Exploring Unhealthy Eating Habits

Patricia Crawford, DrPH, RD

The number of overweight children in the United States today is associated with an unprecedented rise in type II diabetes. Consequently, there is an expectation that many children born today will have shorter life spans than their parents. Today’s children are three to four times more likely to be overweight than children thirty years ago, and rates are particularly high for African American youth. The NHLBI (National Heart, Lung, and Blood Institute) Growth and Health Study, the largest study of African-American girls’ health ever conducted, can provide new insights into this phenomenon and can help us understand how changing patterns of nutrition and physical activity since the 1970s may contribute to the alarming increases in childhood obesity.

In 1987, during the period of soaring pediatric obesity rates, and in response to rapid increases in rates of obesity and heart disease in African American women, the National Heart, Lung, and Blood Institute of the National Institutes of Health funded the ten-year NHLBI Growth and Health Study (NGHS) to look at the development of obesity and Cardio Vascular Disease (CVD) risk factors in black and white girls. Participants were selected from Contra Costa County, California; Cincinnati, Ohio; and Washington, D.C. NGHS recruited more than 2,300 girls, ages nine and ten, at schools (in California and Ohio) and from a large HMO (in D.C.). The study enrolled approximately equal numbers of black and white girls, with a wide range in household income and parental education within each racial group. The study was remarkable in its very high retention rate; 89 percent of the original cohort was measured in the tenth year of follow-up.

During the course of the NGHS study, we measured the girls and collected information on their dietary and physical activity patterns each year for ten years. To our surprise, nearly a quarter of the white girls (22 percent) and nearly a third of the black girls (31 percent) were overweight (BMI > 85th percentile) at ages nine and ten when they were initially enrolled in the study. This was a doubling of the rate from the previous decade, as measured by the CDC's National Health and Nutrition Examination Survey. At ages nineteen and twenty, 41 percent of the white girls and 57 percent of the African American girls were overweight. While rates went up for both black and white girls, the absolute increase was considerably higher for black girls.

Using our wealth of annual NGHS study data on dietary and physical activity patterns, we were able to examine whether children’s food and activity environments varied by race and whether the variations might contribute to differing patterns of weight gain the two groups of girls. What kinds of changes in children’s food and activity environments did we see, and what racial differences did we observe?

First, we observed frequent consumption of food from fast-food restaurants by all children. This agrees with other analyses showing that both children and adolescents are obtaining less of their energy intake at home than in earlier eras, and more at restaurants and fast-food outlets. Portion sizes have increased, and patrons are often encouraged to purchase meals that contain more calories through “value” marketing or “supersizing.” Further, fast food and restaurant foods are typically calorically dense, nutrient-poor, and highly palatable, and frequent patronage has been associated with diets high in fat and calories. Children who ate at fast-food establishments two or more times a week were more likely to increase their relative BMI (body mass index) than those who patronized fast-food establishments once a week or less.

We found that fast-food intake rose with increasing age in girls of both races, but across all ages, from nine to nineteen, black girls consumed fast food more frequently than did white girls.

Second, we observed high consumption of sweetened beverages. In both black and white girls, consumption increased with age. Further, of all beverages consumed, increased soda consumption predicted the greatest increase in BMI. This may be due in part a result of difficulties in physiological compensation for energy consumed as a liquid compared to a solid, and in part a result of the relation between consumption of soda and other eating habits. At each annual visit, black girls consumed less milk than white girls, and the increased rate of soda consumption throughout childhood was greater among black girls than white girls.

Apart from these frequently discussed diet markers, what about overall dietary quality? Dietary patterns of eating were examined using cluster analysis to classify individual girls into discrete patterns over the ten years of the NGHS study. Eight patterns were identified: four for white girls and four for black girls. Only one could be classified as a “healthy pattern,” and it was followed by only white girls. Only 6 percent of the girls in NGHS had diets classified as “healthy.” Further, the healthy pattern was the only pattern associated with a smaller increase in adiposity over the childhood years. On average, more than two-thirds of the girls at any given age were not meeting dietary fat recommendations, with white girls only somewhat more likely to meet the guidelines. Further, nearly half of all girls had diets below recommended levels in vitamins A and C, calcium, iron, and zinc.

We found that black girls were more than twice as likely as white girls to frequently engage in a spectrum of weight-related eating practices, including eating while watching TV, eating while doing homework, skipping meals, eating in the bedroom, eating when not hungry, and eating snacks. For most of the behaviors, girls who frequently practiced a behavior had higher caloric intakes compared with those who practiced it infrequently.

Although the likelihood of a girl frequently engaging in some, but not all, of these eating practices decreased with an increase in parents’ income and education level, even when controlling for socioeconomic status (SES), black girls remained more likely to engage in these eating behaviors than white girls. In fact, NGHS was the first to report that the relationship between SES and weight differed for black and white girls. Both low household income and parental education are risk factors for pediatric overweight in white girls, but no such association was found for black girls.

The NGHS also documented a dramatic decline in leisure time physical activity in both black and white girls from ages nine to twenty, but the decline for African American girls was greater. A direct association between hours of TV viewing and increases in BMI was observed. We documented average daily television time of 3.6 hours per day for white girls aged nine to ten compared with 5.2 hours per day for black girls. The impact of television viewing may be more significant than videogame playing, because it can reduce energy expenditure and increase energy intake: TV influences the type and amount of foods consumed by viewing children. A typical child watches about 40,000 commercials on TV each year, a number that has doubled during the years that have seen a large increase in the rate of pediatric overweight. In focus groups that we have conducted, mothers report knowing what to feed children, but not knowing how to get children to eat the healthy foods. Mothers report that their messages to their children are undermined by ever-present ads for fast foods and junk foods. Furthermore, television programming marketed primarily to African Americans has significantly more food and beverage ads and advertisements for unhealthful foods than general market programming.

Overweight in both black and white girls in the NGHS study increased dramatically from nine to ten years of age to nineteen to twenty years of age. Changes in nutritional patterns are clearly implicated among the reasons for this change. However, we recognize that observed increases in pediatric obesity are not caused by people taking less personal responsibility for their well-being, but rather by profound changes in the world around us. It behooves all of us—health professionals, parents, and citizens—to advocate not only for an environment safe from imminent danger, but also an environment safe from risk of chronic disease. Our nation’s economic and social environment must support people in their actions toward achieving and maintaining healthy lifestyles. Without this, we will all pay the future costs for conditions that are the consequences of an overweight society. In the words of one expert, “If you have malaria, you drain the swamp. If you have death on the roads, you impose seatbelts.…” If you have obesity, you create healthier food and activity environments.

Patricia Crawford, DrPH, RD, is Adjunct Professor in the School of Public Health and the Department of Nutritional Sciences and Toxicology at the University of California, Berkeley. She is Codirector of the Center for Weight and Health and is a Cooperative Extension Nutrition Specialist. Dr. Crawford has led seminal studies in the longitudinal development of obesity and is currently Principle Investigator on studies exploring environmental, family, and policy approaches to childhood obesity. She is widely published, serves on several advisory boards, and is a consultant on numerous collaborative projects.