View Report Details

Changes in snacking nature and habitual energy intake relationships with obesity in 11 to 12 year old children 1990 to 2000

Project Code: N09019

30/09/2005

University of Newcastle
Adams, J ; Adamson, A; O'Keeffe, M

1.1 Introduction

The prevalence of obesity in children in the UK and elsewhere is increasing. Repeat cross sectional studies conducted by the Human Nutrition Research Centre at the University of Newcastle over the past 20 years have shown that the proportions of both boys and girls that are overweight and obese have increased.

1.2 Objectives

The objectives of this work were to use ASH11 data sets of 1980, 1990 and 2000 to describe the nature of snacking and the relative contributions of meals and snacks to the total diet. In addition, we aimed to explore the relationship between snacking and BMI and habitual energy intake, and to determine the potential role of snacking as a risk factor for overweight and obesity.

1.3 Methods

Three dietary surveys were conducted in 1980, 1990 and 2000 in the same seven schools; each included approximately 400 children. The methods used were the same across all three surveys. On two occasions during one academic year, each child recorded all foods and drinks consumed during three consecutive days, each recording period being followed by interview. Height, weight and social class were measured. Two methods of defining meal and snack were used: one applied by the nutritionist using predefined criteria which varied across the study years and a second objective method applied uniformly across the three datasets. All analyses were repeated using both definitions.

1.4 Results

Our analysis found no evidence for the existence of specific snack foods – the types of foods consumed as snacks did not differ from those consumed as meals. If this distinction between meal and snack foods existed previously, it is no longer evident. Similarly there was no evidence that either the total amount, or proportion, of energy derived from snacks or the frequency of snack intake increased between 1980 and 2000. There was some evidence that the energy density of snacks may have increased between 1980 and 2000. Regardless of definition used, the mean daily energy consumed as snacks and frequency of snacks decreased across the study years.

The quantity of food consumed as meals and snacks (total weight and portion size of specific foods) in each survey year was investigated. The total weight of snacks showed no change or a very small decrease. There was an increase in the average portion size for some foods between 1980, 1990 and 2000, including cakes and biscuits, chips, baked beans, fruit and fruit juices, confectionery, carbonated drinks and fruit drinks. Conversely, there was a decrease in portion size for fats, crisps and snacks and sauces. However, for all the foods for which a change (either increase or decrease) in portion size was detected, this change occurred regardless of whether the food was consumed as a meal or a snack.

In terms of nutrient composition, there was little evidence that snacks were consistently less or more ‘healthy’ than meals – snacks were higher in sugars and calcium than meals but lower in fat and sodium. Snacks were also an important contributor to the children’s total daily intake providing approximately 30 percent of all nutrients.

Variations in mean daily proportion of energy from snacks and the energy composition of meals and snacks according to weekday or weekend, gender, and social class in each of the survey years was investigated. Children from social class I and II families obtained a greater proportion of their energy from meals, and so a lower proportion of their energy from snacks, compared with children from social class III, IV and V families. This finding along with social class differences in the nutrient composition of meals and snacks suggests that while the distinction between meals and snacks has reduced overall the distinction may be greater in children from social class I and II than in children from social class III, IV and V.

There was evidence of an effect of social class and year of study on the chance of being overweight or obese. Children from families in social classes I, II and III were less likely to be overweight than children from families in social classes IV or V (p<0.05) and children who took part in the 2000 survey were more likely to be overweight or obese than those who took part in the 1990 and 1980 surveys. More importantly for this study, there was no relationship between the proportion of total energy intake from snacks and the risk of being overweight or obese. This finding was consistent regardless of the definition of snack applied and remained after the exclusion of ‘under-reporters’.

1.5 Discussion

Our analysis found no evidence that ‘snack foods’ are distinct from ‘meal foods’. Foods commonly considered to be eaten as snacks, such as crisps and savoury snacks, confectionery and carbonated drinks were eaten as part of meals just as commonly as they were eaten as part of snacks. This suggests that the term ‘snack foods’ in public health messages is unlikely to be particularly meaningful to children or their parents and carers. Furthermore, we found little evidence that snacks were consistently less or more ‘healthy’ than meals and confirmed that snack make a substantial contribution to the children’s total daily nutrient intake, providing approximately 30 percent of all nutrients.

Our investigations have found no evidence that snack consumption has increased between 1980 and 2000 and, indeed, the frequency of snack intake and total energy derived from snacks appear to have fallen against a background of increased prevalence of overweight and obesity. There was no evidence that the proportion of total energy derived from snacks was related to overweight or obesity. None of the changes in diet observed can explain the rapid increase in overweight and obesity among children over the past twenty years.

As our study appears to be the first work on snacking and body weight from outside the USA. Whilst these findings should be tested using other non-USA datasets and for other age groups, it seems likely that a focus on ‘snack foods’ in public health messages for prevention of obesity directed at children would be ineffective. There is a need for greater understanding of consumer perceptions of ‘snacks’ and the foods to which the term ‘snack’ is ascribed in establishing the evidence base on which messages about dietary intake generally and about food eaten as ‘snacks’ specifically could be formulated.

1.6 Conclusions

Children ate fewer snacks and consumed less energy from snacks in 2000 than in 1980 or 1990. There was no evidence of a relationship between snacking and BMI. In the UK, context these are novel findings and contrary to widely held beliefs which because of their potential to alter public health advice, should be confirmed through further studies. However our results suggest that public health messages for obesity prevention in children should focus on reduction of specific types of foods and/or the portion size of these foods rather than the context (meal or snack) in which any foods are eaten. In the absence of evidence about the role of specific foods, it would seem prudent to focus on low nutrient, high sodium, high energy dense foods as the target for reduction and to promote increased availability and consumption of high nutrient, low sodium, low energy dense foods (including fruits, vegetables and low fat starchy staples) regardless of time of day or context of consumption.

Main File

Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: