Rationale:

• Obesity results from an imbalance of energy intake vs. expenditure and a disturbance in the factors that regulate the feedback process.

Evidence:

• At least 17 genes have been associated with early onset and morbid obesity (47) and FTO is most abundant (47; 48) and in turn predicts more childhood obesity.
• Overweight is increasing in childhood (49; 50; 51; 52; 53), and high childhood weight predicts more adult overweight and obesity (54; 55; 56; 57; 58; 59). Forty percent of children with BMI at or above the 50^th percentile by age 3 were overweight at age 12 (60).
• Rapid weight gain (top 20%) in the first 5 months nearly doubles risk of overweight at age 4.5 years (15), and BMI tracks from age 1 to age 12 (61).
• Children with very high BMI may have genetic defects in melanocortin receptor or other genes (62; 63).
• Childhood obesity predicts adult metabolic syndrome (64), and metabolic syndrome in childhood predicts adult cardiovascular disease 25 years later (65).
• Age at menarche predicts risk of adult obesity (59). The odds ratio of being obese is increased 22% for each year increase in menarche (66).
• The probability of adult obesity (BMI >30 kg/m²) is 50% or more among children over age 13 whose BMI exceeds the 95^th percentile for age and gender (56).
• Four longitudinal studies suggest that depressive symptoms in childhood or adolescence are associated with a 1.9- to 3.5-fold increased risk of subsequent overweight (67).
• A high body weight during adolescence increases health risk, according to the long-term follow-up of the Harvard Growth Study of 1922 (68), a Swedish study (69), a large Danish cohort (70), and a large Norwegian cohort (71).
• Both lower family income (72; 73) and less family education (74; 75) increase risk of obesity.
• Children of overweight parents have a higher risk of becoming overweight adults (35; 54; 74; 76; 77).
• A meta-analysis of 11 studies in children found that less sleep increased the risk of obesity later in life by 58% (OR, 1.58 [CI, 1.26 to 1.98]) (78; 79; 80; 81; 82).
• A large cluster-randomized trial of breast feeding (83) did not confirm a large cross-sectional study with more than 69,000 participants that suggested that breast feeding lowered the incidence of obesity (odds ratio, 0.78 [CI, 0.71 to 0.85]) in participating children (84; 85; 86; 87), but not all studies find this (88).
• Children with a limited ability to delay gratification at age 4 are at a higher risk of overweight at age 11 (89).
• Children who are more physically active and watch less television are less likely to be overweight by age 12 (90). Children who were rewarded for being less sedentary gained less weight in a controlled trial (91; 92).
• Poor hand control, poor coordination, and clumsiness in childhood predicted obesity in adults (93).
• Shift work schedules in mothers are associated with adolescent overweight (94).
• High protein intake at ages 12 to 24 months was related to a higher BMI at age 7 years (95).
• Menarche occurs earlier in overweight girls (66).
• High BMI predicts poorer self-esteem, but those with low self-esteem were not more likely to gain weight (96).
• In epidemiological studies, a high fat diet is associated with a high prevalence of obesity (97; 98; 99).
• In a registry of adults who successfully lost weight, there were significantly more participants who ate a lower fat diet than there were controls who did not (100).
• Overeating causes obesity in controlled metabolic ward studies using identical twins (101).
• A low-glycemic-index diet improved weight loss in one pediatric study (102).
• A longitudinal, observational cohort study found that children eating foods high in energy density (more calories per gram) at age 7 had a higher risk of excess adiposity at age 9 (103).
• The more days cereal was consumed, the less the increase in BMI in adolescents (104).
• Intervention with Head Start children can slow weight gain (105).
• Higher calcium intake in adults is associated with lower body weights in cross-sectional (106) and prospective feeding studies (107; 108).
• In the cross-sectional National Health and Nutrition Examination Survey (109) and in a longitudinal Australian study (76), watching more television was associated with more instances of being overweight in children and adolescents (110), and watching less television was associated with slower weight gain in a randomized clinical trial (111).
• Soft drink consumption is associated with increased energy intake, increased body weight, and weight gain (112; 113; 114; 115). Replacing soft drinks with milk did not change body weight in 16 weeks, although changes in body fat were beneficial (116).
• In a school-based study, drinking fewer carbonated beverages was associated with slower weight gain (112).
• Juice consumption by overweight children in low-income families is associated with further weight gain (117).
• A systematic review of prevention studies in children found that 41% show a positive effect of intervention; none show a negative effect (118).
• Eating fewer family dinners together was associated with more overweight in children (110).
• Aberrant gut (fecal) microflora precede obesity in children (119).
• Mood stabilizers and antipsychotics used to treat pediatric bipolar disease are associated with weight gain (120).

Comments:

• Some types of sexual abuse in girls, but not boys, may increase risk of overweight as adults (121).