3. QUANTITATIVE AND QUALITATIVE APPROACHES TO DIETARY ASSESSMENT

In Chapter 3, both quantitative and qualitative approaches to dietary assessments that are of particular relevance to the Caribbean are described, and their strengths and limitations highlighted. This chapter also discusses how these approaches can be linked in food consumption studies to aid nutrition policy and planning.

Diet is multidimensional and is influenced by biological, psychological, social and cultural factors. Elucidation of diet-disease relationships requires methods that provide adequate specificity for describing food and quantifying nutrient intakes; it also requires methods that minimize systematic error and provide reasonably accurate estimates of the variability between individuals and/or groups.¹

Over the last century attempts have been made to obtain reliable estimates of food and nutrient intakes in order to relate them to the health of individuals and groups.^2-7 Two distinct methods have been practiced: (1) the quantitative method and (2) the qualitative method. A quantitative survey provides data on the amounts of various foods consumed by individuals and or populations; qualitative research provides information on the kind of foods consumed, food preparation procedures, food preferences, cultural influences and attitudes towards foods.^{4, 8} Several reviews of the subject have concluded that no single method of dietary assessment is entirely satisfactory^{2-5, 9} and it is therefore recommended that a combination of both methods based upon the type and purpose of information required is used.

Quantitative approaches to dietary assessment

Data from quantitative surveys can help inform food and nutrition policy in several ways.

They can show the:

• average consumption of foods and nutrients by socioeconomic and demographic groups;

• frequency of consumption of specific foods or food groups;

• adequacy of diet for different population groups;

• diet-disease relationships.

A quantitative approach to dietary assessment can do this by assessing both current and past intake. These methods shall be examined in more detail below (see also Boxes 2.1 and 2.2).

ASSESSING CURRENT INTAKE

Current intake can be assessed using three main methods:

• 24-hour recall;

• estimated food records;

• weighed food records.

24-hour recalls

In a 24-hour recall, the individual is asked to provide estimates of the amount of food and drink they have consumed during the previous 24-hour period. The greatest value of the 24-hour recall method is its ability to estimate nutrient intakes of population groups. This method is used widely to compare nutrient intakes with specific dietary recommendations. Repeated 24-hour recalls are often used as a comparison method to represent usual intake in food frequency questionnaires (FFQ) validity studies.¹⁰ The major limitation of recalls is that they are seldom representative of usual intake.

Estimated food records

An estimated food record is similar to a 24-hour recall: it consists of a detailed description of food and drink consumed over a period (usually three to five days). The options advantage of this method is that it provides detailed dietary intake data that are more representative of usual intake than a single 24-hour recall. The disadvantages are that a high degree of respondent cooperation is necessary, and the act of recording may alter the usual diet.

Weighed food records

The weighed food record or duplicate-food collection method is often regarded as the most precise method for estimating food and/or nutrient intakes of individuals. Here, the amounts of food consumed are actually weighed and recorded so more accurate measurements of actual intake than calculations based on food composition tables can be obtained. Weighed intakes may be used as the comparison method with other dietary methods. The disadvantage of this method, however, is that it is costly in both time and money, and requires highly motivated subjects with high levels of literacy.

ASSESSING PAST INTAKE

Past intake can be assessed using two main methods:

• dietary history;

• food-frequency questionnaires.

Dietary history

The dietary history method estimates usual food intakes of individuals over a relatively long period lasting several weeks or months. The advantage of this method is its ability to detect seasonal changes, obtain data on all nutrients and to correlate well with biochemical measures. However, major limitations are the necessity for a lengthy interview process (and a corresponding respondent burden) and the difficulty and expense of coding the data gathered.

Food-frequency questionnaire (FFQ)

The FFQ is the best available method for conducting studies on diet and disease relationships: it assesses habitual dietary intakes. The underlying principle of the food-frequency approach is that the average long-term diet (intake over weeks, months or years) is a more important exposure period than short-term intakes. The benefit of this method is that it can provide more representative information on usual intake than a few days of diet records or recalls. However, the method is

The Caribbean experience has highlighted the need for practical guidance in the development of validated FFQ instruments. Annex 5 provides an overview of the steps necessary to develop a quantitative food-frequency questionnaire, and shows how to collect the information necessary to analyse and assess diet.

Qualitative approaches to dietary assessment^[2]

Qualitative research is the examination of people's beliefs, perceptions and behaviours. Initial qualitative data collection can be gathered rapidly, and can therefore provide information that is up to date. The approach can also facilitate inquiry into sensitive issues, which are often difficult to investigate through standard survey methods.

Qualitative research can inform food and nutrition policy options in several ways. They can:

• identify specific healthy food options for government subsidy and support;

• distinguish foods most suitable for fortification;

• inform decisions on whether to tax or forbid imports or sales of 'unhealthy' foods at the national, regional, local or institutional levels;

• guide the development and format of food labeling so that it is culturally appropriate and understandable;

• pinpoint the most inappropriate marketing tactics and effective controls that could be implemented;

• suggest the format and content of national or regional campaigns to improve diet;

• indicate alterations to existing food assistance programs and how they are implemented as a way to improve food security (e.g. dispersals twice a month rather than once a month).

HOW TO MAKE THE BEST USE OF QUALITATIVE RESEARCH

On their own, qualitative approaches can generate an extremely rich source of evidence and data and as a result make a valuable contribution to the development of nutrition and food policies and interventions (Box 3.3 provides some examples). In addition, qualitative narratives provide an opportunity to address the non-quantifiable features of food consumption studies such as settings and backgrounds to dietary practices, interesting anecdotes and beliefs about foods as well as the historical evolution of cooking practices.

