Development of a Food Frequency Questionnaire: For Toddlers of Low-German-Speaking Mennonites from Mexico

Report

Rapport

40 Revue canadienne de la pratique et de la recherche en diététique – Vol 73 nº 1, printemps 2012

ABSTRACTPurpose: Little is known about dietary intakes in toddlers of Low-German-Speaking Mennonites from Mexico, although some of these toddlers might be at risk for nutritional deficien-cies. A 97-item, culturally sensitive, interviewer-administered food frequency questionnaire (FFQ) was developed and vali-dated for health professionals to assess dietary intake in these children aged 12 to 36 months.

Methods: Cultural foods on the FFQ were determined via focus groups; a pilot study tested content and formatting. The FFQ was administered to parents/caregivers of 22 toddlers in a southern Ontario community of Low-German-Speaking Mennonites from Mexico. Validity was determined by compar-ing nutrient intakes from the FFQ and from the 24-hour recalls, using Bland-Altman plots, Pearson correlations, and Student’s t-tests. Test-retest reliability was compared between two FFQ administrations (n=14) one month apart, via intraclass correla-tions (ICCs).

Results: Bland-Altman plots showed good agreement between the FFQ and the 24-hour recall; Pearson correlations between methods were significant for protein, folate, calcium, and caf-feine. Student’s t-tests were not significantly different between methods for 11 of 12 nutrients. Test-retest reliability was good on the basis of acceptable ICC for eight of 12 nutrients.

Conclusions: The prevalence of nutrient inadequacies was low, except for folate. These results are promising for implementa-tion of a simple, quick, culturally sensitive FFQ with the potential to provide reliable estimates of mean intakes in toddlers of Low-German-Speaking Mennonites from Mexico.

(Can J Diet Pract Res. 2012;73:40-44) (DOI: 10.3148/73.1.2012.40)

RéSuméObjectif. On sait très peu de choses sur les apports alimen-taires des jeunes enfants mennonites originaires du Mexique parlant le bas-allemand, mais certains d’entre eux pourraient être à risque de carences nutritionnelles. Un questionnaire de fréquence alimentaire (QFA) de 97 questions adapté à la réalité culturelle et administré par les intervieweurs a été mis au point et validé afin que les professionnels de la santé puissent évalu-er l’apport alimentaire de ces enfants âgés de 12 à 36 mois.

Méthodes. Des groupes de discussion ont été utilisés pour choisir les aliments culturels inclus dans le QFA; une étude pilote a été mise sur pied pour tester le contenu et le format. Le QFA a été administré aux parents/personnes qui pren-nent soin de 22 jeunes enfants d’une communauté du sud de l’Ontario formée de mennonites originaires du Mexique parlant le bas-allemand. La validité a été établie en comparant les apports en nutriments du QFA et ceux des rappels alimentaires de 24 heures à l’aide de graphiques de Bland-Altman, des coef-ficients de corrélation de Pearson et des tests t. La fiabilité de la méthode test-retest a été comparée, par corrélations intraclas-ses (CIC), entre deux administrations du QFA (n = 14) effectuées à un mois d’intervalle.

Résultats. Les graphiques de Bland-Altman montraient une bonne cohérence entre le QFA et le rappel alimentaire de 24 heures; les coefficients de corrélation de Pearson entre les deux méthodes étaient significatifs pour les protéines, le folate, le calcium et la caféine. Les tests t n’étaient pas significative-ment différents d’une méthode à l’autre pour 11 des 12 nutri-ments. La fiabilité de la méthode test-retest était bonne sur la base d’un CIC acceptable pour huit des 12 nutriments.

Conclusions. La prévalence de carences en nutriments était faible, sauf pour le folate. Ces résultats sont prometteurs pour l’implantation d’un QFA simple, rapide, adapté à la réalité culturelle et ayant le potentiel de fournir des estimations fiables des apports moyens chez les jeunes enfants mennonites origi-naires du Mexique parlant le bas-allemand.

