|
|
Volume
3:
No. 3, July 2006
ORIGINAL RESEARCH
Development and Reliability
of Brief Dietary Assessment Tools for Hispanics
Patricia Wakimoto, DrPH, Gladys Block, PhD, Shelly Mandel, Norma Medina
Suggested citation for this article: Wakimoto P, Block G, Mandel S, Medina N. Development and reliability of brief dietary assessment tools for
Hispanics. Prev Chronic Dis [serial online] 2006 Jul [date cited]. Available from: URL: http://www.cdc.gov/pcd/issues/2006/
jul/05_0117.htm.
PEER REVIEWED
Abstract
Introduction The Hispanic population is the most rapidly growing ethnic group in the United States. Culturally appropriate and efficient strategies for dietary assessment for this population are currently lacking. To address this issue and promote
a healthy diet for disease prevention, we developed screening tools to assess the fruit, vegetable, and fat intake of
Mexican Americans.
Methods Brief screening tools (screeners) were developed based on national data on Mexican Americans’ dietary intake and were then modified after interviews and field testing. The screeners take less than 10 minutes to administer. A reliability study was conducted from June through September 2000, during which 93 Mexican Americans (39 men, 54 women) completed the screeners twice, 1 month apart. The mean age of the study participants was
36.5 years (range 18–71 years), and 91.4% had been born in Mexico.
Results Correlations between the first and second administration of the screeners were
r = 0.64 for fruits and vegetables and r = 0.85 for dietary fat contributors. In addition, estimates of fruit and vegetable consumption frequency were similar to statewide estimates for Hispanics in California. Reproducibility of reported use of vitamin
supplements at least once per week was high; 84% were classified in the same way both times (P < .001).
Conclusion The screening tools provide a reliable assessment of selected dietary factors
among Mexican Americans. The tools can be scored immediately to provide feedback to respondents. They may be useful in situations requiring easily administered and economical
assessment tools, such as in large-scale studies or in community situations.
Back to top
Introduction
Any effort to improve public health in the United States must consider the diversity of the U.S. population and the varying disease incidence among ethnic groups. Among the ethnic groups in the United States, the Hispanic population is the most rapidly growing. Hispanics constitute 32.4% of the population in California and 13.0% of the U.S. population overall. Currently, 37.4
million Hispanics are living in the United States — one in eight people (1). By the year 2050, it is estimated that the Hispanic population will increase to 102 million, or 24.5% of the U.S. population (2), one of the major changes that is occurring in the composition of the U.S. population. The population shift has been accompanied by substantial health disparities. In terms of
age-adjusted years of potential life lost, Hispanics experience disproportionately more strokes, cervical cancer, perinatal conditions, and other conditions
(2). Conditions related to dietary habits have taken a particularly heavy toll.
Age-adjusted years of potential life lost due to diabetes is 41% greater among
Hispanics than among non-Hispanic whites; the rates of overweight and obesity
are 11% and 7% higher among Hispanic men and 26% and 32% higher among Hispanic
women than among non-Hispanic whites. Thus, accurate and easy ways to assess the
dietary habits of the rapidly growing Hispanic communities in the United States
are needed, as are appropriate strategies that promote healthy lifestyles and
prevent disease.
Although numerous studies have investigated the relationship between dietary intake and disease prevention, many of them were conducted with the
non-Hispanic population, so it is unclear whether the conclusions are generalizable to other ethnic groups. Many of the studies relied on dietary assessment instruments that are widely used and accepted in the
non-Hispanic population. However,
appropriate dietary assessment tools suitable for studying ethnic minority communities in the United States are lacking. Developing and validating such research-appropriate tools are key steps toward addressing health issues of the Hispanic population. In addition, because of the limited education level of many Hispanics, brief screening tools are needed that can be administered in a few
minutes or self-administered in community settings.
We developed brief fat and fruit and vegetable intake screening tools (screeners) appropriate for a subpopulation of the Hispanic community and conducted a reliability study of the
screeners in a population of Mexican Americans in California. In addition, we used the instruments in community settings and for health education. The
screeners are easy to administer or can be
self-administered and take only a few minutes to complete. The scoring system is simple, and the participant receives immediate feedback on dietary intake of fat and fruit and vegetables.
