![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Department of Nutritional Sciences (U-17), University of Connecticut, Storrs, CT 06269
3To whom correspondence should be addressed. E-mail: rperez{at}canr.uconn.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • CSFII • DHKS • dietary quality • food labels • Healthy Eating Index • nutrition knowledge
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). It has also been estimated that one third to one half of the premature incidence of these conditions could be prevented through changes in lifestyle behaviors including diet and physical activity (2
). Both poor nutrition and resulting conditions are disproportionately suffered by low income groups including ethnic minorities (3
). Low income individuals are likely to have poorer dietary practices than their wealthier counterparts (4
–7
). A greater acquisition of nutrition awareness and knowledge by higher income groups has been hypothesized to be responsible, at least in part, for the inverse association between social class, life style behaviors, and the risk of ischemic heart disease mortality seen today in most industrialized nations (8
).
Food labels (FL)5
are found on most food products sold in the United States (9
) It has been hypothesized that use of the current FL could result in a decrease in chronic, diet-related diseases, such as coronary heart disease and some cancers (10
). This is consistent with research that has shown that at least some use of the FL is associated with diets higher in overall dietary quality, lower in fat and/or higher in fruits and vegetables (11
–15
) and with higher serum carotenoids (15
). Of particular importance is that low income groups, which are at greater risk for poor health outcomes, are less likely to use the FL (16
). It is important to further examine the link between FL use and dietary quality, specifically among low income groups because FL have the potential to become one of the most important tools for providing nutrition education and information to the public (16
) and to equalize income disparities.
Associations between income and dietary quality and between FL nutrition panel use and dietary quality have been previously established. We are unaware, however, of any published research that has examined whether there is a statistical interaction between income category and FL use on dietary quality. Thus, the main objective of this study was to analyze data from the 1994–1996 Continuing Survey of Food Intake by Individuals (CSFII) and the Diet and Health Knowledge Survey (DHKS) to examine whether the relationship between income and dietary quality is modified by FL use.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The 1994–96 CSFII and DHKS are two major nationally representative surveys that form part of the National Nutrition Monitoring and Related Research Program in the United States. Both surveys combined are known as the 1994–1996 "What We Eat in America Survey" coordinated by the Agricultural Research Service of the USDA. This nationally representative survey was designed to collect extensive data on the types and amounts of foods eaten by Americans as well as their knowledge and attitudes about diet and health (17
).
Using a master sample of 1990 U.S. population estimates from the March 1994 Current Population Survey, a nationally representative sample was obtained for the 1994–1996 CSFII/DHKS. A stratified, multistage, area probability sample design was used to draw the sample. This sample consisted of noninstitutionalized persons living in households in the United States from 40 analytic domains defined by sex, age (10 groups) and income levels (17
).
For the DHKS, sample individuals were selected from among CSFII Day 1 intake respondents (excluding proxies) 
20 y old. If more than one eligible sample individual lived in a particular household, one was randomly selected using a specially designed sampling program in the interviewer’s laptop computer (17
). The distribution of age, sex and income characteristics for the DHKS respondents was similar to that of the CSFII respondents. The overall response rates were 76.1% for the CSFII and 73.5% for DHKS. The DHKS participation rate for those who completed the Day 1 24-h recall was 91.6% (17
).
Variables and statistical modeling.
The hypothesis explored in this paper was that the known association between income and dietary quality is modified by FL use. This hypothesis was tested on the subsample of 20- to 60-y-old DHKS respondents (n = 5765) who were either the household meal preparers, meal planners or food shoppers (n = 2952). The CSFII included three specific questions in the household CSFII survey to identify who were the household meal preparers, shoppers and/or planners (17
).
The 1994–96 CSFII/DHKS data set was purchased in CD-ROM format from the USDA (18
). SPSS for Windows (19
) was used to handle and merge data subsets and to run the preliminary multivariate logistic regression analyses. Final descriptive statistics by food stamp (FS) status and the mutlivariate logistic regression models were run with WesVar software to take into account the design effect and adjust the standard errors of the regression parameters accordingly (20
).