(Rev can prat rech diétét. 2012;73:40-44)(DOI: 10.3148/73.1.2012.40)

Development of a Food Frequency Questionnaire For Toddlers of Low-German-Speaking Mennonites from Mexico

ANDREA D’AMBROSIO, MAN, RD, AMANDA TIESSEN, MSc, RD, JANIS RANDALL SIMPSON, PhD, RD, Department of Family Relations and Applied Nutrition, university of Guelph, Guelph, ON

Report

Rapport

41Canadian Journal of Dietetic Practice and Research – Vol 73 No 1, Spring 2012

Development of a Food Frequency Questionnaire For Toddlers of Low-German-Speaking Mennonites from Mexico

INTRODuCTIONLittle is known about dietary intakes in toddlers of Low- German-Speaking Mennonites from Mexico. These toddlers may be at risk for nutritional deficiencies because of food in-security, isolation, language/literacy barriers, lack of nutrition knowledge, and consumption of energy-dense staples (1-4). Approximately 57,000 Low-German-Speaking Mennonites from Mexico live in Canada; migration to rural communities in southern Ontario is increasing because of poor economic con-ditions in Mexico (5,6).

Public health registered dietitians (RDs) have expressed concerns about childhood nutrition in these communities. Because of challenges related to language and literacy, we determined, in consultation with public health RDs, that an interviewer-administered food frequency questionnaire (FFQ) would be ideal.

PuRPOSEThe primary objective was to develop, validate, and test the reliability of a culturally sensitive, semiquantitative FFQ for assessing mean dietary intakes in toddlers (children aged 12 to 36 months) of southern Ontario Low-German-Speaking Mennonites from Mexico. A secondary objective was to eval-uate nutrient intakes.

METHODSParticipants and proceduresThis study was approved by the research ethics board at the University of Guelph; informed written and/or verbal con-sent was obtained from mothers who were Low-German-Speaking Mennonites from Mexico, had healthy toddlers, and had immigrated to Canada within the preceding 10 years. Participants were contacted through local health agencies and church leaders; an incentive was $10 grocery vouchers. Demographic information on a subset was obtained with a questionnaire adapted from Statistics Canada (7). Data were collected by one of the authors (A.T.), a community member, or trained research assistants with a translator.

Food frequency questionnaireDevelopment: In winter and spring 2008, six focus groups of two to eight people each were conducted to determine commonly consumed cultural foods (8). The FFQ was based on the Children’s Nutrition Questionnaire (CNQ) (Harvard School of Public Health) (9), with a consideration of established methods (8,10-12). Portion sizes were estimated using measuring utensils and food models (8). The FFQ was pilot tested (n=8) before validation.

Validation and reliability: In 2008 and 2009, validity was determined by comparing the FFQ with multiple-pass 24-hour recalls (13), with repeated 24-hour recalls (two and four weeks later) for a subset (n=14). Fourteen subjects completed two FFQs, one month apart, which allowed assessment of test-retest reliability.

Data analysisNutrient intakes were calculated with The Food Processor (version 10.1, ESHA Research, Salem, OR, 2002), using 2007 Canadian Nutrient File data (14). The Nutrition Research Council method (15) was used to adjust intakes from the 24-hour recalls statistically for intraindividual variation, yielding usual intakes. Bland-Altman plots were used to assess agreement between the FFQ and the 24-hour recall (16). Pearson correlations were used to assess associations between methods; mean nutrient intakes (FFQ versus 24-hour recall) were compared with Student’s paired t-tests. The intraclass correlation coefficient (ICC) was used to evaluate FFQ reliability (17). SPSS, version 16.0 (SPSS Inc., Chicago, IL, 2004), was used for data analysis (statistical sig-nificance, p<0.05).

Prevalence of nutrient inadequacies in comparison with Dietary Reference Intakes was determined for the adjusted 24-hour nutrient intakes (18).

RESuLTSMothers’ average annual income was $15,000 to $29,999. Other demographic characteristics for the subset of moth-ers (n=14) appear in Table 1. Nutrient intakes are reported in Table 2. Bland-Altman plots showed good agreement be-tween the FFQ and the 24-hour recall for all nutrients. Plots for energy and selected nutrients are shown in Figures 1 to 4. No mean difference bias for Bland-Altman plots was detected for vitamin B12, calcium, fat, and carbohydrate, but such a bias was detected for energy, protein, iron, and fibre. Pearson correlations were significant for protein, folate, calcium, and caffeine (Table 2). With Student’s paired t-test, no significant differences were found between the two dietary assessment methods for any nutrients except caffeine (p<0.025) (data not shown). Good test-retest reliability was demonstrated, as ICCs were acceptable (>0.70) for most nutrients, with lower coefficients for energy (0.63), vitamin C (0.53), and folate (0.56). Data on the prevalence of nutrient inadequacies ap-pear in Table 2.