Back to top
Methods
The food list for the two screeners was developed using previously described methods (3).
The list was created based on national data on the Mexican American population from the
Third National Health and Nutrition Examination Survey (NHANES III)
(4), which included an oversampling of the Mexican American population. For the fruit and vegetable screener, we
included the foods most commonly eaten. For the fat screener, we included the foods
making the largest contribution to grams-of-fat intake in this population, including foods contributing the top 60% of all fat intake, which has been the standard approach in developing Block screeners (3). This approach has produced good correlations with longer food lists and with reference data.
We also asked a simple question about vitamin supplement usage during the previous
week.
After creating a food list based on NHANES III, we conducted additional development and testing in three phases: interviews and focus groups, field testing, and a reliability study. For all three phases, participants were recruited from community-based organizations primarily serving the low-income
Hispanic population, including a community center, local health clinics, and adult education
classes (including English as a Second Language classes). The study was conducted from June through September 2000.
During the first phase, individual interviews and small group discussions were conducted to obtain information on cultural values, beliefs, and behaviors related to food, nutrition, and health and to refine the food list and instructions. Thirty-five individual interviews were conducted,
and a total of 70 men and women participated in the small group discussions. Discussions were conducted in
Spanish or English, depending on the preference of the participants. Participants were from
both rural and urban areas and were primarily Mexican and Mexican Americans. Participants completed the two screeners, and we obtained feedback about the screeners’ ease of administration (when self-administered), format, appropriateness of food wording, and usefulness as dietary assessment
instruments.
Field tests were conducted as a community service and were used as an opportunity to informally observe the screeners’ usability in real-world situations. Several hundred people participated at health fairs and other community gatherings. The screeners were administered and scored immediately, and feedback about participants’ fat and fruit and vegetable intakes was provided.
Administration of the screener took 5 to 10 minutes. Participants were asked for feedback about their satisfaction with the screener and its results as well as its ease of use.
For the reliability study, participants were recruited from three community-based organizations, including an organization providing referral services, adult education classes, and two health clinics (one urban and one semirural). Interviewers administered the two screeners (fat intake and fruit and vegetable intake) twice, 1 month apart.
We examined results using two different scoring systems, simple and
times per week. The simple method scored the frequency responses as 0 to 4
across the five categories of the fat screener or 0 to 5 across the six
categories of the fruit and vegetable screener. Thus, for the simple fat score, each food was given a score from 0 to 4, and then the scores
were summed for all the 16 fat screener foods. The same procedure was followed
for the fruit and vegetable screener. The advantage of the simple (0 to 4)
scoring system is that it only involves adding one-digit numbers and is
therefore easier to self-score. The times-per-week scoring system used the reported frequency of each category to assign the score. For example, 2 to 3 times per week was assigned a score of 2.5. The
times-per-week scores for each food were
summed using all foods
in the fat or fruit and vegetable screeners, resulting in an estimate of
times per week for each screener. This scoring system is not appropriate for
self-scoring but does permit an estimate of servings per week.
Descriptive statistical analyses were performed. For the reliability analysis, agreement between the two administrations was assessed using Pearson correlation coefficients. Both scores were approximately normally distributed and were not transformed. To estimate the meaning of the simple score in relation to recommended levels of intake, we also conducted a regression analysis of the
relationship between the simple score and estimated times per day.
The University of California Committee for the Protection of Human Subjects approved the research, and participants signed informed consent forms.
Back to top
Results
Interviews and field testing
Based on the individual interviews and small group discussions, only minor modifications were made to the original food list. The original
food list derived from the national data was found to represent the foods important to the study population. Formatting changes were made and words were added
or modified to clarify certain food items. Overall, the participants reported that the screeners were easy to complete and
that the process helped them evaluate their diets.
