Abstract
This cross-sectional study assessed: (a) awareness and knowledge of federal dietary tools (MyPyramid, Food Guide Pyramid (FGP), and food labels (FL)), and (b) the influence of acculturation and state of residence on FGP knowledge (FGPK) indicators among low-income Latina WIC participants (N = 479) living in Connecticut, Ohio, Texas, and California. Participants were familiar with FGP but only 37% recognized MyPyramid. FGPK was highest for fruits (71%) and lowest for the ‘breads and cereals’ group (12%). Less than half (47%) used FL’s when grocery shopping. Living in OH, was associated with the lowest FGPK. Multivariate analyses showed that more acculturated individuals living in CT/CA had better FGPK than participants living in TX and their less acculturated counterparts in CT/CA. The forthcoming revised federal dietary tools need to be adequately disseminated among Latinos, with special emphasis on those with lower acculturation levels, living in rapid emerging Latino communities or in the US–Mexico border.
Keywords: Dietary Guidelines, Hispanic, Latino, Food Guide Pyramid, Food label, Nutrition education/knowledge, WIC
Introduction
Latinos are the largest minority group representing 15% of the US population and it is projected that by 2050 they will represent 25% of the population [1, 2]. In 2007–2008, obesity rates for Latinas were 43% compared to 33% among their white counterparts [3]. Behavior change theories posit that even though dietary informational and applied (instrumental) knowledge are not a sufficient condition, they are necessary for addressing the obesity epidemic [4, 5]. Very few studies have examined food labels (FL), the FGP, and Dietary Guidelines knowledge [6–17] among Latinos.
Latinos share a common language, however, different subgroups differ substantially in dietary and other lifestyle behaviors as well as in poverty levels and disease risks [1, 18]. Accordingly, effective family nutrition education efforts targeting Latinos need to take into account acculturation [19] and geographic area of residence.
Federal Dietary Tools
The Federal government issued the current Dietary Guidelines for Americans in 2005. These guidelines provide the scientific basis for the development of tools that can assist the general population with the adoption of healthier dietary practices that can help curb the obesity and chronic disease epidemic facing the nation [20]. Two key dietary tools available to consumers to translate the guidelines into improved practices are food labels (FL) [21] and MyPyramid [22]. Pérez-Escamilla and Haldeman [10] showed that FL use buffers against the negative impact of poverty on dietary quality. They also showed that hands-on FL education improves FL knowledge and self-efficacy among Latinas [13]. MyPyramid is the current official government visual tool that is intended to be used together with the MyPyramid.gov website, a “one-stop shopping” internet site with the goal of personalizing information regarding nutrition and physical activity according to the 2005 Dietary Guidelines for Americans [22].
Objectives
The objectives of our study were to examine: (a) awareness and knowledge of federal dietary tools (MyPyramid, Food Guide Pyramid (FGP), and Food labels (FL)), and (b) the influence of acculturation and state of residence on FGP knowledge indicators among low-income Latina WIC participants (N = 479) living in Connecticut, Ohio, Texas, and California. This is a very timely topic as the 2005 Dietary Guidelines for Americans are currently being revised and will be released in 2010.
Methods
This was a cross-sectional, multi-state study conducted from March to August 2006 that included 479 participants. Participants were non-pregnant Latinas 16 years or older who were receiving WIC benefits recruited in CT (n = 101), OH (n = 100), TX (n = 174) and CA (n = 104). Recruitment took place at: (1) WIC offices, local supermarkets, community agencies (CT); (2) community-based agencies working with Latino families (OH); (3) hospital where majority of women were WIC clients (TX); and (4) WIC Clinic (CA). The research protocol was approved by the human subjects Institutional Review Board (IRB) in all research centers (University of Connecticut, The Ohio State University, The University of Texas at El Paso, and Loma Linda University).
The 24-item survey developed for this study was organized into 3 domains: (1) dietary tools (2) socio-demographics, and (3) family cohesiveness and acculturation. The time needed to complete the questionnaire was 10–20 min. These domains were chosen because both theory and empirical evidence indicated that collecting data on these constructs was essential for understanding federal nutrition awareness tools and knowledge among Latinos. The questionnaire was tested for construct validity by the authors of this work with expertise in the fields of Latino acculturation, nutrition, and health. The survey was applied in either Spanish (65%) or English (35%). The survey was first developed in English, then translated into Spanish and finally back translated into English by the study PI (RPE) to ensure that both versions had equivalent meanings. No discrepanies were found between English and Spanish versions. This was not surprising as the questionnaire was brief and very simple, and most importantly most items were drawn from instruments used previously by the study PI and co-PIs in the target communities.
The survey was then pre-tested with 10 WIC Latina clients in CA to assess readability and required level of literacy (i.e., less than high school education). In all sites, the questionnaire was intially applied/responded in front of the field work supervisors to find out if there were any items that were not properly understood in the final questionnaire. Results of this “qualitative” verification showed that all questionnaire items were properly understood across sites. Participants were asked if they recognized the dietary tools (FGP, MyPyramid, and FL) and whether they knew what the FGP is intended for and the recommended daily serving ranges per food group included in FGP and MyPyramid.
Data were collected on participant’s age, education, household size, monthly food expenditures, and country of birth. Birth place and time living in the US were included as acculturation proxies. The acculturation scale included: reading/speaking language preferences, country where respondent grew up, ethnicity of friends, and ethnic pride. Family cohesiveness was measured by asking if family members: (a) do things together; (b) discuss problems together; (c) feel close to each other. Consistent with Balcazar et al. [23–25] the Cronbach’s alpha coefficients in our study were acceptable; 0.90 for the acculturation scale and 0.71 for the family cohesiveness scale.
In CT interviews were conducted by three community interviewers. In CA, TX and OH interviews were self-administered under the supervision of community outreach workers.
