Health Education Research, Vol. 15, No. 1, 39-44,
February 2000
© 2000 Oxford University Press
An experimental investigation of the influence of health information on children's taste preferences
Health Behaviour Unit, Department of Epidemiology and Public Health, University College London, 2–16 Torrington Place, London WC1E 6BT, UK
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and proceduresResultsDiscussionReferences |
|---|
Promotion of healthy diets often involves provision of information about which food types are most favourable for health. This is based on the assumption that the rational consumer will, other things being equal, choose the food that they know is healthier. However, health information may not always have a positive effect, since there is evidence that some people, particularly children, believe that healthiness and tastiness are mutually exclusive characteristics. To the extent that taste governs preferences and consumption, the characterization of a food as healthy could reduce its anticipated pleasantness. The present study tested the idea that a `healthy' label would reduce liking for a novel drink. The results showed that the children rated a `healthy labelled' drink as less pleasant and said they would be less likely to ask their parents to buy it than the same drink presented with control information. These results suggest that care may need to be exercised in promoting foods to children through an emphasis on health, unless the implications of healthiness can be rendered more positive.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and proceduresResultsDiscussionReferences |
|---|
Diets that are high in fat and sodium, and low in fibre, fruits and vegetables have been found to be associated with an increased incidence of, and mortality from, cardiovascular disease and cancer (WHO Study Group, 1990
). These observations have led to the formulation of dietary guidelines in many countries and the active promotion of `healthy' diets. However, current evidence suggests that fewer than 15% of the population of the US (Block et al., 1988) or the UK (Gregory et al., 1990) meet the recommended limitations on saturated fat, and fewer than 10% eat as much fruit and vegetables as is recommended (Patterson et al., 1990). There is particular concern about the diets of children and adolescents, which are often markedly discrepant from the recommendations (Committee on Medical Aspects of Food Policy, 1989; Kimm et al., 1990). The promotion of healthier eating patterns among children is considered to be a priority in view of the mounting evidence for `tracking' both of food choices (Kelder et al., 1994) and risk factors (Berenson et al., 1989), which indicate that children with poor diets are likely to become adults with poor diets.
Many programmes directed towards reducing children's consumption of fat and lowering of their cholesterol have been carried out [e.g. (Simmons-Morton et al., 1991; Snyder and Trenkner, 1992)]. Efforts have also been directed towards increasing fruit and vegetable consumption; the `5 a Day for Better Health' programme was introduced in the US in 1991 to encourage Americans to increase their fruit and vegetable consumption. Four of the nine projects spawned by that initiative are directly targeting children in elementary schools (Havas et al., 1994).
Not surprisingly, there has also been considerable interest in identifying the factors that influence children's eating patterns. There have been a number of studies concerned with the external factors affecting children's food choices or those that impinge on children in various ways from the environment in which they live. Taras et al. (Taras et al., 1989), for example, showed that children's weekly television viewing hours correlated significantly with their requests for purchases of food advertised on television. When children were told that their mothers would be monitoring their food choices, they were found to choose fewer non-nutritious foods than when not provided with that `threat' (Klesges et al., 1991) and peer modelling of specific food choices has also been show to have a marked influence on what children eat (Birch, 1980).
Many paediatric health promotion initiatives rely, at least in part, on increasing children's food-related knowledge or their understanding of diet–disease relationships. However, few studies have explicitly investigated the mediating role of changes in attitude and beliefs, and there is no direct evidence that changes in knowledge are either necessary or sufficient for behaviour change. This issue becomes particularly important in the light of the findings from research concerned with the socialization of children's health attitudes and behaviour. Tinsley (Tinsley, 1992), for example, has shown that, overall, children are not especially concerned about health outcomes and that health is not a salient issue or a priority for them. This is especially so for those younger than 9 years (Gochman, 1987). An understanding of the way health-related beliefs interact with food preferences and choices would seem to be crucial for programmes aimed at increasing children's intake of the food that might enhance their well-being and health.
In a study of 9- to 10-year-old children's expectations and preferences, there was evidence that many of them believed `if a food tastes good, it must not be good for me; and if a food tastes bad, it is probably good for me' (Baranowski et al., 1993). This raises the possibility of the reverse effect: `if a food is said to be healthy then I probably won't like it'. If health labelling does induce negative expectations in children, then this would have important implications for the way that healthy diet should be promoted to children and how parents might best get their child to eat a healthy diet.