The fat screener had 16 items, including eggs, whole milk, flour tortillas, hamburgers, tacos and burritos, other mixed dishes with meat, pork, fried chicken, cheese, pizza, refried beans, French
fries, chips, cake, fat in cooking, and salad dressing. The fat screener had five frequency response categories, from once a month or less to five or more times per week.
The fruit and vegetable screener had 7 items (Appendices), including fruit
juice, other fruit, green salad, tomatoes or salsa, potatoes, soups or stews
with vegetables, and any other vegetables. The fruit and vegetable screener had
six frequency response categories, from less than once per week to two or more
times per day. The frequency categories are the same as those used in the
English versions of Block screening questionnaires (3,5). Portion sizes were not
assessed.
For the interviews and field testing of the screeners, almost 300 participants were included, of whom 49% were
men. The majority of the respondents were married. Age distribution was as follows: 38%
were younger than 30 years of age, 47% were between 30 and 49 years of age,
and 15% were 50 years or older. Eighty-five percent of the participants
completed the Spanish-language version, and 15% chose to complete the English
version. Almost 42% had an eighth-grade education or less, and 21% had completed
high school (data not shown).
Reliability study
For the reliability study, 93 people completed the two brief questionnaires twice,
4 weeks apart. Men comprised 42% of the sample, and 91% of participants had been born in Mexico (Table 1).
To estimate the average frequency of consumption, we used the times-per-week scoring system. The average consumption frequency of fruits and vegetables was 4.2 times per day at the first visit (Table 2). On the first administration of the screener, 32%
of participants scored at 5 or more times per day, and 30% scored at fewer than 3 times per day (data not shown). On average, fat sources were consumed
3.8 times per day. Twenty-four percent of participants reported eating fat sources 5 or more times per day, and 14% reported eating them fewer than twice a day (data not shown). Men and women both reported a lower intake of fruits, vegetables, and fat on the second administration.
During the first administration of the screener, 44% of participants reported
consuming vitamin supplements at least once per week. Women reported similar
usage during the first and second screening. In contrast, 42% of men in the
first screening reported consuming vitamins at least once per week, whereas only 30% of men
participating in the second screening reported consuming vitamins at least once
per week. However, this difference between men and women may not be meaningful because of the small sample size; only 5 of the 39 men gave a different answer on the first and second
screenings.
The reliability correlations showed no important differences between using the simple and
times-per-week scoring systems, so results for the reliability analysis are based on the simple scoring system. The simple scoring system was a continuous variable ranging from 1 to 28
in this sample. The reliability correlation was r = 0.64 for the fruit and vegetable screener and r = 0.85 for the fat screener (Table 3). The
slightly lower correlations for fruits and vegetables could have been caused by variations in their availability because the data collection spanned the months from June through September. Adjustments for age and sex had little effect on the correlation coefficients. The screeners also asked about use of vitamin supplements “at least once a week.” Reproducibility of this question was
also high; 84% were classified in the same way both times (P < .001)
(Table 3).
The two scoring systems were highly correlated (r = 0.99 for the fat scores in the first administration). Regression of the first-administration
times-per-day score on the simple score yielded the following equations:
Fat Times/Day = 0.16 × Fatsimple − 0.48
FV Times/Day = 0.34 × FVsimple − 1.34,
where FV indicates fruit and vegetable; Fatsimple, simple fat score; and FVsimple,
simple fruit and vegetable score. Using these equations, we created usable cutoff points from the simple scoring systems. For fruits and vegetables, the simple scores were broken down as follows:
- ≥18 ≈ 5 per day, Excellent
- 16–17 ≈ 4 per day, Good
- 13–15 ≈ 3 per day, Fair
- <13 ≈ 2 per day, Poor
In the first administration of the first screener, 46% had fruit and
vegetable scores of excellent; 12%, good; 14%, fair; and 28%, poor. In the second administration, scores were 28%, excellent; 15%, good; 20%, fair; and 28%, poor — a more likely distribution.