Statistical Analyses
Once data were entered and cleaned at each site they were sent in electronic form to the study headquarters at the University of Connecticut where a pooled data set was created and analyzed with SPSS for Windows version 17.0. Bivariate analyses were conducted to compare across sites: awareness and knowledge of federal dietary tools as well as socio-economic, cultural, and demographic characteristics. One-way ANOVA was used to compare continuous variables and the chi-square test to compare categorical variables.
Backward stepwise logistic multivariate regressions were conducted to examine if: (a) the association between acculturation and FGP awareness/knowledge was modified by state, and (b) the association between state and FGP awareness/knowledge was modified by acculturation level. These analyses adjusted for respondent’s age, family size and monthly food expenditure. Three regressions were run each using a different measure or proxy of acculturation: (a) acculturation scale (reflecting language preferences, ethnicity of friends and country where respondent grew) (b) birth place and (c) time living in the US. Statistical significance in the bivariate analyses was based on a P value ≤0.05. Multivariate logistic regression results were expressed as odds ratios (OR) and corresponding 95% Confidence Intervals (95% CI). Associations were considered to be significant if the 95% CI excluded the value of 1.
Variable Definitions
Respondents were classified as having either a ‘lower’ or ‘higher’ level of acculturation using a score of 12 as the cutoff (scale score range 0–20). The Food Guide Pyramid score and MyPyramid score were computed based on whether the respondents could correctly recognize what was the dietary tool and whether they could correctly describe the purpose of the FGP and MyPyramid and the recommended daily servings of different food groups. The knowledge score ranged from 0 to 7 for FGP and 0 to 6 for MyPyramid. Those with a score of 4 or higher were considered to have a ’high’ FGP or MyPyramid knowledge score.
For the regression analyses testing the association of acculturation score or birth place with recognition of dietary tools, respondents from CT and CA were pooled into one group and jointly compared with respondents from Texas. This post-hoc decision allowed us to increase the statistical power of our analysesand wasbased on exploratory analyses showing that the association between acculturation score or birth place with dietary tools outcomes among respondents from CT and CA behaved similarly. Likewise, when analyzing the association of time living in the US with recognition of dietary tools, respondents from CT, CA, and OH were pooled in a single group and compared against respondents from TX. Respondents from OH could only be included in the regression model examining time living in the US as 96% had ‘lower’ acculturation and none of them were born in the continental US.
Results
Sample Characteristics
The average age of participants ranged from 26 to 33 years. In all locations the participants had less than a high school education. Household size ranged from around 4 in CT to 5 in CA. The percentage of participants born in the continental US ranged from 0% in OH to 41.3% in CA. As expected among those born outside of the US, participants from CT were more likely to have been born in Puerto Rico and their counterparts from OH, CA and TX were more likely to have been born in Mexico. The percentage of participants living in the US for at least 10 years ranged from 19% in OH to 65% in CA (Table 1).
Table 1.
Socioeconomic and demographic characteristics among Latina WIC participants by site
| All | CT | OH | CA | TX | Pa | |
|---|---|---|---|---|---|---|
| Number of respondents | n = 479 | n = 101 | n = 100 | n = 104 | n = 174 | |
| Age (years) | 27.8 ± 7.5 | 26.3 ± 5.7 | 33.4 ± 9.9 | 26.5 ± 6.8 | 26.3 ± 5.7 | .000 |
| Education (years) | 10.1 ± 3.4 | 9.7 ± 3.2 | 9.5 ± 4.2 | 10.4 ± 3.4 | 10.4 ± 3.0 | .087 |
| Family size (No.) | 4.6 ± 1.8 | 4.2 ± 1.5 | 4.6 ± 1.6 | 5.2 ± 2.3 | 4.5 ± 1.8 | .002 |
| Food expenditure per month | .033 | |||||
| ≤$200 | 122 (26.6%) | 23 (23.0%) | 17 (17.7%) | 28 (27.2%) | 54 (33.8%) | |
| >$200 | 337 (73.4%) | 77 (77.0%) | 79 (82.3%) | 75 (72.8%) | 106 (66.3%) | |
| Birth place | .000 | |||||
| Mexico | 294 (61.6%) | 27 (26.7%) | 76 (76.0%) | 51 (49.0%) | 140 (81.4%) | |
| Continental US | 104 (21.8%) | 29 (28.7%) | 0 (0%) | 43 (41.3%) | 32 (18.6%) | |
| Puerto Rico | 37 (7.8%) | 33 (32.7%) | 3 (3%) | 1 (1.0%) | 0 (0%) | |
| Central America | 21 (4.4%) | 6 (5.9%) | 8 (8.0%) | 7 (6.7%) | 0 (0%) | |
| South America | 11 (2.3%) | 0 (0%) | 11 (11%) | 0 (0%) | 0 (0%) | |
| Other | 10 (2.1%) | 6 (5.9%) | 2 (2.0%) | 2 (1.9%) | 0 (0%) | |
| Years living in the US | .000 | |||||
| <10 years | 247 (53.8%) | 52 (53.6%) | 79 (80.6%) | 36 (34.6%) | 80 (50.0%) | |
| ≥10 years | 212 (46.2%) | 45 (46.4%) | 19 (19.4%) | 68 (65.4%) | 80 (50.0%) | |
| Acculturationb | .000 | |||||
| Low | 333 (72.4%) | 61 (62.2%) | 95 (96.9%) | 50 (51.5%) | 127 (76%) | |
| High | 127 (27.6%) | 37 (37.8%) | 3 (3.1%) | 47 (48.5%) | 40 (24%) | |
| Family cohesionc | .484 | |||||
| Low | 40 (8.4%) | 9 (8.9%) | 12 (12.0%) | 7 (6.7%) | 12 (7.1%) | |
| High | 434 (91.6%) | 92 (91.1%) | 88 (88.0%) | 97 (93.3%) | 157 (92.9%) |
χ2 for categorical variables, one-way ANOVA for continuous variables
Acculturation score based on whether the respondent: (a) speaks, read mainly English or Spanish (b) friends are mainly Americans from the United States or Hispanic (c) spent early life in the United States or Latin America
Family cohesion based on whether the respondent’s family do things or discuss problems together, and feel close to each other
Awareness and Knowledge of Federal Dietary Tools
Food Guide Pyramid
The percentage of participants familiar with the FGP ranged from 76% in OH to 98% in CT and TX. The proportion responding that the FGP is a ‘guide for healthy eating’ ranged from 74% in OH to 95% in CT. The percentage knowing the correct daily recommended servings of fruit fluctuated between 49% in OH and 86% in CT. Correct knowledge of vegetable servings ranged from 41% in OH to 66% in CA. The corresponding figures for ‘grains and cereals’ servings ranged from 6% in OH to 19% in CA. The correct answer for dairy group servings fluctuated between 42% in OH and 60% in CT. The corresponding figure for ‘meat and meat alternates’ servings ranged from 41% in OH to 58% in CT. Consistent with these findings, the proportion with a high overall FGP knowledge score was lowest in OH and highest in CT and CA (Table 2).