Qualitative methods can also be used to complement quantitative approaches. They are particularly useful when an investigator knows little about the subject being investigated - for example, a group's perceptions or beliefs regarding certain foods. Qualitative methods also play confirmatory and elucidating roles as well as add depth, substance and meaning to quantitative results.

Complete descriptions of different types of qualitative approaches and how they can be used may be found in Bernard, 200011 and Weller and Romney, 1988.12 The following methods are the most commonly used in qualitative research:

• in-depth interviewing;

• direct observation;

• participant observation;

• focus group discussions.

These shall be examined in more detail now.

In-depth interviewing

This is a type of exploratory dialogue between interviewer and informant, where the informant is seen as teaching the interviewer about their cultural setting. This method is useful for identifying actual dietary practices and for delineating facilitators and barriers to recommended practices such as in young child feeding.

Direct observations

The direct observation technique entails the recording of actual behaviour, as opposed to reported or recalled behaviour. This method is particularly useful in documenting food behaviours among various population groups. Direct observations allow investigators to go beyond the selective perceptions of others and experience the setting first hand, thereby allowing for more relevant policies and plans to be designed.

Participant observation

The participant observation method involves the observer residing in the community of interest for an extended period of time and, as the name suggests, participating in local activities. In this way, the investigator learns about the local setting through extended and informal contact with community members. By becoming part of the group, the investigator is more likely to see events from the perspective of insiders. In addition, by observing operations and activities directly within the setting, researchers gain a better understanding of the context and process of activities.

Behavioural observations are very useful for examining infant and young child-feeding practices.

Focus-group discussions

Focus-group discussions involve using groups and the group processes to generate discussion on a topic of interest. The aim of the discussion is to explore issues rather than to simply describe or explain. This method allows researchers to reach deeper levels of meaning through the use of the respondents' own words and modes of expressions. The process of local people discussing a topic in their own terms with one another will yield insights and permit natural discourse that would not be available using standard one-on-one interviewing.

The method is exceptionally valuable for obtaining information from children and from individuals who are not highly literate. Qualitative approaches have been used in the Caribbean to examine young child-feeding practices. An example is provided in Box 3.4.

KEY APPLICATIONS FOR THE USE OF QUALITATIVE INFORMATION FOR PROGRAMME DEVELOPMENT

Qualitative information can be used for programme development in several ways. It allows identification of:

• significant subgroups within a population to be targeted by programmes;

• key aspects of how people determine what constitutes a meal;

• relevant systems of food classification;

• important food proscriptions and prescriptions;

• priority food handling-related behaviors (positive and negative) for intervention;

• priority feeding behaviors within the household

• other priority behaviors for intervention;

• key foods for intervention;

• cultural metaphors for communication;

• culturally appropriate intervention materials.

HOW QUALITATIVE APPROACHES CAN BE USED TO AID THE UNDERSTANDING OF CULTURAL FACTORS WHICH MAY BE THE DETERMINANTS OF CHRONIC DISEASES IN THE CARIBBEAN

Cultural norms regarding the types of foods eaten, preparation methods used and when, where and how such foods should be served are crucial in understanding why chronic diseases, for example, exist in certain populations. In Guyana, in collaboration with the Ministry of Health, CFNI conducted a qualitative study of food consumption habits. The study assisted in identifying key cultural norms associated with food preparation and consumption. Some of the key questions this qualitative survey asked were:

• What are local meal patterns and food preferences?

• What are local norms relating to eating style? For instance, are family members encouraged to finish all the food on their plates?

• How are "favorite" foods prepared and served?

• What are local perceptions of preferred body size? What value is placed on fatness?

The study also provided information on commonly eaten foods by ethnic and socio-economic groups and commonly used household measures. These helped guide the development of the data collection instrument for the quantitative phase of the study (see Box 3.5 for further examples). Findings from the survey were also used in the development of food-based dietary guidelines and in designing a pilot community education programme.

Combining qualitative and quantitative approaches in food consumption studies

Ideally, a complete food consumption study will integrate both quantitative and qualitative approaches in order to achieve convergence of results. Qualitative methods will provide context to the quantitative findings and ground them in a specific cultural context, and in this way provide a much broader and deeper analytical and interpretative framework than could be obtained from either of the two methods being used independently.²

Combining the approaches allows "facilitation", in which one approach facilitates or "assists" the other.¹³ For example, results from qualitative studies are valuable for improving the quality of quantitative studies by identifying hypotheses that might be tested as well as to inform and improve the quality of the data collection instrument intended to assess dietary intake. Boxes 3.6 and 3.7 provide details.

The implications of results of quantitative studies and their policy relevancy may be analysed using qualitative approaches. This is especially relevant for the development of programmes to improve diet and nutritional status among sub-population groups. Qualitative results help assist the instrument development for quantitative surveys in a variety of ways. They help:

• ascertain what the main nutritional/dietary issues are in a particular setting;

• select the most appropriate dietary assessment instrument for the task;

• choose appropriate tools for estimating portion size;

• decide the most suitable proxy reporters for the diets of individuals who cannot provide their own report (e.g., small children);

• select foods for inclusion on food-frequency questionnaires and infant-diet histories;

• pick appropriate wording (labels) for foods on a food frequency questionnaire;

• ascertain the appropriate frequency categories for a food frequency instrument;

• decide on appropriate probing questions for 24-hour recalls;

• design additional survey questions that can assess the impact of a nutrition-intervention programme on cognitive and behavioural factors;

• develop questions for structured surveys;

• aid in the interpretation of quantitative findings