For fats, the cutoff points were more arbitrary. We used simple quartiles
based on the distribution from the second administration. This approach may be better than estimating percentage of energy from fat, because a low fat percentage may simply be the result of an overall high energy intake or a high energy intake from soft drinks or alcohol. Following is the score breakdown for the fat
screener:
- ≤18 = Excellent
- 19–24 = Good
- 25–33 = Fair
- >33 = Poor
Back to top
Discussion
Our analyses have shown that the two brief dietary screeners for Mexican Americans
have good reliability. We also tested the screeners in community situations such
as health fairs, community centers, and clinics and found them to be well-received and to provide useful information for the participants.
Many brief dietary assessment tools have been developed over the past few decades,
most of which have been simplified food frequency questionnaires
(FFQs) or questionnaires focusing on eating behaviors. Several were developed to assess dietary fat intake or percentage of energy from fat (3,6-8). A review of validation studies of fat screeners was conducted by Yaroch et al (9). Our correlation of r =
0.85 (Table 3) is comparable to the correlation of r = 0.87 found among white middle-class
participants for the Kristal et al Food Habits Questionnaire (10) and the Connor et al Diet Habit Survey (11).
Brief tools assessing fruit and vegetable intake have also been developed, evaluated, and reviewed (12-15). Our results of r = 0.64 for the fruit and vegetable screener and r = 0.85 for the fat sources screener are similar to the r = 0.62 for a composite index of diet atherogenicity based on the 17 food items, found by Shea et al for the 17-item dietary component of the Behavioral Risk Factor
Surveillance System conducted among white, African American, and Hispanic respondents (16). Smith-Warner et al (17) reported reliability correlations of r = 0.55 and higher.
Few brief screeners have been used with substantial numbers of Mexican Americans, and none (of which we are aware) were systematically developed for this population based on national data. The few Spanish-language screeners and brief tools that have been used with
large numbers of Hispanic
adults include 1) the National Cancer Institute’s 5 A Day for Better Health fruit and vegetable screener
(18) used in the Centers for Disease Control and Prevention state and local Behavioral Risk Factor Surveillance System (BRFSS) (19) and the California Health Interview Survey (CHIS) (20); 2) a modification of the Food Habits Questionnaire originally developed by Kristal et al (10); and 3) the fruit, vegetable, and fat questions of the Food Behavior Checklist (21-24).
Some authors have investigated the reliability of longer instruments in assessing the dietary intake of Hispanics
in the United States. Taren et al (25) studied the reliability of the 159-item Southwest FFQ, a modification of the Block FFQ, among 79 Mexican American and 80 non-Hispanic participants. The second
FFQ was administered
after 2 weeks. The average reliability coefficient for energy and fats was r = 0.81, similar to our r= 0.85 for the fat screener. The average reliability coefficient for vitamin C, folate, beta-carotene, and lycopene was r = 0.72, similar to our fruit and vegetable screener correlation of r = 0.64. Mayer-Davis et al studied the reliability of the Insulin Resistance Atherosclerosis Study (IRAS)
FFQ (also a modification of the Block questionnaire) among 43 rural Hispanic women after a 2- to 4-year interval (26). Reliability coefficients for energy and fats ranged from r = 0.54 to r = 0.64, and for vitamin C the correlation was r = 0.59. Cullen et al studied the reliability of the Youth/Adolescent Questionnaire among 41 Hispanic seventh- and eighth-grade students
after a 3-week interval (27). Questions were read to the students in class. The reliability coefficient for energy was r = 0.61 and for servings of fruits, vegetables, and juices was r = 0.68.
For some research purposes, full-length dietary questionnaires are preferable because they provide
estimates for a range of nutrients. However, some situations call for
self-administration, but among populations with lower education levels (such as
the Mexican American sample in this study), self-administration of long questionnaires
is problematic. In a multiethnic validation study of self-administered full-length questionnaires among low-income
participants in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) (28), researchers found that validity was acceptable among whites and African Americans but not among Hispanics. In addition, in some research situations, such as interventions to increase fruit and vegetable intake or decrease fat intake, a brief,
targeted instrument may be all that is needed.
