«Minimum Dietary Diversity for Women A Guide to Measurement Minimum Dietary Diversity for Women A Guide to Measurement Published by the Food and ...»
When the MDD-W is included in large-scale, multi-topic surveys, it might not be possible to select enumerators with this range of knowledge and survey experience. In any case, it is recommended that enumerators have some post-high school education and experience in survey methodology and interviewing.
TOPICS TO COVER DURING TRAININGThis section highlights a few main points and issues unique to training enumerators for collecting food group diversity data using the recall method21.
Training should include classroom instruction, discussion and field practice. Once the questionnaire rows with locally available foods are reviewed, a fair amount of time should be allotted to discussions, as the trainees need to be familiar with the foods (including commonly consumed mixed dishes) and their classification into rows in order to correctly record data on the MDD-W questionnaire.
The following scheme provides some points to consider when designing enumerator training for the MDD-W.
A. Introduction to and meaning of dietary diversity and the MDD-W • Discuss the objectives of the questionnaire, i.e. to gain information on the foods and food groups consumed by the woman respondent the previous day and night. Explain that healthy diets are diverse and include many foods and food groups. Explain and show that the questionnaire organizes foods into groups by showing food items in rows of similar foods.
• Explain that the final output from the questionnaire is a count of food groups and explain that there are some rows that “count” and others that do not “count”, either because the foods are not nutritious or because people usually consume very small amounts.
Many resources are available that cover enumerator training more generally. See, for example, the Demographic 21 and Health Surveys manual for training field staff.
B. Description of questionnaire rows and exercises in classifying foods • Review the groupings listed on the adapted questionnaire (MDD-W groups and other categories) and clarify any questions regarding why items are placed in the various rows.
Special issues may need to be discussed and specific guidance given, such as classifying beverages, condiments and seasonings, etc. (see Appendix 2 for a list of foods that are difficult to classify).
• In an exercise, have enumerators sort/classify foods into the appropriate rows of the adapted questionnaire (e.g. using a stack of food cards or photographs).
C. Introduction to the open recall method and to recording information on the MDD-W questionnaire • Explain the principles behind the open recall method (to obtain a report of all foods and drinks consumed by the respondent during the day and night at meals, between meals and during food preparation, and consumed both in the home and outside the home).
• Explain the time period of the recall – from the time the respondent woke the previous day through the day and overnight. Explain that the aim is to gather information about a 24-hour period.
• Explain the concept of mixed dishes.
• Discuss how to probe about mixed dishes and where to place ingredients in the questionnaire rows. If available, introduce the guidance sheets on common mixed dishes developed during the adaptation phase and practice using them.
• Explain that certain foods are classified in only one row even if they have several ingredients (e.g. bread).
• Review the “Condiments and seasonings” list developed during the adaptation phase and (if available) explain how to use the guidance sheets to aid in correctly classifying these items.
• Explain and demonstrate an open recall and show how enumerators should record the information on the questionnaire when using a printed questionnaire or tablet.
D. Practice in carrying out the open recall • In an exercise, have trainees practice/role play in pairs and then select several pairs to practice in front of the group. Ask other trainees to comment on the role play; follow with corrections as needed.
• Review the questionnaires marked during the role plays, correcting errors as needed.
• Throughout training, allow sufficient time for questions and comments from enumerator trainees, as this may indicate the need to modify some parts of the questionnaire, enumerator instructions or guidance sheets to improve ease of administration and clarity for the respondent.
E. Final adjustment of questionnaire prior to data collection • Follow standard practices established by the survey organizers for field testing and field practice by trainees; standard practice will vary by the type and scale of the survey in which the MDD-W questionnaire is embedded.
• Following training, field testing and piloting, revise the questionnaire and enumerator instructions as needed and review all changes with enumerator trainees.