Table 2.
Familiarity and knowledge of Food Guide Pyramid, MyPyramid and food label by site
| All | CT | OH | CA | TX | Pa | |
|---|---|---|---|---|---|---|
| Food Guide Pyramid | ||||||
| FGP recognition | .000 | |||||
| No | 37 (7.8%) | 2 (2.0%) | 23 (23.7%) | 9 (8.7%) | 3 (1.7%) | |
| Yes | 438 (92.2%) | 99 (98.0%) | 74 (76.3%) | 94 (91.3%) | 171 (98.3%) | |
| What is FGP | .000 | |||||
| Incorrect | 60 (12.6%) | 5 (5.0%) | 26 (26.0%) | 11 (10.6%) | 18 (10.4%) | |
| Correct | 418 (87.4%) | 96 (95.0%) | 74 (74.0%) | 93 (89.4%) | 155 (89.6%) | |
| Fruit servings | .000 | |||||
| Incorrect | 129 (27.4%) | 13 (13.7%) | 50 (51.0%) | 26 (25.0%) | 40 (23.1%) | |
| Correct | 341 (72.6%) | 82 (86.3%) | 48 (49.0%) | 78 (75.0%) | 133 (76.9%) | |
| Vegetable servings | .001 | |||||
| Incorrect | 208 (44.4%) | 48 (50.5%) | 58 (59.2%) | 35 (33.7%) | 67 (39.2%) | |
| Correct | 260 (55.6%) | 47 (49.5%) | 40 (40.8%) | 69 (66.3%) | 104 (60.8%) | |
| Grains & cereals servings | .014 | |||||
| Incorrect | 407 (87.0%) | 89 (91.8%) | 92 (93.9%) | 83 (81.4%) | 143 (83.6%) | |
| Correct | 61 (13.0%) | 8 (8.2%) | 6 (6.1%) | 19 (18.6%) | 28 (16.4%) | |
| Dairy servings | .047 | |||||
| Incorrect | 224 (47.5%) | 40 (40.4%) | 57 (58.2%) | 43 (41.7%) | 84 (48.8%) | |
| Correct | 248 (52.5%) | 59 (59.6%) | 41 (41.8%) | 60 (58.3%) | 88 (51.2%) | |
| Meat & alternates servings | .080 | |||||
| Incorrect | 228 (48.1%) | 42 (42.4%) | 58 (59.2%) | 46 (44.2%) | 82 (47.4%) | |
| Correct | 246 (51.9%) | 57 (57.6%) | 40 (40.8%) | 58 (55.8%) | 91 (52.6%) | |
| FGP scoreb | .000 | |||||
| Low | 119 (26.0%) | 12 (12.9%) | 41 (42.7%) | 20 (19.8%) | 46 (27.4%) | |
| High | 339 (74.0%) | 81 (87.1%) | 55 (57.3%) | 81 (80.2%) | 122 (72.6%) | |
| MyPyramid | ||||||
| MyPyramid recognition | .018 | |||||
| No | 299 (63.1%) | 66 (65.3%) | 72 (72.0%) | 52 (51.0%) | 109 (63.7%) | |
| Yes | 175 (36.9%) | 35 (34.7%) | 28 (28.0%) | 50 (49.0%) | 62 (36.3%) | |
| Fruit servings | .033 | |||||
| Incorrect | 396 (83.7%) | 90 (89.1%) | 90 (90.0%) | 79 (79.0%) | 137 (79.7%) | |
| Correct | 77 (16.3%) | 11 (10.9%) | 10 (10.0%) | 21 (21.0%) | 35 (20.3%) | |
| Vegetable servings | .005 | |||||
| Incorrect | 382 (80.8%) | 85 (84.2%) | 90 (90.0%) | 70 (70.7%) | 137 (79.2%) | |
| Correct | 91 (19.2%) | 16 (15.8%) | 10 (10.0%) | 29 (29.3%) | 36 (20.8%) | |
| Grains & cereals servings | .418 | |||||
| Incorrect | 446 (94.1%) | 98 (97.0%) | 95 (95.0%) | 92 (92.0%) | 161 (93.1%) | |
| Correct | 28 (5.9%) | 3 (3.0%) | 5 (5.0%) | 8 (8.0%) | 12 (6.9%) | |
| Dairy servings | .134 | |||||
| Incorrect | 406 (85.8%) | 91 (90.1%) | 89 (89.9%) | 85 (85.0%) | 141 (81.5%) | |
| Correct | 67 (14.2%) | 10 (9.9%) | 10 (10.1%) | 15 (15.0%) | 32 (18.5%) | |
| Meat & alternates servings | .379 | |||||
| Incorrect | 448 (94.5%) | 98 (97.0%) | 96 (96.0%) | 94 (94.0%) | 160 (92.5%) | |
| Correct | 26 (5.5%) | 3 (3.0%) | 4 (4.0%) | 6 (6.0%) | 13 (7.5%) | |
| MyPyramid scorec | .064 | |||||
| Low | 423 (90.4%) | 96 (95.0%) | 93 (93.9%) | 87 (88.8%) | 147 (86.5%) | |
| High | 45 (9.6%) | 5 (5.0%) | 6(6.1%) | 11 (11.2%) | 23 (13.5%) | |
| Food label | ||||||
| Food label recognition | .748 | |||||
| No | 38 (8.0%) | 10 (10.0%) | 6 (6.0%) | 9 (8.7%) | 13 (7.5%) | |
| Yes | 438 (92.0%) | 90 (90.0%) | 94 (94.0%) | 94 (91.3%) | 160 (92.5%) | |
| Food label utilization | .043 | |||||
| Rare or sometimes | 383 (80.3%) | 79 (78.2%) | 73 (73.0%) | 82 (78.8%) | 149 (86.6%) | |
| Often or quite often | 94 (19.7%) | 22 (21.8%) | 27 (27.0%) | 22 (21.2%) | 23 (13.4%) |
χ2 for categorical variables, one-way ANOVA for continuous variables
Food Guide Pyramid knowledge score based on whether the respondents: (a) could correctly recognize the Food Guide Pyramid and (b) knew the Food Guide Pyramid recommended daily servings of different food groups