Table 1Demographic characteristics of a subset of mothers of

toddlers in southern Ontario Low-German-Speaking Mennonites from Mexico

Characteristic Mean (n=14) SD

Mothers’ age (years) 28 5

Children’s age (years) 2.17 0.55

Years of residence in Canada 5 4

Number of children 3.6 1.7

Years of education 5 1

SD = standard deviation

Report

Rapport


Figure 1Bland-Altman plot for 24-hour recall intake of caloric intake versus food frequency questionnaire (FFQ) for 22 toddlers

of Low-German-Speaking Mennonites from Mexico

Solid horizontal lines represent mean differences between methods and dashed lines represent the limits of agreement (mean ± 1.96 standard deviation).

Figure 3Bland-Altman plot for 24-hour recall intake of vitamin B12 intake versus food frequency questionnaire (FFQ) for 22

toddlers of Low-German-Speaking Mennonites from Mexico


Figure 2Bland-Altman plot for 24-hour recall intake of protein intake versus food frequency questionnaire (FFQ) for 22 toddlers

of Low-German-Speaking Mennonites from Mexico


Figure 4Bland-Altman plot for 24-hour recall intake of iron intake

versus food frequency questionnaire (FFQ) for 22 toddlers of Low-German-Speaking Mennonites from Mexico


24-h

our r

ecal

l cal

oric

inta

ke v

ersu

s FF

Q c

alor

ic in

take

(kca

l)24

-hou

r rec

all v

itam

in B

12 in

take

ver

sus

FFQ

vita

min

B12

in

take

(mg)

24-h

our r

ecal

l pro

tein

inta

ke v

ersu

s FF

Q p

rote

in in

take

(g)

24-h

our r

ecal

l iro

n in

take

ver

sus

FFQ

iron

inta

ke (m

g)

Mean of caloric intake estimates (kcal)

Mean of vitamin B12 intake estimates (mg)

Mean of protein intake estimates (g)

Mean of iron intake estimates (mg)

Report

Rapport

43Canadian Journal of Dietetic Practice and Research – Vol 73 No 1, Spring 2012

DISCuSSIONA great need exists for practical, valid, reliable, and cultur-ally sensitive measures of food intake at the population level (19,20), especially among toddlers of Low-German-Speaking Mennonites from Mexico, as virtually no dietary information is available for these children. This research provides findings on development, preliminary validation, and repeatability of a culturally sensitive FFQ with low respondent and adminis-trative burdens.

Numerous validation studies have demonstrated the valid-ity of FFQs (19,21-24). The current study results are similar to those reported in other validation studies among toddlers and school-aged children (19,21,22,24). For example, Pearson cor-relations are consistent with those reported for 151 five- to ten-year-old Brazilian children for energy, protein, carbohydrate, sodium, calcium, and folate intakes (21). In the current study, correlation coefficients for calcium were higher (r2=0.71) than they were for 18 six- to 10-year-old Italian children (r2=0.5) (22) and 224 four- to five-year-old New York children (r2=0.61) (24).

To the best of our knowledge, we are the first to report nu-trient intakes in toddlers of Low-German-Speaking Menno-nites from Mexico. Evaluation of nutrient intakes revealed re-sults similar to those of the 2004 Canadian Community Health

Survey, in which the prevalence of inadequacies was below 10% for all nutrients with an Estimated Average Requirement (25). The exception in our study was folate; the prevalence of folate inadequacy was 18%.

Study limitationsA major limitation is the small sample size (8) because of dif-ficulties recruiting participants. Other investigators have had similar difficulties (26). Nevertheless, validity and reliability were acceptable, likely because of modification of the validat-ed CNQ. Participants were recruited through public health programs, which limited generalizability. In addition, poor perception of portion sizes, language barriers, parental recall error, and social desirability bias may have affected results (6,13,19,21,27,28). The 24-hour recall was completed on the same day as the FFQ, which could have produced higher cor-relations because the caregiver was sensitized to the child’s diet by having completed the FFQ (24). The length of the FFQ (97 items) could have caused respondent fatigue, al-though other studies have included more items (8,11,18,22). Despite the limitations, this study has generated valuable pre-liminary findings on the use of this FFQ in a nutritionally vulnerable group.