Are there plausible mechanisms whereby such expectations might arise in everyday life? There are at least two possible processes. One is that `less healthy' foods are probably more palatable than healthier foods, because most foods in the `less healthy' category are processed foods, manufactured specifically to appeal to the palate through the selection of well-liked proportions of fats, sugar and salt. A second possibility is that children often experience parents' exhortations to `eat it up because it's good for you', which might also promote an expectation that `healthy' equals `not tasty'.
Although the existing literature suggests that `healthiness' and `tastiness' tend to be seen as opposites by children, the fact that most of the research is correlational compromises the interpretation of the results. An experimental study in which perceived `healthiness' is experimentally controlled while the effect on taste expectation is assessed, offers the best way to detect an impact of a health label on taste expectations.
This study was, therefore, concerned with the relationship between information about health properties and liking. The hypothesis to be tested was that offering a positive health message in association with a novel drink would have an adverse effect on anticipated liking relative to a novel drink without the health label.
|
Subjects and procedures |
|---|
|
TopAbstractIntroductionSubjects and proceduresResultsDiscussionReferences |
|---|
Forty children (22 girls and 18 boys) aged 9–11 years attending a junior school in Oxfordshire in the UK took part in the study. The school drew pupils from a wide range of social backgrounds and, although social status was not formally assessed, the sample were estimated to be reasonably representative of 9–11 year olds in the UK population. Permission was obtained from the parents of all children in Years 6 and 7, informing them that the research was on children's reactions to the tastes of new drinks. No parents refused permission for their child to take part. The children were asked to taste and rate one of two drinks; one described either as `a new health drink' and the other as `a new drink'. Children were randomly assigned to each drink group. The children were told that we were interested in their reactions to some new drinks that were likely to be available in the near future, and we wanted to see what boys and girls of their age thought about them. There were no comments from the children indicating that they ascertained the purpose of the study.
Each child was individually taken through the procedure by the second author, which lasted approximately 10 min. At the start of the experimental session, the child was asked to drink 25 ml of water to ensure that they were not too thirsty and to practise the use of the rating scales. They were then asked to indicate how pleasant the water was using a five-point `smiley-face' rating scale, one end of which there was a smiling face labelled `I like it a lot' and at the other end, a scowling/unhappy face was labelled `I dislike it a lot'. Intermediate points were indicated by more or less upturned/downturned mouths. Two different versions varied the direction of these scales, in order to avoid response bias.
In the experimental procedure that followed, the child was presented with a small glass containing 50 ml of drink, that was presented and described as either `a new health drink' or simply as `a new drink'. The drinks were made either from Redoxon® effervescent, lemon-flavoured tablets or Lucozade Sports® powder in lemon flavour, which effervesced when added to water. The label on the clear plastic sandwich bag from which the tablet or powder was taken (either `new health drink' or `new drink') served to emphasize to each child the kind of drink chosen for him or her. Drink type and label were counter-balanced, so that the tablets and the powder were each used for half of each drink type. The child was asked to consume all of the drink, and then to complete rating of (1) how healthy the drink was from very bad for you to very good for you, (2) how much they liked the drink (rated as above), (3) whether they would like a parent to buy the drink (rated on five points from definitely no to definitely yes) and (4) if they would recommend it to a friend (same rating system). Finally, the children were asked hypothetically whether they thought that they would like a drink more, the same or less, if they were told it was a health drink. At the end of the study, the children were provided with a glass containing another 50 ml of the drink and they were invited to drink as much of it as they liked.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and proceduresResultsDiscussionReferences |
|---|
Two forms of the drink (tablet and powder) were used but there were no differences in any ratings of the two forms. There were also no sex differences. Data from both drink types and from boys and girls are therefore combined.
Ratings of the `healthiness' of the two drinks confirmed the impact of the intervention, the new health drink being rated as significantly more healthy than the new drink (mean health rating = 4.77 ± 0.45 for the `new health drink' versus 4.10 ± 0.79 for the `new drink'; t[38]= 3.39, P < 0.001).
Because of the non-normal distribution of the evaluations of responses to the drink, they were grouped into negative (including the midpoint) and positive responses, and the results were analysed with non-parametric statistics. Hedonic responses varied across the two drinks and are shown in Table I. The new health drink attracted fewer positive responses than the new drink (
2[1]= 5.7, P < 0.05). Likelihood of asking parents to buy the new health drink was also significantly lower than for the new drink (2[1]= 4.3, P < 0.05). The perceived likelihood that friends would like the new health drink was not significantly different than for the new drink.
|
Consumption levels after the taste rating had been completed were in the predicted direction, with slightly lower consumption of the new health drink (34.5 ± 19.1 ml) than of the new drink (38.2 ± 17.6 ml), but the difference did not reach statistical significance.
