference), there may be other factors which were not assessed that influenced results, such as diet and genetic factors.
  • Sleep was only assessed over a one-week period, and may not be representative of longer-term sleep patterns. 
  • The newspapers have suggested that staying up playing video games may be to blame, but the study did not assess why the teens who slept less did so – they could have been doing homework or had part-time jobs in the evening.
  • Teens were recruited from gym and health classes and therefore may have been healthier than other teens.
  • The teens were all of low-to-middle socio-economic status, and all came from one school. Just over half of them were of African American origin – an ethnic group known to be at increased risk of developing type 2 diabetes. The results may therefore not be representative of the teenage population in general.

Although getting enough sleep is clearly important, this study alone cannot tell us whether doing so will lower teens’ risk of diabetes. Studies that follow teens over time, such as a cohort study, will be needed to determine whether this is the case. 


NHS Attribution