|Page tools: Print Page Print All|
3. Comprehensive nutrition surveys, such as the National Nutrition Survey (NNS) conducted as a joint project with the then Commonwealth Department of Health and Family Services and the Australian Bureau of Statistics (ABS) in 1995, are expensive to develop and implement, and impose a substantial reporting load on the selected respondents. In contrast, short dietary questions about food habits can be included in population health surveys at a much reduced cost. They can provide specific information on aspects of food habits such as food diversity (e.g. fruit and vegetable intake) or social behaviours that affect food intake. While they provide an incomplete picture compared to the detailed information of a nutrition survey, they are useful for monitoring trends in population behaviours linked to important aspects of nutrition such as food diversity.
4. This paper provides a detailed explanation of the methodological differences between the short Dietary Indicators module in the 2001 National Health Survey (NHS) and the equivalent questions in the 1995 National Nutrition Survey (NNS). The focus is on changes in methodology for the short dietary questions in the move from a self-enumeration form in 1995 NNS to a personal interview in 2001; and the development of new questions on deliberate intake of folate fortified foods and beverages. This information will support the development of short dietary questions used in other personal interview based population health surveys.
2. THE 2001 NATIONAL HEALTH SURVEY (NHS)
5. The 2001 National Health Survey was conducted over the period from February to November 2001, with development commencing in 1998. It is the fifth survey of its type conducted by the ABS since 1977, and the first in a new series of health surveys planned to be conducted every three years. (Refer to 5.8 for details of sampling frame and standard error margins).
6. The surveys are designed to obtain national benchmark information on a wide range of health issues, and to enable trends in aspects of health to be monitored over time. The questionnaires for the 2001 survey were developed in consultation with key government, professional, community and industry bodies, to address their highest priority information needs as appropriate to a survey of this type. Representatives from these organisations form the nucleus of the National Health Survey Reference Group, which meets to discuss aspects of the survey and advise the ABS on consumer/client needs, as part of the survey development and evaluation process.
7. The survey collects indicators covering three broad areas of health:
8. A range of demographic and other characteristics is also obtained. Where appropriate, the survey's content and methodologies are consistent with those of past surveys, to facilitate comparisons of results over time. (Refer to 2001 NHS Users' Guide)
9. The NHS Reference Group endorsed the ABS proposal to include a new Dietary Indicators module as a core module in the 3-yearly NHS. The focus of the Dietary Indicators module is to monitor food habits in Australia and thereby provide regular indicators for the Australian population. Consultations with stakeholders had identified the creation of a time series on dietary habits as a high priority. Given the limited space and the wide range of health topics to be covered in the NHS, the size of the Dietary Indicators module was restricted to approximately 6 to 8 questions.
10. The 2001 NHS also collected information on breastfeeding. This module is not discussed here, as a separate paper addressing this aspect of nutrition is planned for release in September 2003. (Refer to ABS cat. no. 4810.0.55.001).
3. SELECTION OF DIETARY INDICATORS
11. The National Public Health Nutrition Strategy and policies such as Eat Well Australia and the National Fruit and Vegetable initiative provided the policy focus for the identification of relevant topics in the Dietary Indicators module. The ABS, in consultation with a small technical working group of representatives from the Strategic Inter Governmental Nutrition Alliance (SIGNAL), the then Department of Health and Aged Care and the National Food and Nutrition Monitoring and Surveillance Project, identified the following topics for inclusion in the Dietary Indicators module:
12. From the topics listed above, folate was identified as a high priority. The aim was to measure whether women of child-bearing age (15 to 49 years of age) were deliberately consuming foods, beverages and supplements which had been fortified with folate. In 1994, the National Health and Medical Research Council Expert Panel on Folate Fortification had recommended a target for folate consumption among women of child-bearing age of 400 micrograms of folate per day, including natural folate and fortified foods. The aim of this target was to reduce the incidence of neural tube defects (NTDs) in children. Changes to the Food Standards Code permitted the fortification of relevant foods on a voluntary basis by the food industry. As this was a relatively new measure, most of the initial efforts to collect data focused on measuring the awareness of folate and its link to preventing NTDs and whether women of child-bearing age would increase their folate intake once they understood the link. So the priority was to collect information on the intake of folate fortified products.
13. The Australian Guide to Healthy Eating (Smith 1998) recommends that the Australian population choose different varieties of foods from each of the five main food groups:
14. The collection of information on cereal intake was considered important to provide additional information on food diversity. An added benefit was that by collecting information on both the usual fruit, vegetable and cereal intakes, data on the intake of foods which contain naturally occurring folate would also be available to users.
15. Table 1 lists the six Dietary Indicators collected in the 2001 NHS and identifies the 5 indicators common to the 2001 NHS and the 1995 NNS.The preliminary design of the Dietary Indicators module included 4 new questions and 5 existing questions from the 1995 National Nutrition Survey (NNS), although one was later omitted (i.e. eight questions provided the basis for six indicators). This section briefly outlines the design of the new questions and also discusses methodological and contextual differences between the questions common to the 2001 NHS and the 1995 NNS.
Table 1. Dietary indicator questions used in the 2001 NHS and the 1995 NNS
3.1 Cereal intake (new indicator)
16. The usual intake of cereals was considered an important indicator of food diversity. A single question asked the respondent to report usual intake of the main forms of cereals and cereal-based foods.