MyPyramid knowledge score based on whether the respondents: (a) could correctly recognize the MyPyramid and (b) knew the MyPyramid recommended daily servings of different food groups
My Pyramid
As expected, MyPyramid recognition was significantly lower than for FGP in the pooled sample (37 vs. 92%). MyPyramid recognition ranged from 28% in OH to 49% in CA. The percentage who knew the daily recommended fruit servings fluctuated between 10% in OH and 21% in CA. Likewise, whereas only 10% of participants from OH knew the recommended vegetable servings, this was true for 29% of CA respondents. Knowledge of ‘grain and cereals’ recommended servings was uniformly low with only 6% of the pooled sample answering correctly. The percentage knowing the correct dairy servings ranged from 10% in CT and OH to 18% in TX. Only 5% answered correctly the recommended servings of ‘meat and meat alternates’ and this knowledge was uniformly low across states. The proportion of the sample with a high MyPyramid score ranged from 5% in CT to 13% in TX (Table 2).
Food Labels
The overwhelming majority of respondents (92%) recognized the nutrition facts panel of the FL. However, only 20% used PL’s ‘often’ or ‘quite often’ when purchasing foods with this figure ranging from 13% in TX to 27% in OH (Table 2).
Acculturation and FGP Awareness and Knowledge: Multivariate Analyses
Multivariate regression results examining the interaction between study site and acculturation indicators are presented in Tables 3, 4 and 5. In these tables the reference groups were the TX respondents with the higher acculturation score (Table 3), born outside the US (Table 4), and living in the US for >10 years (Table 5). Results from analyses that used a different reference group presented below (but not shown in tables) are marked with an asterisk.
Table 3.
Association between site and acculturation level with FGP awareness and knowledge among Latinas in the US
| Outcome variables | n | OR | 95% Confidence interval |
|
|---|---|---|---|---|
| Lower | Upper | |||
| What is FGP | ||||
| CT&CA low | 110 | 1.276 | .418 | 3.899 |
| CT&CA high | 84 | 3.125 | .788 | 12.387 |
| TX low | 112 | 2.031 | .621 | 6.647 |
| TX high | 37 | 1.000 | – | – |
| Fruits | ||||
| CT&CA low | 105 | 1.469 | .652 | 3.310 |
| CT&CA high | 83 | 3.410 | 1.344 | 8.657 |
| TX low | 113 | 2.676 | 1.154 | 6.202 |
| TX high | 37 | 1.000 | – | – |
| Vegetables | ||||
| CT&CA low | 105 | 1.104 | .517 | 2.359 |
| CT&CA high | 83 | 1.501 | .680 | 3.311 |
| TX low | 113 | 1.571 | .736 | 3.354 |
| TX high | 36 | 1.000 | – | – |
| Grains & cereals | ||||
| CT&CA low | 105 | .500 | .133 | 1.881 |
| CT&CA high | 83 | 2.794 | .885 | 8.822 |
| TX low | 112 | 2.132 | .686 | 6.629 |
| TX high | 37 | 1.000 | – | – |
| Dairy | ||||
| CT&CA low | 107 | .946 | .444 | 2.016 |
| CT&CA high | 84 | 1.789 | .807 | 3.967 |
| TX low | 113 | .944 | .445 | 2.000 |
| TX high | 36 | 1.000 | – | – |
| Meat & alternates | ||||
| CT&CA low | 108 | 1.454 | .682 | 3.098 |
| CT&CA high | 84 | 1.161 | .532 | 2.534 |
| TX low | 113 | 1.227 | .581 | 2.594 |
| TX high | 37 | 1.000 | – | – |
| FGP scorea | ||||
| CT&CA low | 101 | 1.795 | .776 | 4.154 |
| CT&CA high | 82 | 4.056 | 1.523 | 10.802 |
| TX low | 111 | 1.556 | .687 | 3.523 |
| TX high | 36 | 1.000 | – | – |
Backward stepwise multivariate logistic regression, covariates included in all models: age, family size, monthly food expenditure
Food Guide Pyramid knowledge score based on whether the respondents: (a) could correctly recognize the Food Guide Pyramid and (b) knew the Food Guide Pyramid recommended daily servings of different food groups
Table 4.