The key dependent variable used was the Healthy Eating Index (HEI); the index measures how closely individuals meet recommended Food Guide Pyramid intakes (n = 5 items) and dietary guidelines regarding fat, cholesterol and sodium intakes (n = 5 items). Each correct item counts 10 points, giving a maximum possible score of 100. The HEI data set used was obtained from the USDA Center for Nutrition Policy and Promotion and merged with the CSFII/DHKS data set using the individual identifier variables. The HEI was computed for each of the two CSFII intake days based on the 24 h-recall data (21
).
Multivariate logistic regression was used to test for the FL · income category interaction on the HEI (expressed as < vs. 62.8; the value corresponds to the unweighted median of the 2-d HEI average of all 1994–1996CSFII/DHKS respondents) after controlling for respondent’s level of education, gender and age, race, interview language, household FS status, nutrition knowledge and season of the year in which the interview was conducted. Covariates were selected on the basis of a review of previous empirical evidence and exploratory bivariate analyses with analytical sample. Except for nutrition knowledge, which was computed from DHKS, these variables were derived from the1994–1996 CSFII household questionnaire. Two income categories were used [
350% poverty line (PL) vs. >350% PL] for these analyses.
DHKS survey respondents were asked five questions about FL. One of the questions asked was: "When you buy foods, do you use the nutrition panel that tells the amount of calories, protein, fat, and such in a serving of a food often, sometimes, rarely, or never?" For the purpose of these analyses, those who indicated using the FL often or sometimes were classified as users and those answering rarely, never, or don’t know were classified as nonusers. DHKS included four sets of nutrition knowledge questions on food fat content (14 items), food groups (5 items), obesity/health relationships (open ended with a maximum of four possible correct answers) and FL (9 items). The scoring system used in these analyses was based first on computing an additive score per set in which each item was coded as either 0 if incorrect or 1 if correct. The median split within each set was used to classify the "set score" as either low (0) or high (1). Finally, a global score was obtained by adding up the sets scores, i.e., possible global score range (0–4). Based on bivariate analyses with HEI, the nutrition knowledge global score was recoded as low (score, 0–3) or high (score, 4).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
.
|
, Fig. 1
). Those who were wealthier and used FL were significantly less likely to have a low HEI compared with the reference group formed by those in the lower income category who did not use the FL nutrition panel [odds ratio (OR) = 0.42; 95% confidence interval (CI): 0.31, 0.56]. By contrast, those who were wealthier but did not use the FL were as likely as the reference group to have a low HEI. Those who were in the lower income category but used the FL were significantly less likely to have a low HEI compared with the reference group (OR = 0.62; 95% CI: 0.48, 0.80 ). Thus, among FL users, those with a higher income level were less likely than their lower income counterparts to have a lower HEI (OR: 0.42 vs. 0.62, respectively). Among higher income respondents, those who used FL were less likely to have a low HEI compared with their counterparts who did not use the FL nutrition panel (OR: 0.42; 95% CI: 0.31, 0.56 vs. OR: 1.08; 95% CI: 0.74, 1.54, respectively).
|
|
. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
).
Again, as expected (4
–6
), the level of formal education was positively associated with dietary quality. An intriguing finding, however, was that those with the lowest level of education were not as affected as those with intermediate levels of formal education. This result should be interpreted with caution because only a small fraction of respondents were in the lowest category of the education classification (i.e., <8th grade). However, it may suggest that those with the lowest levels of education may be more likely to access programs that may buffer the effect of low social status on dietary quality.
African-Americans were at substantially higher risk of consuming diets of suboptimal nutritional value compared with Caucasians even after controlling for income and level of education. It is possible that some food choices among African-Americans are driven by cultural habits and beliefs that may not be consistent with dietary guidelines (e.g., high consumption of saturated fat).
DHKS respondents interviewed in Spanish had better dietary quality than those interviewed in English after adjusting for numerous socioeconomic and demographic confounders including race/ethnicity. This suggests positive dietary behaviors that may be attributed to the Hispanic culture. This result, however, must be interpreted with caution because a very small proportion of respondents were interviewed in Spanish.