The children's answers to the question concerning the influence of a health label on liking revealed little awareness of the adverse impact identified in the intervention. Six children could not answer the question at all. Of those who did answer, three (9%) said it would make them like the drink less, 17 (50%) said it would have no effect and 14 (41%) said that it would make them like it more. As this question was asked after the intervention and ratings, it is possible that their judgements were influenced by their recent experience. No relationship was observed between actual responses to the drink and responses to the hypothetical question.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and proceduresResultsDiscussionReferences |
|---|
The results of this study supported the principal hypothesis that labelling a novel drink a `new health drink' reduced the frequency of a positive hedonic response, i.e. fewer children expected to like the drink and the reported likelihood of asking parents to buy it in future was lower. There was no significant effect on the `friends liking' variable, but if it is possible that this was a difficult judgement for children of this age. The children's comments made it clear that they interpreted `like' and `dislike' predominantly in relation to taste, but it is possible that the hedonic responses included other characteristics. These results provide some experimental support for Baranowski et al.'s (Baranowski et al., 1993
) observation that `healthy' and `tasty' might be considered by children to be incompatible characteristics. They also go further and suggest that this belief could be instrumental in modifying responses to novel foods or drinks—with children being more likely to have a negative hedonic response to products labelled `healthy'. These results are consistent with the finding in adults, that labelling a food as `low-fat' reduced its perceived pleasantness (Wardle and Solomons, 1994; Westcombe and Wardle, 1997). The effect of the health label on consumption was not significant, which may indicate that its effect extend only to expectations and not to behaviour. Two possible alternative explanations for the negative result should be considered. First, there may have been a ceiling effect; consumption was limited to one more glass (50 ml) and more than half the children drank all that they were offered (65% of those in the `new drink' group and 50% of those in the `new health drink' group). Thirst and politeness may also have contributed to over-riding the hedonic considerations. Second, most children liked the drink reasonably well, perhaps sufficiently to encourage them to drink all of it. The use of a preference test, comparing the two novel drinks with a familiar drink, might have been more useful. In future research, a preference test with a larger volume of drink and the opportunity to consume outside of the presence of the experimenter would provide a more definitive test of the effect of the labelling process on behaviour.
The hypothetical question concerning what effects children thought a health label might have was posed to them after the ratings and consumption. We had no clear expectations of the outcome of this question, but thought it possible that most children would give the `socially desirable' answer. In the event, we observed quite a strong social desirability effect, i.e. children's responses suggested that they either would not acknowledge, or did not perceive, an adverse effect of a health label. In other words they may not spontaneously verbalize their negative expectations of the tastiness of healthy foods.
The possible implications of this work for health promotion and for parents who are trying to modify their children's diets are important. First, it raises the whole issue of what health means to children and, in particular, what a `healthy food' or a `healthy drink' might indicate. This issue was not addressed in the present study, which focused particularly on likings, but it is an important issue for future research. Secondly, there is the implication that if a health label gives an implicit, negative hedonic message, then caution should be exercised in promoting foods as healthy. Parents should especially avoid invoking health in circumstances where the child is refusing (and perhaps disliking) a food, since that may serve to reinforce an association between dislike and healthiness. Further work is necessary, however, to establish the processes whereby children evaluate, and are influenced by, health information, and especially to explore the opportunities for framing health-related messages in a positive fashion.
There are limitations to this study, which should be acknowledged. The second author carried out the experiment, so it is possible that there were experimenter effects. It would be very hard to avoid this in work with children, since it is difficult to use written or taped instructions. A second limitation is that the study took place over 2 days in one school and it is possible that children discussed the study between themselves, although we asked the children what they knew and no child verbalized our hypothesis. It is also unlikely that this would have systematically biased the results, since both drink types were used in each experimental condition. A third limitation is the sample size; it would be useful to replicate the study on a larger sample and perhaps to look at moderating effect of social background. On the positive side, this is the first study of its kind to use an experimental methodology in a research area which has perhaps been over-dependent on correlational studies and it should provide a useful stimulus to further research.
|
Acknowledgments |
|---|
The help of the staff and pupils at Willsmead Middle School is gratefully acknowledged.
|
References |
|---|
|
TopAbstractIntroductionSubjects and proceduresResultsDiscussionReferences |
|---|
Baranowski, T., Domel, S., Gould, R., Baranowski, J., Leonard, S., Triber, F. and Mullis, R. (1993) Increasing fruit and vegetable consumption among 4th and 5th grade students: results from focus groups using reciprocal determinism. Journal of Nutrition Education, 25, 114–120.