For interventions designed to improve dietary habits, it is necessary to know the sensitivity of the instrument in detecting change. The sensitivity can be calculated from the data presented here. The following formula estimates the sample size needed to detect change within a single group:
N = ([Zα/2 + Zβ]2
× σ2d) / Δ2
where Zα/2 indicates the upper α/2
percent point of the normal distribution; Zβ, the lower β
percent point of the normal distribution; σ2d, the variance of the differences in the before and after estimates, which in this analysis was 2.407 for fruits and vegetables per day and 0.959 for fat sources per day;
and
Δ, the size of the change you would like to achieve.
For example, to detect a change of 0.5 times per day in fat sources and fruit and vegetables sources,
with a two-sided alpha and 90% power, would require sample sizes of 41 (for fat) and 102 (for fruit and vegetables). The sample sizes needed to detect the difference between the changes in two groups (e.g., intervention and control) can be calculated as follows:
Neach = ([Zα/2 + Zβ]2
× 2 ×
σ2d) /
Δd2 ,
where Zα/2 indicates the upper α/2
percent point of the normal distribution; Zβ, the lower β
percent point of the normal distribution; σ2d, the variance of the differences in the before and after estimates;
Δd, the difference between the two change scores.
For example, if the control group increases by 0.25 times per day and the intervention group increases by 0.75 times per day, the
Δd is 0.50. The study would need 81 and 203 people in each
group to detect that degree of difference in the changes in the two groups for fat intake and fruit and vegetable intake, respectively.
The reason for the lower intake estimates for fat sources and fruits and vegetables in the second administration
of the screeners is unclear. However, this pattern has been observed repeatedly (although not exclusively) in dietary data, with
both FFQs and 24-hour recalls (27). Given this pattern of respondent behavior, it would be prudent to administer a baseline questionnaire twice and use
the second administration as the estimate of before in an intervention study.
Our data comparing fruit, vegetable, and fat intakes and supplement use in
two questionnaire administrations to 93 individuals demonstrate good
reliability. The fact that the sample included both men and women and comprised
a population whose primary language was Spanish suggests broad usefulness among
U.S. Hispanic Americans, of whom 66.9% are of Mexican origin (29). In addition, most of the food names used
in the screeners are commonly used among all U.S. Hispanic populations. However, the cultural diversity in the
U.S. Hispanic population suggests the desirability of testing the instrument
among Cuban, Puerto Rican, and Central American subgroups.
The suggested cutoff points for the simple scoring system may be used to give immediate feedback to respondents when the screeners are scored by the respondents themselves or by an assistant. The resulting estimated fruits and vegetables times per day from the
time-per-week system and the corresponding simple system may be interpreted to approximately equal servings per day, because one piece of
fruit equals one serving, and most vegetables are not eaten more than once per day. The times-per-day estimates of 4.2 at the first administration and 3.6 at the second administration correspond reasonably well with estimates of 3.8 servings of fruits and vegetables
per day among Hispanic adults in a 1999 study in California (30).
For the fat-intake screener, either scoring system translates to times per day, and the foods often have no recommended serving size or frequency. Furthermore, the
fat-intake scores are based on fat grams, whereas recommendations are based on percent of calories. However, during the development of the original English-language screener
for fat intake, it was shown that the screener score based on fat grams correlated at r
= 0.54 with percentage of calories from fat as estimated by the mean of three 4-day records (3). Thus, the fat-intake score (either simple or
times per week) may be used as a continuous variable to rank participants’ fat intake.
In addition to reliability, another key feature of the brief screeners is ease of administration. Although screeners are not substitutes for more comprehensive dietary assessment instruments that assess the entire diet,
they fill the practical need for easily administered, economical, and less
burdensome tools. These features are particularly relevant
to the practical use of these tools among larger populations and in community
settings. Our preliminary results with community-based populations show the screeners to be well-received. Screeners can identify individuals at high risk
of disease due to too-frequent consumption of high-fat foods or infrequent consumption of
fruits and vegetables. Because they are simple, easy to administer, and easy to
understand,
screeners can be useful in community settings where precise or accurate
estimates are unnecessary. The screeners themselves can provide content for nutrition
education programs. With their immediate feedback on dietary intake, the screeners
can
stimulate respondents’ interest, which may facilitate dietary change. When
resources are limited, or among
subpopulations such as minority populations or those with limited English language proficiency, brief screeners can be useful instruments for dietary assessment, raise awareness of individuals’ food intakes, and heighten interest in and motivation for making changes.