Respondent’s gender was not associated with dietary quality. This is not surprising because this analytical sample was based on household food shoppers, handlers and/or meal preparers. Contrary to what is found in many developing countries, the season of the year when the interview was conducted was unrelated to the respondent’s dietary quality. This may be explained by a more stable supply of a variety of foods in the United States than in less developed countries.
The interactive model tested indicates that the influence of income on dietary quality is modified by use of FL. Specifically, wealthier individuals that do not use FL are as likely as lower income individuals who do not use FL to have a suboptimal dietary quality. This suggests that income does not make a difference in dietary quality if FL are not used. Among those who used FL, the positive influence of this practice on dietary quality was stronger if they had higher rather than lower income. This finding suggests that the benefit of FL use is greater among higher than among lower income individuals.
These analyses have several methodological limitations. First, they are based on a cross-sectional sample, which does not assist with establishing the temporal sequence of events. Thus, the question remaining is: does FL use lead to improved dietary quality? Do individuals who have healthier lifestyles, including dietary intakes compatible with dietary guidelines, choose to use FL? Are both forces operating in a cyclical manner? Undoubtedly, these are public health nutrition questions of great importance that can be answered conclusively only through sound prospective intervention studies (23
). Second, both the dependent and key independent variables are based on self-reported behaviors such as dietary intake and FL use. Thus, it is reassuring that a recent study has found substantial agreement between self-reported FL use and serum carotenoids levels (15
). Further validation studies involving direct observations of self-reported behaviors are warranted in different socioeconomic and cultural contexts. Third, these analyses do not explain whether FL are associated with general dietary quality improvements or whether there are specific dietary components that are more responsive to FL use. Future analyses concentrating on individual components of the HEI are likely to shed some light on this question.
In conclusion, FL use is associated with improved dietary quality among all income groups with a greater benefit of use among higher income individuals. Income is not associated with improved dietary quality in the absence of FL use.
|
FOOTNOTES |
|---|
2 Funded by the U.S. Department of Agriculture Economic Research Service through UC Davis Research contract no. K-981834–07.
4 Present address: Department of Nutrition and Foodservice Systems, University of North Carolina at Greensboro.
5 Abbreviations used: 95% CI, 95% confidence interval; CSFII, Continuing Survey of Food Intake by Individuals; DHKS, Diet and Health Knowledge Survey; FL, food label; FS, food stamps; HEI, Healthy Eating Index; NFS, non-FS; PL, poverty line; OR, odds ratio.
Manuscript received 13 November 2001. Initial review completed 4 December 2001. Revision accepted 14 January 2002.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Siegel, M. & Doner, L. (1998) Emerging threats to the public’s health: the need for social change. Siegel, M. Doner, L. eds. Marketing Public Health: Strategies to Promote Social Change 1998 Aspen Publishers Gaithersburg, MD. .
2. U.S. Department of Health and Human Services (1996) Healthy People 2000: Midcourse Review and 1995 Revisions 1996 Jones and Bartlett Publishers Sudbury, MA. .
3. Brown, J. & Pollitt, E. (1996) Malnutrition, poverty, and intellectual development. Sci. Am. 247:38-43.
4.
Buttriss, J. L. (1997) Food and nutrition: attitudes, beliefs, and knowledge in the United Kingdom. Am. J. Clin. Nutr. 65(suppl.):1985S-1995S.
5. Kirby, S. D., Baranowski, T, Reynolds, K. D., Taylor, G. & Binkley, D. (1995) Children’s fruit and vegetable intake: socioeconomic, adult-child, regional, and urban-rural influences. J. Nutr. Educ. 27:261-271.
6. Morton, J. F. & Guthrie, J. F. (1997) Diet-related knowledge, attitudes and practices of low income individuals with children in the household. Fam. Econ. Nutr. Rev. 10:2-16.
7. Zive, M. M., Frank-Spohrer, G. C., Sallis, J. F., McKenzie, T. L., Elder, J. P., Berr, C. C., Broyles, S. L. & Nader, P. R. (1998) Determinants of dietary intake in a sample of white and Mexican-American children. J. Am. Diet. Assoc. 98:1282-1289.[Medline]
8.