Association between site and birth place with FGP awareness and knowledge among Latinas in the US
| Outcome variables | n | OR | 95% Confidence interval |
|
|---|---|---|---|---|
| Lower | Upper | |||
| What is FGP | ||||
| CT&CA US | 71 | 1.117 | .324 | 3.847 |
| CT&CA other places | 132 | .667 | .263 | 1.689 |
| TX US | 27 | .293 | .088 | .980 |
| TX other places | 128 | 1.000 | – | – |
| Fruits | ||||
| CT&CA US | 70 | 2.438 | .946 | 6.286 |
| CT&CA other places | 127 | .679 | .373 | 1.233 |
| TX US | 27 | .543 | .213 | 1.386 |
| TX other places | 129 | 1.000 | – | – |
| Vegetables | ||||
| CT&CA US | 70 | 1.204 | .655 | 2.214 |
| CT&CA other places | 127 | .771 | .468 | 1.272 |
| TX US | 27 | 1.068 | .452 | 2.521 |
| TX other places | 127 | 1.000 | – | – |
| Grains & cereals | ||||
| CT&CA US | 70 | 1.764 | .866 | 3.593 |
| CT&CA other places | 127 | .389 | .171 | .885 |
| TX US | 27 | 1.158 | .394 | 3.403 |
| TX other places | 128 | 1.000 | – | – |
| Dairy | ||||
| CT&CA US | 71 | 1.527 | .843 | 2.768 |
| CT&CA other places | 129 | 1.150 | .703 | 1.881 |
| TX US | 27 | .950 | .414 | 2.181 |
| TX other places | 128 | 1.000 | – | – |
| Meat & alternates | ||||
| CT&CA US | 71 | .679 | .378 | 1.221 |
| CT&CA other places | 130 | 1.204 | .728 | 1.990 |
| TX US | 27 | .510 | .218 | 1.192 |
| TX other places | 129 | 1.000 | – | – |
| FGP scorea | ||||
| CT&CA US | 70 | 2.719 | 1.123 | 6.584 |
| CT&CA other places | 122 | 1.234 | .670 | 2.270 |
| TX US | 27 | .514 | .212 | 1.244 |
| TX other places | 125 | 1.000 | – | – |
Backward stepwise multivariate logistic regression, covariates included in all models: age, family size, monthly food expenditure
Food Guide Pyramid knowledge score based on whether the respondents: (a) could correctly recognize the Food Guide Pyramid and (b) knew the Food Guide Pyramid recommended daily servings of different food groups
Table 5.
Association between site and staying time in the US with FGP awareness and knowledge among Latinas in the US
| Outcome variables | n | OR | 95% Confidence interval |
|
|---|---|---|---|---|
| Lower | Upper | |||
| What is FGP | ||||
| CT&OH&CA <10 years | 162 | .390 | .155 | .982 |
| CT&OH&CA >10 years | 131 | 1.251 | .427 | 3.670 |
| TX <10 years | 74 | .884 | .282 | 2.770 |
| TX >10 years | 71 | 1.000 | – | – |
| Fruits | ||||
| CT&OH&CA <10 years | 156 | .518 | .275 | .976 |
| CT&OH&CA >10 years | 130 | 1.259 | .629 | 2.521 |
| TX <10 years | 75 | 1.259 | .574 | 2.763 |
| TX >10 years | 71 | 1.000 | – | – |
| Vegetables | ||||
| CT&OH&CA <10 years | 156 | .580 | .326 | 1.033 |
| CT&OH&CA >10 years | 130 | .964 | .531 | 1.752 |
| TX <10 years | 75 | 1.020 | .521 | 1.994 |
| TX >10 years | 69 | 1.000 | – | – |
| Grains | ||||
| CT&OH&CA <10 years | 158 | .412 | .163 | 1.040 |
| CT&OH&CA >10 years | 128 | 1.130 | .497 | 2.569 |
| TX <10 years | 74 | 1.683 | .706 | 4.009 |
| TX >10 years | 71 | 1.000 | – | – |
| Milk | ||||
| CT&OH&CA <10 years | 159 | .893 | .507 | 1.573 |
| CT&OH&CA >10 years | 130 | 1.620 | .897 | 2.926 |
| TX <10 years | 75 | 1.304 | .672 | 2.532 |
| TX >10 years | 70 | 1.000 | – | – |
| Meat | ||||
| CT&OH&CA <10 years | 159 | .816 | .464 | 1.436 |
| CT&OH&CA >10 years | 131 | 1.411 | .786 | 2.532 |
| TX <10 years | 75 | 1.410 | .731 | 2.718 |
| TX >10 years | 71 | 1.000 | – | – |
| FGP scorea | ||||
| CT&OH&CA <10 years | 153 | .760 | .406 | 1.421 |
| CT&OH&CA >10 years | 126 | 2.140 | 1.043 | 4.393 |
| TX <10 years | 73 | 1.161 | .549 | 2.458 |
| TX >10 years | 69 | 1.000 | – | – |
Backward stepwise multivariate logistic regression, covariates included in all models: age, family size, monthly food expenditure
Food Guide Pyramid knowledge score based on whether the respondents: (a) could correctly recognize the Food Guide Pyramid and (b) knew the Food Guide Pyramid recommended daily servings of different food groups
Acculturation Score
Multivariate analyses identified an interaction between state and acculturation score on diverse FGP knowledge indicators (Table 3). Specifically, the odds of knowing the daily recommended servings of fruits were significantly higher among respondents living in CT and CA who had a high acculturation score (OR = 3.4) as well as among those living in TX who had a low acculturation score (OR = 2.7) compared with TX participants with a high acculturation score.
Compared with respondents living in TX who had a high level of acculturation, participants from CT and CA with a high level of acculturation tended to have higher odds of knowing the recommended daily servings of grains and cereals (OR = 2.8). The odds for knowing the grains and cereals servings were also higher among those from CT and CA with a high acculturation score (OR = 2.3; 95% CI: 1.1–5.0)* and among those from TX with a low acculturation score (OR = 1.8; 95% CI: 0.9–3.6)* compared with participants from CT and CA with a low acculturation score.
Respondents from CT and CA who had a high acculturation score also had higher odds of having a higher FGP knowledge score (OR = 4.0) compared with their counterparts from TX who also had a high acculturation score (Table 3). Respondents from CT and CA who had a high acculturation score also had higher odds of having a higher FGP knowledge score compared with their state counterparts with a low acculturation score (OR = 2.3; 95% CI: 1.0–5.2)* and participants from TX also with a low acculturation score (OR = 2.6; 95% CI: 1.2–5.9)*.