Back to top
Author Information
Corresponding Author: Patricia Wakimoto, National Center for
Minority Health and Health Disparities, Children’s Hospital Oakland
Research Institute, 5700 Martin Luther King Jr. Way, Oakland, CA 94609.
Telephone: 510-428-3885, ext. 5463. E-mail: pwakimoto@chori.org.
Author Affiliations: Gladys Block, Shelly Mandel, Norma Medina, University of California, Berkeley, School of Public Health, Division of Community Health and Human Development, Berkeley, Calif.
Back to top
References
- U.S. Census Bureau. United States census. Washington (DC): U.S. Department of Commerce, Economics and
Statistics Administration, U.S. Census Bureau; 2000. Available from: URL: http://www.census.gov/main/www/cen2000.html.
- Centers for Disease Control and Prevention (CDC).
Health disparities experienced by Hispanics — United States. MMWR Morb Mortal
Wkly Rep 2004;53(40):935-7.
- Block G, Clifford C, Naughton MD, Henderson M, McAdams M. A brief dietary
screen for high fat intake. J Nutr Educ 1989;21(5):199-207.
- U.S. Department of Health and Human Services. National Center for Health
Statistics. Third National Health and Nutrition Examination Survey, 1988–1994.
NHANES III examination data file (CD-ROM). Hyattsville (MD): Centers for Disease
Control and Prevention, 1996.
- Block Dietary Data Systems, www.nutritionquest.com.*
- Ammerman AS, Haines PS, DeVellis RF, Strogatz DS, Keyserling TC, Simpson RJ Jr, et al.
A brief dietary assessment to guide cholesterol reduction in low-income individuals: design and validation. J Am Diet Assoc 1991;91(11):1385-90.
- Coates RJ, Serdula MK, Byers T, Mokdad A, Jewell S, Leonard SB, et al.
A brief, telephone-administered food frequency questionnaire can be useful for surveillance of dietary fat intakes. J Nutr 1995;125(6):1473-83.
- Thompson FE, Midthune D, Subar AF, Kahle LL, Schatzkin A, Kipnis V.
Performance of a short tool to assess dietary intakes of fruits and vegetables, percentage energy from fat and fibre. Public Health Nutr 2004;7(8):1097-105.
- Yaroch AL, Resnicow K, Khan LK.
Validity and reliability of qualitative dietary fat index questionnaires: a review. J Am Diet Assoc 2000;100(2):240-4.
- Kristal AR, Shattuck AL, Henry HJ.
Patterns of dietary behavior associated with selecting diets low in fat: reliability and validity of a behavioral approach to dietary assessment. J Am Diet Assoc 1990;90(2):214-20.
- Connor SL, Gustafson JR, Sexton G, Becker N, Artaud-Wild S, Connor WE.
The Diet Habit Survey: a new method of dietary assessment that relates to plasma cholesterol changes. J Am Diet Assoc 1992;92(1):41-7.
- Serdula M, Coates R, Byers T, Mokdad A, Jewell S, Chavez N, et al.
Evaluation of a brief telephone questionnaire to estimate fruit and vegetable consumption in diverse study populations. Epidemiology 1993;4(5):455-63.
- Block G, Gillespie C, Rosenbaum EH, Jenson C.
A rapid food screener to assess fat and fruit and vegetable intake. Am J Prev Med 2000;18(4):284-8.
- Thompson FE, Subar AF, Smith AF, Midthune D, Radimer KL, Kahle LL, et al.
Fruit and vegetable assessment: performance of 2 new short instruments and a food frequency questionnaire. J Am Diet Assoc 2002;102(12):1764-72.
- Kim DJ, Holowaty EJ.
Brief, validated survey instruments for the measurement of fruit and vegetable intakes in adults: a review.
[Published erratum in: Prev Med 2003;37(1):72]. Prev Med 2003;36(4):440-7.