Table 2Nutrient intakes and prevalence of nutritional inadequacy in toddlers of southern Ontario Low-German-Speaking

Mennonites from Mexico

Prevalence ofNutrient Nutrient intake inadequacya

24-hour recallb (n=22) FFQ (n=22) Pearson correlationsc

Mean ± SD Mean ± SD r2

Energy (kcal) 1330 ± 210 1458 ± 466 0.338 No EAR

Protein (g) 49 ± 5.5 50 ± 5 0.516d No EAR

Carbohydrate (g) 173 ± 38 196 ± 70 0.301 9%

Fat (g) 50 ± 8 47 ± 2 0.251 No EAR

Fibre (g) 9 ± 2 8 ± 3 0.377 No EAR

Vitamin B12 (mg) 3 ± 7 3 ± 1 0.346 0%

Vitamin C (mg) 55 ± 20 51 ± 29 0.192 0%

Folate (mg) 171 ± 36 169 ± 56 0.484d 18%

Calcium (mg) 797 ± 255 924 ± 407 0.741d 5%

Sodium (mg) 1500 ± 212 1295 ± 417 0.230 No EAR

Iron (mg) 7 ± 2 6 ± 3 0.303 9%

Caffeine (mg) 13 ± 14 25 ± 25 0.797d No EAR

EAR = Estimated Average Requirement; FFQ = food frequency questionnaire; SD = standard deviationa Prevalence of inadequacy calculated by the EAR cut-point method (18)b Adjusted nutrient intake value using Nutrition Research Council method (15)c For comparison of FFQ and 24-hour recalld p<0.02

Report

Rapport


RELEVANCE TO PRACTICEThis FFQ has potential uses in needs assessments, evaluations of community outreach programs, and policy interventions aimed at improving dietary intakes for toddlers of Low- German-Speaking Mennonites from Mexico. More research is needed to validate this FFQ further. Educational initiatives should focus on increasing dietary folate in the diets of this group of toddlers.

AcknowledgementsWe thank the mothers of the southern Ontario Low- German-Speaking Mennonites from Mexico who partici-pated in the study, the Canadian Foundation for Dietetic Research, the Perth District Health Unit, and the Woolwich Community Health Centre.

References1. Martens HM. Mennonites from Mexico: immigrations and settlement in

Canada: a research report. Waterloo, ON: Conrad Grebel University College; 1975.

2. Sider SR. Experiences with language retention: case studies of Mennonites from Mexico. Ann Arbor, MI: ProQuest; 2006.

3. Perth District Health Unit. Descriptive profile of Amish and Mennonite com-munities in Perth County. Stratford, ON: Perth District Health Unit; 2006.

4. Tarasuk V. Discussion paper on household and individual food security. Health Canada; 2001 [cited 2011 Jan 26]. Available from: http://www.hc-sc.gc.ca/fn-an/nutrition/pol/food_sec_entire-sec_aliments_entier-02-eng.php

5. Kulig JC, Wall M, Hill S, Babcock R. Childbearing beliefs among Low- German-Speaking Mennonite women. Int Nurs Rev. 2008;55(4):420-6.

6. Hall BL, Kulig JC. Kanadier Mennonites: a case study examining research challenges among religious groups. Qual Health Res. 2004;14(3):359-68.

7. Statistics Canada. Canadian Community Health Survey (CCHS) question-naire for Cycle 1.1 (September 2000-November 2001). Ottawa: Statistics Canada; 2001.

8. Cade J, Thompson R, Burley V, Warm D. Development, validation and uti-lization of food-frequency questionnaires—a review. Public Health Nutr. 2002;5:567-87.

9. Blum RE, Wei EK, Rockett HRH, Leppert J, Langiers J, Gardner JD, et al. Vali-dation of a food frequency questionnaire in Native American and Caucasian children 1 to 5 years of age. Matern Child Health J. 1999;3:167-72.