Birth Place
Significant interactions were also found between respondent’s state and place of birth on diverse FGP knowledge indicators (Table 4). Respondents from CT and CA who were born in the US had higher odds of knowing the daily fruit recommended servings in relationship to: respondents from TX born outside the US (OR = 2.4 Table 4), participants from CT and CA born outside the US (OR = 3.6; 95% CI: 1.4–9.1)* and respondents from TX born in the US (OR = 4.5; 95% CI: 1.4–14.6)*.
Respondents from CT and CA who were born outside the US had lower odds of knowing the daily grains and cereals recommended servings (OR = 0.4) compared with respondents from TX born outside the US. They also had lower odds in relationship to participants from CT and CA (OR = 0.2; 95% CI: 0.1–0.5)*, and TX (OR = 0.3; 95% CI: 0.1–1.1)* born in the US.
Respondents from CT and CA who were born in the US had higher odds of having a higher FGP knowledge score compared with respondents from TX born outside the US (OR = 2.7) and also with respondents from TX born in the US (OR = 5.3; 95% CI: 1.8–16.0)*.
Time Living in US
Significant interactions were also identified between respondent’s state and time living in the US on diverse FGP knowledge indicators (Table 5). Participants from CT, OH and CA who had lived for <10 years in the US had lower odds of being aware about the purpose of the FGP compared with respondents from TX that had lived in the US for >10 years (OR = 0.4) or <10 years (OR = 0.4; 95% CI: 0.2–1.1)*. Participants from CT, OH and CA who had lived for <10 years in the US had lower odds of knowing the daily recommended servings of fruits compared with their counterparts from the same states that had lived in the US for >10 years (OR = 0.4; 95% CI: 0.2– 0.7)*, as well as their counterparts from TX who had lived in the US for <10 years (OR = 0.4; 95% CI: 0.2–0.8)* or for >10 years (OR = 0.5).
Participants from CT, OH and CA who had lived for <10 years in the US had lower odds of knowing the recommended daily ‘grains and cereals’ servings compared with respondents from TX who had lived in the US for <10 years (OR = 0.2; 95% CI: 0.1–0.6)* or for > 10 years (OR = 0.4).
Respondents from CT, OH and CA who had lived for >10 years in the US had higher odds of having a higher FGP score compared with their counterparts from TX that had lived for >10 years in the US (OR = 2.1), or compared with participants from Texas who had lived in the US for <10 years (OR = 1.8; 95% CI: 0.9–3.8)* or in CT, OH and CA but who had lived in the US for <10 years (OR = 2.8; 95% CI: 1.6–5.1)*.
Discussion
To our knowledge, this is the first study to examine awareness and knowledge of federal dietary tools in a multi-sate sample of Latina WIC participants. Overall, the dietary tools knowledge score was low for both MyPyramid and FGP. As expected, given national trends of the growth of Latino communities in different regions of the US [1], our findings showed that participants from OH had immigrated much more recently than participants from CA, CT, and TX. The fact that respondents from OH had substantially lower FGP knowledge scores indicates the need for additional culturally appropriate nutrition education efforts in the rapid emerging Latino communities in the US [1, 26]. It is unfortunate that even though researchers have called for major improvements in the development of nutrition guidance to the public including the Food Guide Pyramid their recommendations have not specifically addressed the needs of the culturally diverse low-income groups in the country [27].
This study confirms that even though Latinas are quite familiar with food labels very few are actually using them often to help with food selections. This finding contrasts with the fact that in the general US population the percent that ‘often’ reads food labels increased from 44% in 2002 to 54% in 2008 [28]. This finding, together with evidence showing that food label reading is associated with better nutrition knowledge among Latinos [29] and dietary quality among low-income groups in general [10] calls for reconsidering the way food labels are designed and the way low-literacy individuals are taught how to use them [30].
Only about one-third of participants had seen MyPyramid about 1 year after it was released. This calls for reconsidering the overall communication approaches that will be used to disseminate the forthcoming revised dietary tools among Latinos. Because MyPyramid is a visual tool that is intended to be jointly used with a computer and the MyPyramid.gov internet site [22] this effort will require addressing low computer access and literacy among Latinos and other socio-economically disadvantaged groups.
In our study, the relationship between acculturation and dietary tools knowledge was context specific. For example, more acculturated individuals living in CT or CA had a higher FGP knowledge score compared with their less acculturated counterparts. However, in TX acculturation score did not influence FGP knowledge. It is possible that in El Paso acculturation may have a different influence given its location right at the US-Mexico border.
Overall, our findings suggest that participants residing in CA or CT with higher levels of acculturation (either high acculturation score, being born in US or living in US for longer) had better FGP knowledge scores than their less acculturated counterparts in their own states or participants residing in TX (with either low or high acculturation levels Tables 3, 4, 5). It is possible that more acculturated individuals in CT or CA have had more access to timely FGP information as a result of being more fluent in English and more likely to include non-Latinos in their social networks. This finding supports the need to improve access to timely and sound culturally appropriate nutrition education and dietary tools among Latinos with lower levels of acculturation or living in the US–Mexico border.
Our study had several limitations. Firstly, because of the logistic complexity of conducting this type of multi-state projects involving partnerships between academic and community/WIC agencies, recruitment strategies varied by site. However, differences in key socioeconomic, demographic, and acculturation factors were consistent with expectations according to respondents’ geographic area of residence. Secondly, the survey was applied by community interviewers in CT but self-applied in the remaining states. However, we do not expect that a bias occurred as a result because our bilingual instrument was very short, simple and self-explanatory, and even when self-applied there were trained community outreach workers overseeing the process. Thirdly, our study did not use a multi dimensional scale to measure acculturation as previously recommended [19]. However, it is encouraging that our findings were highly consistent across acculturation indicators.