- Shea S, Melnik TA, Stein AD, Zansky SM, Maylahn C, Basch CE.
Age, sex, educational attainment, and race/ethnicity in relation to consumption of specific foods contributing to the atherogenic potential of diet. Prev Med 1993;22(2):203-18.
- Smith-Warner SA, Elmer PJ, Fosdick L, Tharp TM, Randall B.
Reliability and comparability of three dietary assessment methods for estimating fruit and vegetable intakes. Epidemiology 1997;8(2):196-201.
- Thompson B, Demark-Wahnefried W, Taylor G, McClelland JW, Stables G, Havas S, et al.
Baseline fruit and vegetable intake among adults in seven 5 a Day study centers located in diverse geographic areas. J Am Diet Assoc 1999;99(10):1241-8.
- Centers for Disease Control and Prevention (CDC). Behavioral Risk Factor
Surveillance System survey data. Atlanta (GA): U.S. Department of Health and
Human Services, Centers for Disease Control and Prevention; 1998. Available
from: URL: http://www.cdc.gov/brfss/technical_infodata/ surveydata/1998.htm.
- California Health Interview Survey adult questionnaire. Los Angeles:
University of California, UCLA Center for Health Policy Research; 2001.
Available from: URL: http://www.chis.ucla.edu/pdf/CHIS2001_adult_q.pdf.*
- Neuhouser ML, Thompson B, Coronado GD, Solomon CC.
Higher fat intake and lower fruit and vegetables intakes are associated with greater acculturation among Mexicans living in Washington State. J Am Diet Assoc 2004;104(1):51-7.
- Gans KM, Burkholder GJ, Risica PM, Lasater TM.
Baseline fat-related dietary behaviors of white, Hispanic, and black participants in a cholesterol screening and education project in New England. J Am Diet Assoc 2003;103(6):699-706; discussion 706.
- Murphy SP, Kaiser LL, Townsend MS, Allen LH.
Evaluation of validity of items for a food behavior checklist. J Am Diet Assoc 2001;101(7):751-61.
- Hartline-Grafton H, Nyman R, Briefel R, Cohen R. Prototype notebook: short
questions on dietary intake, knowledge, attitudes and behaviors. Washington
(DC): Mathematica Policy Research, Inc; 2004. Available from: URL:
http://www.mathematica-mpr.com/publications/PDFs/protonote.pdf.*
- Taren D, de Tobar M, Ritenbaugh C, Graver E, Whitacre R, Aickin M. Evaluation of the southwest food frequency questionnaire.
Ecol Food Nutr 2000;38(6):515-547.
- Mayer-Davis EJ, Vitolins MZ, Carmichael SL, Hemphill S, Tsaroucha G, Rushing J,
et al.
Validity and reproducibility of a food frequency interview in a multi-cultural epidemiologic study. Ann Epidemiol 1999;9(5):314-24.
- Cullen KW, Zakeri I.
The Youth/Adolescent Questionnaire has low validity and modest reliability among low-income African-American and Hispanic seventh- and eighth-grade youth. J Am Diet Assoc 2004;104(9):1415-9.
- Block G, DiSogra C. WIC dietary assessment validation study. Final report.
Alexandria (VA): U.S. Department of Agriculture, Food and Nutrition Service, 1995.
Available from: URL: http://www.nutritionquest.com/research/WIC_Dietary_ Assessment_Validation_Study_Final_Report.zip.*
- U.S. Census Bureau. The Hispanic population in the United States: March
2002. Washington (DC): U.S. Department of Commerce, Economics and Statistics Administration, U.S.
Census Bureau; 2003. Available from: URL: http://www.census.gov/prod/2003pubs/p20-545.pdf.
- Oppen M, Sugerman S, Foerster SB. Fruit and vegetable consumption in California adults: Ten-year highlights from the California Dietary Practices Surveys 1989–1999.
Sacramento (CA): California Department of Health Services, Cancer Prevention and
Nutrition Section; 2002. Available from: URL: http://www.phi.org/pdf-library/fruit_survey1102.pdf.*
Back to top
|
|