Kunzt, A. E., Groenhof, F., Andersen, O., Borgan, J. K., Costa, G., Desplanques, G., Filakti, H., Giraldes, M. D., Faggiano, F, Harding, S., Junker, C, Martikainen, P., Minder, C., Nolan, B., Pagnanelli, F., Regidor, E., Vågerö, D, Valkonen, T. & Mackenbach, J. P. (1999) Occupational class and ischemic heart disease mortality in the United States and 11 European countries. Am. J. Public Health 89:47-53.
9. Brecher, S. J., Bender, M. M., Wilkening, V. L., McCabe, N. M. & Anderson, E. M. (2000) Status of nutrition labeling, health claims, and nutrient content claims for processed foods: 1997 food label and package survey. J. Am. Diet. Assoc. 100:1057-1062.[Medline]
10.
Zarkin, G. A., Dean, N., Mauskopf, J. & Williams, R. (1993) Potential health benefits of nutrition label changes. Am. J. Public Health 83:717-724.
11. Neuhouser, M., Kristal, A. & Patterson, R. (1999) Use of food labels is associated with lower fat intake. J. Am. Diet. Assoc. 99:45-46.[Medline]
12. Krueter, M. W., Brennan, L. K., Scharff, D. P. & Lukwago, S. N. (1997) Do nutrition label readers eat healthier diets? Behavioral correlates of adults’ use of food labels. Am. J. Prev. Med 13:277-283.[Medline]
13.
Kristal, A., Levy, L., Patterson, R., Li, S. & White, E. (1998) Trends in food label use associated with new nutrition labeling regulations. Am. J. Public Health 88:1212-1215.
14. Haldeman, L. & Pérez-Escamilla, R. () Does Food Label Use and Nutrition Knowledge Affect Dietary Quality?. (1999) Results from the 1994–96 CSFII/DHKS, p. 37 (abs.). Society for Nutrition Education Annual Meeting Proceedings, Baltimore, MD, July 24–28.
15. Murphy, S. P., Kaiser, L. L., Townsend, M. S. & Allen, L. H. (2001) Evaluation of validity of items for a food behavior checklist. J. Am. Diet. Assoc. 101:751-761.[Medline]
16. Brunt, A. R. & Shafer, E. (1997) Helping consumers with low literacy interpret the new food label. J. Nutr. Educ. 29:224A-225A.
17. Tippett, K. S. & Yasmin, S. C. (1998) Design and Operation: The Continuing Survey of Food Intakes by Individuals and the Diet and Health Knowledge Survey, 1994–1996 1998 U.S. Department of Agriculture, Agricultural Research Service Nationwide Food Surveys Report No. 96–1..
18. U.S. Department of Agriculture, Agriculture Research Service (1998) 1994–96 Continuing Survey of Food Intakes by Individuals and 1994–96 Diet and Health Knowledge Survey. CD-ROM 1998Available from National Technical Information Service, Springfield, VA..
19. SPSS Inc. (1996) SPSS for Windows: version 7 1996 SPSS Chicago IL. .
20. SPSS, Inc. (1998) WesVar. Complex Samples 3.0 1998 SPSS Chicago, IL. .
21. Bowman, S. A., Lino, M., Gerrior, S. A. & Basiotis, P. P. (1998) The Healthy Eating Index:1994–1996 1998 U.S. Department of Agriculture, Center for Nutrition Policy and Promotion CNPP-5.
22. Joy, A. B. & Doisy, C. (1996) Food Stamp Nutrition Education Program: assisting food stamp recipients to become self-sufficient. J. Nutr. Educ. 28:123-126.
23. Haldeman, L. A. (2001) Development and Impact of a Food Label Education Intervention on Food Label Knowledge, Self-Efficacy and Stage of Change among Latinos: a Randomized Community Trial 2001 Doctoral thesis, University of Connecticut Storrs, CT. .
This article has been cited by other articles:
|
|
 |
|
  Y. H. Chu, E. A. Frongillo, S. J. Jones, and G. L. Kaye Improving Patrons' Meal Selections Through the Use of Point-of-Selection Nutrition Labels Am J Public Health, November 1, 2009; 99(11): 2001 - 2005. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