34 Section 5 Selection and training of enumerators
Normal practice for large-scale, multi-topic surveys is to conduct thorough enumerator training that may last 2 weeks or more, with pilot testing of the entire survey (all survey modules) prior to starting data collection. When the MDD-W is included, it would be ideal, although not always realistic, to schedule up to 2 days on the MDD-W into the overall enumerator training schedule. If the enumerators are nutritionists or have experience with nutrition questionnaires, initial training could be accomplished in 1 day. In all cases, enumerators can continue to learn during survey implementation through direct feedback from supervisors, debriefing and group discussions during field staff meetings and, in the case of extended periods of data collection, through periodic retraining.
3536Section 6. Tabulation, presentation and interpretation
Constructing the MDD-W indicator To construct the MDD-W indicator, the first step is to combine (aggregate) questionnaire rows (food groups and subfood groups) into the 10 MDD-W food groups, as shown in Table 4.
Table 4. Aggregation to construct Minimum Dietary Diversity for Women of Reproductive Age (MDD-W)
For example, if a questionnaire is coded “1” for “yes” for either subgroup “A” or “B”, the woman receives a point for the first MDD-W group (“Grains, white roots and tubers, and plantains”). She does not receive an additional point if she consumed food items from both subgroups. The 10 MDD-W groups are first summed into a score ranging from 0 to 10. Each woman is then coded “yes” or “no” for scoring ≥ 5, followed by calculation of the proportion of women who score from 5 to 10.
The order of items/rows differs slightly on the list-based questionnaire in Appendix 3. When using the list-based 22 method, for example, it is better to have “Vitamin A-rich vegetables, roots and tubers” precede “White roots and tubers and plantains” to avoid misclassification of orange-/yellow-fleshed sweet potatoes, carrots, etc.
37 Minimum Dietary Diversity for Women A Guide to Measurement Section 6 Tabulation, presentation and interpretation Since many users may also calculate the infant and young child feeding (IYCF) indicator for MDD, Appendix 4 provides a table showing how the ten food groups in the MDD-W compare with the seven groups in the IYCF MDD indicator. See also Table 1 in Section 1 for a comparison of the indicators.
Presentation and interpretation of results Presentation can be as simple as the percent of WRA achieving MDD-W or “minimum dietary diversity”. The indicator was developed for exactly this purpose, i.e. when a single, simple, dichotomous indicator is needed.
The interpretation of the indicator is: “X% of women achieved minimum dietary diversity, and they are more likely to have higher (more adequate) micronutrient intakes than the X% of women who did not”.
In some cases, it may be useful to present results separately by selected geographic, socioeconomic or household characteristics (e.g. urban vs. rural households, by region, by wealth quintile or by level of education), but decisions on appropriate disaggregation will be survey- and context-specific and will depend on objectives, sampling and sample sizes. Example figures on the following pages present hypothetical data for urban and rural households.
While designed to meet the need for a single, simple indicator, the data collected to construct the indicator also provide a rich description of diet patterns. The information may also reflect specific food groups of interest in particular contexts (e.g. animal-source foods, fruits and vegetables, nutrient-poor and/or energy-dense groups and other specific food groups promoted in interventions).
The following figures are illustrative and are not an exhaustive set of presentation options.
The data used to generate the graphs are from two data sets, one urban and one rural, with a sample size of approximately 400 in each site23. Figures show percent and means (standard deviation), along with 95% confidence intervals.
FOOD GROUP DIVERSITYIn addition to presenting the percent of WRA achieving minimum dietary diversity, it can be useful to present the average (mean) diversity score and a histogram illustrating the distribution of scores.
This is especially useful where the percentage of women consuming foods from five or more food groups is low, as in the rural site shown below.
Real data were used, but to create data sets of equal size in urban and rural sites, data were randomly replicated 23 (repeated) within the data sets.
Figure 1. Percent achieving Minimum Dietary Figure 2.
Mean (SD) number of Diversity for Women of Reproductive Age food groups yesterday (MDD-W) (≥5 food groups yesterday) (Error bars indicate 95% confidence interval) Figure 3. Food group diversity scores for yesterday (out of 10 groups)
It may also be useful to present the average number of fruit/vegetable groups consumed out of the four groups (“Dark green leafy vegetables”, “Other vitamin A-rich fruits and vegetables” [usually dark yellow/orange/red], “Other fruits” and “Other vegetables”)24.