Policy Implications
Our findings call for the development, testing, and implementation of low-literacy, culturally appropriate nutrition education tools and curriculums; and facilitating access to them in schools and diverse community settings including the use of community outreach models like community health workers/promotores de salud [31, 32] and social marketing campaigns such as SALUD! [33]. This is of special relevance now as the US government is in the process of revising the Dietary Guidelines for Americans and the federal nutrition tools derived from them. This effort will require community based participatory research (CBPR) approaches based on focus groups to determine culture-specific issues regarding nutrition, as well as cognitive interviews to better understand how to effectively address the needs of culturally diverse low-literacy audiences.
Acknowledgments
We thank Predrag Putnik for support with data management and preliminary analyses. This work was partially funded by the Connecticut NIH EXPORT Center for Eliminating Health Disparities among Latinos, grant # NIH-NCMHD P20MD001765, PI Rafael Pérez-Escamilla.
Contributor Information
Rafael Pérez-Escamilla, Email: rafael.perez-escamilla@yale.edu, Yale School of Public Health, New Haven, CT 06510, USA; Connecticut Center for Eliminating Health Disparities among Latinos, University of Connecticut, 3624 Horsebarn Hill Rd, Storrs, CT 06269-4017, USA.
David Sheng Song, Connecticut Center for Eliminating Health Disparities among Latinos, University of Connecticut, 3624 Horsebarn Hill Rd, Storrs, CT 06269-4017, USA.
Christopher Alan Taylor, Medical Dietetics, The Ohio State University, 453 W 10th Ave, 306A Atwell Hall, Columbus, OH 43210-1234, USA.
Alfredo Mejia, Department of Nutrition, School of Public Health, Loma Linda University, 24493 University Ave, Loma Linda, CA 92354, USA.
Hugo Melgar-Quiñonez, Department of Human Nutrition, Ohio State University, 325 Campbell Hall, 1787 Neil Ave, Columbus, OH 43210-1295, USA.
Hector G. Balcazar, School of Public Health, University of Texas Health Science Center at Houston, 1100N. Stanton, Suite 100, El Paso, TX 79902, USA
Robert L. Anders, School of Nursing, University of Texas at El Paso, 1101N Campbell, El Paso, TX 79902, USA
Sofia Segura-Pérez, Hispanic Health Council, Inc, 175 Main Street, Hartford, CT 06106, USA.
Maria O. Duarte-Gardea, Health Promotion Program, College of Health Sciences, University of Texas at El Paso, 1101N. Campbell Street, El Paso, TX 79902-0581, USA
Jorge M. Ibarra, Health Promotion Program, College of Health Sciences, University of Texas at El Paso, 1101N. Campbell Street, El Paso, TX 79902-0581, USA
References
- 1.Perez-Escamilla R. Health care access among Latinos: implications for social and health care reforms. J Hispanics Higher Educ. 2010;9:43–60. [Google Scholar]
- 2.US Government. [Accessed 17 April 2010];US Minority Population Continues to Grow. Available at: http://www.america.gov/st/diversity-english/2008/May/20080513175840zjsredna0.1815607.html.
- 3.Flegal KM, Carroll MD, Ogden CL, Curtin LR, Glanz K, Lewis FM, Rimer BK. Prevalence and trends in obesity among US adults, 1999–2008. JAMA. 2010;303(3):235–241. doi: 10.1001/jama.2009.2014. [DOI] [PubMed] [Google Scholar]
- 4.Glanz K, Lewis FM, Rimer BK, editors. Health behavior and health education: theory, research, and practice. 2nd ed. San Francisco: Jossey-Bass Publishers; 1997. [Google Scholar]
- 5.Dickin KL, Dollahite JS, Habicht JP. Nutrition behavior change among EFNEP participants is higher at sites that are well managed and whose front-line nutrition educators value the program. J Nutr. 2005;135(9):2199–2205. doi: 10.1093/jn/135.9.2199. [DOI] [PubMed] [Google Scholar]
- 6.Contento IR, Randell JS, Basch CE. Review and analysis of evaluation measures used in nutrition education intervention research. J Nutr Educ Behav. 2002;34(1):2–25. doi: 10.1016/s1499-4046(06)60220-0. [DOI] [PubMed] [Google Scholar]
- 7.Contento I, Balch G, Bronner Y, Lytle L, Maloney S, Olson C, Swadener S. The effectivenes of nutrition education and implications for nutrition education policy, programs, and research: a review of research. J Nutr Educ. 1995;26(6) [Google Scholar]
- 8.Cowburn G, Stockley L. Consumer understanding and use of nutrition labelling: a systematic review. Public Health Nutr. 2005;8(1):21–28. doi: 10.1079/phn2005666. [DOI] [PubMed] [Google Scholar]
- 9.Yancey AK, Kumanyika SK, Ponce NA, McCarthy WJ, Fielding JE, Leslie JP, et al. Population-based interventions engaging communities of color in healthy eating and active living: a review. Prev Chronic Dis. 2004;1(1):A09. [PMC free article] [PubMed] [Google Scholar]
- 10.Perez-Escamilla R, Haldeman L. Food label use modifies association of income with dietary quality. J Nutr. 2002;132(4):768–772. doi: 10.1093/jn/132.4.768. [DOI] [PubMed] [Google Scholar]
- 11.Bermudez-Millan A, Perez-Escamilla R, Damio G, Gonzalez A, Segura-Perez S. Food safety knowledge, attitudes, and behaviors among Puerto Rican caretakers living in Hartford, Connecticut. J Food Prot. 2004;67(3):512–516. doi: 10.4315/0362-028x-67.3.512. [DOI] [PubMed] [Google Scholar]