10. Teufel N. Development of culturally competent food-frequency question-naires. Am J Clin Nutr. 1997;65(Suppl):1173S-8S.

11. Williams PL, Innis SM. Food frequency questionnaire for assessing infant iron nutrition. Can J Diet Pract Res. 2005;66:176-82.

12. Heath ALM, Skeaff CM, Gibson RS. The relative validity of a computerized food frequency questionnaire for estimating intake of dietary iron and its absorption modifiers. Eur J Clin Nutr. 2000;54:592-9.

13. Gibson RS. Principles of nutritional assessment. 2nd ed. New York: Oxford

University Press; 2005.

14. Health Canada. Canadian Nutrient File (CNF), 2007b; 2007 [cited 2010 Feb

7]. Available from: http://www.hc-sc.gc.ca/fn-an/nutrition/fiche-nutri-data/

cnf_downloads-telechargement_fcen-eng.php

15. Food and Nutrition Board, Institute of Medicine. Dietary Reference Intakes:

applications in dietary planning. Appendix E: adjustment of observed intake

data to estimate the distribution of usual intakes in a group. Washington:

National Academy Press; 2003. p. 196-208.

16. Bland JM, Altman DG. Statistical methods for assessing agreement between

two methods of clinical measurement. Lancet. 1986;1:307-10.

17. Landis JR, Koch GG. The measurement of observer agreement for categorical

data. Biometrics. 1977;33:159-74.

18. Food and Nutrition Board, Institute of Medicine. Dietary Reference Intakes:

applications in dietary assessment. Washington: National Academy Press;

2000.

19. Sudha V, Radhika G, Sathya RM, Ganesan A, Mohan V. Reproducibility and

validity of an interviewer-administered semi-quantitative food frequency

questionnaire to assess dietary intake of urban adults in southern India. Int J

Food Sci Nutr. 2006;57(7-8):481-93.

20. Hanning RM, Royall D, Toews JE, Blashill L, Wegner J, Driezen P. Web-based

food behaviour questionnaire: validation with grades six to eight students.

Can J Diet Pract Res. 2009;70(4):172-8.

21. Fumagalli F, Pontes Monteiro J, Sartorelli DS, Vieira MN, de Lourdes Pires

Bianchi M. Validation of a food frequency questionnaire for assessing dietary

nutrients in Brazilian children 5 to 10 years of age. Nutrition. 2008;24(5):427-

32.

22. Bertoli S, Petroni ML, Pagliato E, Mora S, Weber G, Chiumello G, et al. Vali-

dation of food frequency questionnaire for assessing dietary macronutrients

and calcium intake in Italian children and adolescents. J Pediatr Gastroen-

terol Nutr. 2005;40(5):555-60.

23. Segovia-Siapco G, Singh P, Jaceldo-Siegl K, Sabaté J. Validation of a food-

frequency questionnaire for measurement of nutrient intake in a dietary in-

tervention study. Public Health Nutr. 2007;10(2):177-84.

24. Stein AD, Shea S, Basch CE, Contento IR, Zybert P. Consistency of the Wil-

lett semiquantitative food frequency questionnaire and 24-hour dietary re-

calls in estimating nutrient intakes of preschool children. Am J Epidemiol.

1992;135(6):667-77.

25. Kirkpatrick SI, Tarasuk V. Food insecurity is associated with nutrient inad-

equacies among Canadian adults and adolescents. J Nutr 2008;138:604-12.

26. Rowden PM, Paschal AM, Hawley SR, Hsiao T. Oral health attitudes and

practices among a German Mexican Mennonite farmworker community.

J Immigr Minor Health. 2011;13:1159-67.

27. Livingstone MBE, Robson PJ, Wallace JMW. Issues in dietary intake assess-

ment of children and adolescents. Br J Nutr. 2004;92:S213-22.

28. Treiber FA, Leonard SB, Frank G, Musante L, Davis H, Strong WB, et al. Di-

etary assessment instruments for preschool children: reliability of parental

responses to the 24-hour recall and a food frequency questionnaire. J Am

Diet Assoc. 1990;90:814-20.

Copyright of Canadian Journal of Dietetic Practice & Research is the property of Dietitians of Canada and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.

Documents

Development of a Food Frequency Questionnaire: For Toddlers of Low-German-Speaking Mennonites from Mexico