Berenson, G. S., Srinivasan S. R., Hunter, S. M., Nicklas, T. A., Freedman, D. S., Shear, C. L. and Webber, L. S. (1989) Risk factors in early life as predictors of adult heart disease: the Bogalusa Heart Study. American Journal of Medical Sciences, 298, 141–151.
Birch, L. L. (1980) Effects of peer models' food choices and eating behaviors on preschoolers' food preferences. Child Development, 51, 489–496.[ISI]
Block, G., Rosenberger, W. F. and Patterson, B. H. (1988) Calories, fat and cholesterol: intake patterns in US population by race, sex and age. American Journal of Public Health, 78, 1150–1155.
Committee on Medical Aspects of Food Policy (1989) The Diets of British Schoolchildren. HMSO, London.
Gochman, D. S. (1987) Youngsters' Health Cognitions: Cross-sectional and Longitudinal Analyses. Health Behavior Systems, Louisville, KY.
Gregory, J., Foster, K., Tyler, H. and Wisemen, M. (1990) The Dietary and Nutritional Survey of Adults. HMSO, London.
Havas, S., Heimendinger, J., Reynolds, K., Baranowski, T., Nicklas, T. A., Bishop, D., Buller, D., Sorensen, G., Beresford, S. A. A., Cowan, A. and Damron, D. (1994) 5 a Day for Better Health: a new research initiative. Journal of the American Dietetic Association, 94, 32–36.[ISI][Medline]
Kelder, S. H., Perry, C. L., Klepp, K. I. and Lytle, L. L. (1994) Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. American Journal of Public Health, 84, 1121–1126.
Kimm, S. Y. S., Gergen, P. J., Malloy, M., Dresser, C. and Carrol, M. (1990) Dietary patterns of US children: implications for disease prevention. Preventive Medicine, 19, 432–442.[ISI][Medline]
Klesges, R. C., Stein, R. J., Eck, L. H., Isbell, T. R. and Klesges, L. M. (1991) Parental influence on food selection in young children and its relationships to childhood obesity. American Journal of Clinical Nutrition, 53, 859–864.
Patterson, B., Block, G., Rosenberger, W. F., Pee, D. and Kahle, L. L. (1990) Fruit and vegetables in the American diet: data from the NHANES II Survey. American Journal of Public Health, 80, 1443–1449.
Simons-Morton, B. G., Parcel, G. S., Baranowski, T., Forthofer, R. and O'Hara, N. M. (1991) Promoting physical activity and a healthful diet among children: results of a school-based intervention study. American Journal of Public Health, 8, 986–991.
Snyder, M. P., M, S. and Trenkner, L. L. (1992) Reducing fat and sodium in school lunch programs: the LUNCHPOWER! Intervention Study. Journal of the American Dietetic Association, 92, 1087–1091.[ISI][Medline]
Taras, H., Sallis, J., Patterson, T., Nader, P. and Nelson, J. (1989) Television's influence on children's diet and physical activity. Journal of Developmental Behavioral Pediatrics, 10, 176–180.[ISI][Medline]
Tinsley, B. J. (1992) Multiple influences on the acquisition and socialization of children's health attitudes and behavior: an integrative review. Child Development, 63, 1043–1069.[ISI][Medline]
Wardle, J. and Solomons, W. (1993) Naughty but nice: a laboratory study of health information and food preferences. Health Psychology, 13, 180–183.
Westcombe, A. and Wardle, J. (1997) The influence of fat content information on responses to three foods. Appetite, 28, 49–62[ISI][Medline]
WHO Study Group (1990) Diet, Nutrition and the Prevention of Chronic Diseases. WHO, Geneva.
Received on November 20, 1998; accepted on April 8, 1999
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. K. Kant Reported Consumption of Low-Nutrient-Density Foods by American Children and Adolescents: Nutritional and Health Correlates, NHANES III, 1988 to 1994 Arch Pediatr Adolesc Med, August 1, 2003; 157(8): 789 - 796. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||