Eat to dream: Study shows dietary nutrients associated with certain sleep patterns

The new research is published online, ahead-of-print in the journal Appetite.

"Although many of us inherently recognize that there is a relationship between what we eat and how we sleep, there have been very few scientific studies that have explored this connection, especially in a real-world situation," said Michael A. Grandner, PhD, instructor in Psychiatry and member of the Center for Sleep and Circadian Neurobiology at Penn. " In general, we know that those who report between 7 -- 8 hours of sleep each night are most likely to experience better overall health and well being, so we simply asked the question "Are there differences in the diet of those who report shorter sleep, longer sleep, or standard sleep patterns?"

To answer this question, the research team analyzed data from the 2007-2008 National Health and Nutrition Examination Survey (NHANES) sponsored by the Centers for Disease Control and Prevention. NHANES includes demographic, socioeconomic, dietary, and health-related questions. The sample for the survey is selected to represent the U.S. population of all ages and demographics. For the current study, researchers used the survey question regarding how much sleep each participant reported getting each night to separate the sample into groups of different sleep patterns. Sleep patterns were broken out as "Very Short'' (<5 h per night), ''Short'' (5-6 h per night), ''Standard' (7-8h per night), and ''Long'' (9 h or more per night). NHANES participants also sat down with specially trained staff who went over, in great detail, a full day's dietary intake. This included everything from the occasional glass of water to complete, detailed records of every part of each meal. With this data, the Penn research team analyzed whether each group differed from the 7-8 hour "standard" group on any nutrients and total caloric intake. They also looked at these associations after controlling for overall diet, demographics, socioeconomics, physical activity, obesity, and other factors that could have explained this relationship.

The authors found that total caloric intake varied across groups. Short sleepers consumed the most calories, followed by normal sleepers, followed by very short sleepers, followed by long sleepers. Food variety was highest in normal sleepers, and lowest in very short sleepers. Differences across groups were found for many types of nutrients, including proteins, carbohydrates, vitamins and minerals.

In a statistical analysis, the research team found that there were a number of dietary differences, but these were largely driven by a few key nutrients. They found that very short sleep was associated with less intake of tap water, lycopene (found in red- and orange-colored foods), and total carbohydrates, short sleep was associated with less vitamin C, tap water, selenium (found in nuts, meat and shellfish), and more lutein/zeaxanthin (found in green, leafy vegetables), and long sleep was associated with less intake of theobromine (found in chocolate and tea), dodecanoic acid (a saturated fat) choline (found in eggs and fatty meats), total carbohydrates, and more alcohol.

"Overall, people who sleep 7 -- 8 hours each night differ in terms of their diet, compared to people who sleep less or more. We also found that short and long sleep are associated with lower food variety," said Dr. Grandner. "What we still don't know is if people altered their diets, would they be able to change their overall sleep pattern? This will be an important area to explore going forward as we know that short sleep duration is associated with weight gain and obesity, diabetes, and cardiovascular disease. Likewise, we know that people who sleep too long also experience negative health consequences. If we can pinpoint the ideal mix of nutrients and calories to promote healthy sleep, the healthcare community has the potential to make a major dent in obesity and other cardiometabolic risk factors."

Other authors for Penn include Nicholas J. Jackson and Jason R. Gerstner, PhD.

This research was supported grants from National Institutes of Health (T32HL007713, 12SDG9180007 and P30HL101859).