Figure 4. Mean (SD) number of fruit/vegetable groups yesterday (out of 4 groups) Consumption patterns for specific food groups Presentation and examination of the percent of WRA consuming foods from specific food groups and subgroups provide a good qualitative description of the diet.
Both the nutrient-rich food groups in the MDD-W and the optional, low nutrient density food groups may be of interest.
Different fruit/vegetable groups have different nutrient profiles, so consumption of a variety best ensures good 24 intakes of micronutrients, as well as of other phytochemicals and fibre. Many national FBDG explicitly advise consumption of a variety of types or colours of fruits and vegetables, and several specifically advise consumption of dark green leafy vegetables; see global FBDG compiled by FAO at http://www.fao.org/nutrition/nutritioneducation/food-dietary-guidelines/en/.
The results shown in this section could also be presented in a tabular format.
The data from the MDD-W questionnaire also allow for a variety of context-specific descriptive analyses. For example, differences in food group consumption for those above or below the threshold can be explored and will vary by context. Table 5 shows these differences in the same urban and rural sites used in the examples above.
All data presentation choices will depend on the audience and objectives for presentation.
These data were not collected in the source data set used to construct other figures, so values are not real.
Agudo, A. 2005. Measuring intake of fruit and vegetables. Background paper for the Joint FAO/WHO Workshop on Fruit and Vegetables for Health, September 2004, Kobe, Japan. Rome, FAO.
Alasalvar, C. & Bolling, B.W. 2015. Review of nut phytochemicals, fat-soluble bioactives, antioxidant components and health effects. Br J Nutr., 113(Suppl 2): S68–78. Doi: 10.1017/ S0007114514003729.
Albert, J. 2007. Global patterns and country experiences with the formulation and implementation of food-based dietary guidelines. Ann Nutr Metab. 51(Suppl 2): 2–7.
Arimond, M., Wiesmann, D., Becquey, E., Carriquiry, A., Daniels, M.C., Deitchler, M., FanouFogny, N., Joseph, M.L., Kennedy, G., Martin-Prével, Y. & Torheim, L.E. 2010. Simple food group diversity indicators predict micronutrient adequacy of women’s diets in 5 diverse, resource-poor settings. J Nutr. 140(11): 2059S–69S.
Bouvard, V., Loomis, D., Guyton, K.Z., Grosse, Y., Ghissassi, F.E., Benbrahim-Tallaa, L., Guha, N., Mattock, H., Straif, K. & the International Agency for Research on Cancer Monograph Working
Group. 2015. Carcinogenicity of consumption of red and processed meat. Lancet Oncol. pii:
S1470–2045(15)00444-1. Doi: 10.1016/S1470-2045(15)00444-1. [Epub ahead of print].
Del Gobbo, L.C., Falk, M.C., Feldman, R., Lewis, K. & Mozaffarian, D. 2015. Effects of tree nuts on blood lipids, apolipoproteins, and blood pressure: systematic review, meta-analysis, and doseresponse of 61 controlled intervention trials. AJCN. Doi: 10.3945/ajcn.115.110965.
FAO. 2011. Guidelines for measuring household and individual dietary diversity. Rome, FAO.
FAO. 2013. Edible insects: Future prospects for food and feed security. Rome, FAO.
George, S.M., Ballard-Barbash, R., Manson, J.E., Reedy, J., Shikany, J.M., Subar, A.F., Tinker, L.F., Vitolins, M. & Neuhouser, M.L. 2014. Comparing indices of diet quality with chronic disease
mortality risk in postmenopausal women in the Women’s Health Initiative Observational Study:
evidence to inform national dietary guidance. Am J Epidemiol. 180(6): 616–25. Doi: 10.1093/aje/ kwu173.
Gewa, C.A., Murphy, S.P., Weiss, R.E. & Neumann, C.G. 2014. Determining minimum food intake amounts for diet diversity scores to maximize associations with nutrient adequacy: an analysis of schoolchildren’s diets in rural Kenya. Public Health Nutr. 17(12): 2667–73. Doi: 10.1017/ S1368980014000469.