- Click here for explanation of classification scheme.

To assess epidemiological evidence systematically on the relationship between sleep duration and childhood obesity.

• Observational study, cohort study, randomized trial, cross-sectional and case-control study designs
• Studied subjects that were children between the ages of zero and 18 years
• Studied more than 100 subjects
• Included a measure of overweight/obesity using BMI.

• Systematic Review:
  • Studied subjects with ages greater than 18 years
  • Reported results on fewer than 100 subjects
  • Reported measures of overweight and obesity other than BMI.
• Meta-analysis: Did not report measures of odds ratio (or relative risk or hazard ratio) with a 95% confidence interval.

 

Recruitment

Comprehensive literature search of the PubMed database using related MeSH keywords for papers published between January 1980 and May 2007.  Searches used keywords "sleep AND obesity AND child," "sleep AND overweight AND child," "sleep AND obesity," "sleep AND overweight," "paediatric," "pediatric."  Articles were also searched using the "related article" option in PubMed.  Studies were screened for inclusion and exclusion criteria.

Design

Systematic review and meta-analysis.

Statistical Analysis

Fixed and random effects models were applied to estimate the pooled ORs and 95% CI of overweight/obesity for each sleep duration category compared with the reference group. The pooled OR was obtained averaging the natural logarithm of ORs, weighted by the inverse of their respective variances. Publication bias was assess using two formal tests:

• Beggs-adjusted rank correlation test
• Egger's regression asymmetry test.

Timing of Measurements

Papers were included if published between January 1980 and May 2007.

Dependent Variables

• Systematic Review: Presence of overweight/obesity (defined by BMI or CDC Prevention Growth Chart data indicating that a child is above the 85th percentile for overweight or above the 95th percentile for obesity)
• Meta-analysis: Associated overweight and obesity (OR/relative risk/hazard ratio).

Independent Variables

Sleep duration.

Control Variables

• Age
• Gender
• Country of origin.

• Initial N: 17 eligible observational studies (three cohort studies, 12 cross-sectional studies, two case-control studies) were included in the systematic review. 11 eligible studies were included in the meta-analysis (two cohort studies and nine cross-sectional studies).
• Attrition (final N): 17 studies (systematic review); 11 studies (meta-analysis)
• Age: Studies with subjects aged zero to 18 years
• Location: Six studies conducted in the United States, five studies conducted in Europe (France, Germany, Portugal and UK), five studies conducted in Asia (Japan, China, Hong Kong and Taiwan), one study conducted in Australia and one study conducted in Canada.

 

Key Findings

• Systematic Review:
  • Several studies presented strong evidence of the association between short sleep duration and childhood obesity
  • A prospective cohort study of 7,758 children aged three years at baseline conducted in the UK found that short sleep duration (10.5 hours) at age three was associated with obesity at age seven (OR=1.45; 95% CI: 1.10, 1.89)
  • A clear dose-response relationship was observed between sleep and obesity in Japanese children aged six to seven years. Compared with sleep duration of 10 hours or more, the adjusted OR was 1.49 (95% CI: 1.08, 2.14) for nine to 10 hours of sleep; 1.89 (95% CI: 1.34, 2.73) for eight to nine hours of sleep and 2.87 (95% CI: 1.61, 5.05) for less than eight hours of sleep after adjustment for age, gender, physical activity, screen time, parental obesity and other lifestyle factors
  • Gender differences were observed in Australia children aged seven to 15 years, with an inverse relation observed between hours of sleep and obesity in boys, but not in girls
  • In general, studies of children aged less than 10 years found in inverse relationship between sleep duration and obesity. Finding in adolescents were somewhat inconsistent.
• Meta-analysis:
  • The pooled OR and 95% CI using each studies specific criterion for short sleep duration was 1.58 (95% CI: 1.26, 1.98) for the total general pediatric population
  • The findings did not show a clear dose-response relationship, though it existed in some sub-groups
  • Compared with children having recommended sleep duration, those with much shorter sleep duration had significantly higher risk of overweight/obesity (OR=1.60; 95% CI: 1.22, 2.10). The shortest sleep duration had a much higher risk for overweight/obesity (OR=1.92; 95% CI: 1.115, 3.20).
  • The association between shortest vs. longest sleep duration and overweight/obesity was stronger in boys than in girls (OR=2.50; 95% CI: 1.88, 3.34) vs (OR=1.24; 95% CI: 1.07,1.45), P<0.001
  • For each hour unit increase in sleep duration, the pooled OR for overweight/obesity was 0.91 (95% CI: 0.84, 1.00), P=0.044
  • The analysis showed no publication bias.

• Meta-analysis provides strong evidence to help quantify the relationship between sleep duration and overweight/obesity in children and adolescents
• Children with shorter sleep duration had a 58% (pooled OR=1.58; 95% CI: 1.26, 1.98) higher risk for overweight or obesity
• Children with the shortest sleep duration had an even higher risk (92%) when compared with children having longer sleep duration
• For each hour increase in sleep, the risk for overweight/obesity was reduced on average by 9% (pooled OR=0.91; 95% CI: 0.84, 1.00)
• A significant linear dose-response relationship can be found only in young children (less than 10 years old) or for combined overweight and obese children; but not for children as a whole group.
• A significant gender difference in the association between sleep and obesity was observed, with boys having a stronger inverse association than girls (OR=2.50 vs. 1.24).

Government:	National Institutes of Diabetes and Digestive and Kidney Diseases; USDA
University/Hospital:	Johns Hopkins Bloomberg School of Public Health

• Authors noted limitations of the study, including:
  • Selection bias of papers included
  • Differences in study populations
  • Differences in assessment of co-variates
  • Exposure variables among papers
  • Differences in the classification of obesity
• Meta-analysis was limited only to observational studies, as no intervention trials were identified through the literature search
• Studies with fewer than 100 subjects were excluded, thus the studies remaining relied on sleep diaries and logs. Only one study measured sleep with wrist actigraphy, introducing the possibility of recall bias into the results.