- 12.Dharod JM, Perez-Escamilla R, Bermudez-Millan A, Segura-Perez S, Damio G. Influence of the fight BAC! food safety campaign on an urban Latino population in Connecticut. J Nutr Educ Behav. 2004;36(3):128–132. doi: 10.1016/s1499-4046(06)60149-8. [DOI] [PubMed] [Google Scholar]
- 13.Haldeman L. Ph. D. Dissertation, University of Connecticut, Department of Nutrition Science. 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. [Google Scholar]
- 14.Pérez-Escamilla R, Himmelgreen D, Ferris A. Community nutritional problems among Latino children in Hartford, Connecticut. Storrs and Hartford, CT: Connecticut Family Nutrition Program Technical Report # 1. 1997
- 15.Perez-Escamilla R, Himmelgreen D, Segura-Millan S, Gonzalez A, Mendez I, Haldeman L. Knowledge of folic acid and neural tube defects among inner-city residents: have they heard about it? J Am Diet Assoc. 1999;99(1):80–83. doi: 10.1016/S0002-8223(99)00021-8. [DOI] [PubMed] [Google Scholar]
- 16.Perez-Escamilla R, Ferris AM, Drake L, Haldeman L, Peranick J, Campbell M, et al. Food stamps are associated with food security and dietary intake of inner-city preschoolers from Hartford, Connecticut. J Nutr. 2000;130(11):2711–2717. doi: 10.1093/jn/130.11.2711. [DOI] [PubMed] [Google Scholar]
- 17.Tanasescu M, Ferris AM, Himmelgreen DA, Rodriguez N, Perez-Escamilla R. Biobehavioral factors are associated with obesity in Puerto Rican children. J Nutr. 2000;130(7):1734–1742. doi: 10.1093/jn/130.7.1734. [DOI] [PubMed] [Google Scholar]
- 18.Aguirre-Molina M, Molina C. Latino population: Who are they? In: Molina C, Aguirre-Molina M, editors. Latino health in the United States: a growing challenge. Washington, DC: American Public Health Association; 1994. pp. 3–22. [Google Scholar]
- 19.Perez-Escamilla R, Putnik P. The role of acculturation in nutrition, lifestyle, and incidence of type 2 diabetes among Latinos. J Nutr. 2007;137(4):860–870. doi: 10.1093/jn/137.4.860. [DOI] [PubMed] [Google Scholar]
- 20.Dietary Guidelines for Americans 2005. [Accessed 8 March 2006];US. Department of Health and Human Services US Department of Agriculture. www.healthierus.gov/dietaryguidelines.
- 21.Tillotson JE. United States nutrition labeling and education act of 1990. Nutr Rev. 1991;49(9):273–276. doi: 10.1111/j.1753-4887.1991.tb07419.x. [DOI] [PubMed] [Google Scholar]
- 22.McDermott AY. MyPyramid.gov. Nutr Clin Care. 2005;8(2):102–104. [PubMed] [Google Scholar]
- 23.Balcazar H, Krull JL, Peterson G. Acculturation and family functioning are related to health risks among pregnant Mexican American women. Behav Med. 2001;27(2):62–70. doi: 10.1080/08964280109595772. [DOI] [PubMed] [Google Scholar]
- 24.Balcazar H, Krull JL. Determinants of birth-weight outcomes among Mexican-American women: examining conflicting results about acculturation. Ethn Dis. 1999;9(3):410–422. [PubMed] [Google Scholar]
- 25.Balcazar H, Peterson G, Krull J. Acculturation and family cohesiveness in Mexican American pregnant women: social and health implications. Fam Community Health. 1997;20:16–31. [Google Scholar]
- 26.Pérez-Escamilla R, Himmelgreen D, Bonello H, González A, Haldeman L, Méndez I, Cruz J, Segura-Millán S. Nutrition knowledge, attitudes, and behaviors among Latinos in the USA: influence of language. Ecol Food Nutr. 2001;40:321–325. [Google Scholar]
- 27.Kinney JM. Challenges to rebuilding the US food pyramid. Curr Opin Clin Nutr Metab Care. 2005;8(1):1–7. [PubMed] [Google Scholar]
- 28.FDA. [Accessed 17 April 2010];Key Findings from 2002 and 2008 US Food and Drug Administration’s Health and Diet Survey. Available at: http://www.fda.gov/Food/LabelingNutrition/ucm202780.htm.
- 29.Boulanger P, Pérez-Escamilla R, Himmelgreen D, González A, Mendez I, Segura-Millán S, Haldeman L. Determinants of nutrition knowledge among low-income, Latino caretakers in Hartford, Conn. J Am Diet Assoc. 2002;102:978–981. doi: 10.1016/s0002-8223(02)90223-3. [DOI] [PubMed] [Google Scholar]
- 30.Haldeman L, Pérez-Escamilla R, Ferris AM, Drake L, Himmelgreen D, Bonello H, González A, Segura-Millán S, Méndez I, Bermúdez A, Romero-Daza N. Development of a color-coded bilingual food label for low-literacy Latino caretakers. J Nutr Educ. 2000;32:152–160. [Google Scholar]
- 31.Balcazar H, Byrd T, Ortiz M, Tondapu S, Chavez M. A randomized community intervention to improve hypertension control among Mexican Americans: using the promotoras de salud community outreach model. J Health Care Poor Underserved. 2009;20:1079–1094. doi: 10.1353/hpu.0.0209. [DOI] [PubMed] [Google Scholar]
- 32.Pérez-Escamilla R, Hromi-Fiedler A, Vega-López S, Bermúdez-Millán A, Segura-Pérez S. Impact of peer nutrition education on dietary behaviors and health outcomes among Latinos: a systematic literature review. J Nutr Educ Behav. 2008;40(4):208–225. doi: 10.1016/j.jneb.2008.03.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Pérez-Escamilla R, Himmelgreen D, Bonello H, Peng YK, Mengual G, González A, Méndez I, Cruz J, Phillips LM. Marketing nutrition among urban Latinos: the SALUD! campaign. J Am Diet Assoc. 2000;100(6):698–701. doi: 10.1016/s0002-8223(00)00203-0. [DOI] [PubMed] [Google Scholar]