3.2 Folate (new indicator)
17. In the latter part of 1999 questions on the intentional consumption of folate fortified foods and supplements were designed by the ABS in consultation with the technical working group (see Section 3). These questions were developed from surveys such as the 1997 Eat Well Tasmania survey and the 1995 and 1997 Health Department of Western Australia evaluation of folate health promotion program, which had included questions designed to measure the awareness, attitudes and the intention to consume folate fortified products by women of child bearing age.
18. The technical working group determined that the main priority was to identify the population of women of child bearing age (18-49 years) who had intentionally consumed folate fortified products and/or folate supplements. Three questions were designed to measure the deliberate intake of three different types of folate products, foods, beverages and folate supplements. A short recall period of two weeks prior to interview was implemented to maximise the ability of the respondent to remember consumption of these products.
19. If respondents were unable to answer any of the questions because they did not understand the concept of folate, then it was assumed that they had not intentionally consumed folate fortified products or folate supplements.
3.3 Questions common to the 2001 NHS and the 1995 NNS
20. Apart from the folate questions and the usual cereal intake question, all items on dietary habits in the 2001 NHS were included in the 1995 National Nutrition Survey (NNS). The questions on adding salt to cooked food and whether the respondent had run out of food in the last 12 months and couldn't afford to buy more (food insufficiency) are exactly the same as the 1995 NNS. However, the question on type of milk usually consumed in the 2001 NHS includes a new category for soy milk and therefore the category 'None of the above' is not comparable between the 1995 NNS and the 2001 NHS. Another important difference was that in the 1995 NNS, respondents were able to select multiple categories of milk consumption. However, the indicator of more interest is 'usual consumption of milk' rather than a description of all milk products consumed. To reflect this, NHS respondents in 2001 were required only to select a single category of milk consumption which best matched their usual intake of milk product.
3.4 Methodology and contextual changes of questions common to the 2001 NHS and 1995 NNS
21. In the 1995 NNS, the questions on usual fruit and vegetable intake were included at the end of the Food Frequency Questionnaire (FFQ). The FFQ was administered as a self-completion questionnaire and was given to respondents aged 12 years and over at the end of the face-to-face interview. This placement ensured that the respondent had been introduced to the definitions of fruit and vegetables and other dietary concepts through the earlier FFQ questions and the face to face interview. The average total interview time per respondent was 43 minutes which included 26 minutes for the Individual Food Intake Questionnaire (IFIQ), 9 minutes for the physical measurements and 8 minutes for the questions on food habits and attitudes.
22. To understand the contextual effects of the different components of the 1995 NNS on the respondent's ability to answer the questions on usual fruit and vegetable intake, it is important to understand the different methodology used in the 1995 NNS (see Appendix D). In the 2001 NHS, the questions on Dietary Indicators were placed approximately one quarter of the way into the questionnaire in a designated module. Respondents did not have the assistance of undertaking an IFIQ prior to answering the questions on usual fruit and vegetable intake.
4. DEVELOPMENT AND TESTING OF THE DIETARY INDICATORS MODULE
23. As a new module in the NHS, the ABS had to test the placement of the questions and respondent understanding of the questions in a cognitive testing laboratory. The aim of the cognitive interviewing process was to assess respondent understanding of concepts and question wording and assess question module sequencing and context effects on the answers using a standard cognitive interview method. A number of cognitive testing techniques were used to explore the processes of memory storage and retrieval, and how the information required was processed by the respondent (e.g. "think aloud", retrospective and concurrent probing). Further information on the results of the cognitive testing are included in Appendix E.
24. Results from the cognitive testing provided a focus for further development of the questions used in the Dietary Indicators module. Both the questions and interviewer instructions were revised and prompt cards were developed for the questions on the usual intake of vegetables, fruit and cereals to assist respondents to calculate the number of serves usually eaten.
25. The revised Dietary Indicators module was then included in a subsequent field test of the 2001 NHS on approximately 400 respondents. The results of this field test indicated that respondents had difficulty with determining the number of serves they usually consumed for the questions on the usual intake of fruit, vegetables and cereals. (See Appendix E). The ABS sought advice from the technical working group on the validity of the concept of 'usual intake' and whether analysis of the results of the usual intake questions in the 1995 NNS supported the continued inclusion of these questions in the 2001 NHS.
26. The Australian Food and Nutrition Monitoring Unit was then able to provide the preliminary results of an analysis of the usual fruit and usual vegetable intake questions in the 1995 NNS compared to the information in the 24-hour dietary recall (IFIQ) (Rutihauser et al. 2001). The overall conclusion of this analysis was that responses to the short questions on fruit and vegetable intake can provide reliable information on fruit and vegetable intake across a range of population sub-groups that are generally consistent with group level differences in fruit and vegetable intake, as determined by the 24-hour recall approach.
27. Further discussion of this analysis by the technical working group determined that the mean response category for population sub-groups in a time series would provide a good indicator of changes in dietary habits over time. The ABS then decided to retain the questions on the usual fruit intake and usual vegetable intake, subject to finding a new method to assist respondents to determine the number of serves they usually consume. The question on usual cereal intake was omitted because of continued concerns about the question's complexity since results from both cognitive testing and the field test indicated respondents had major difficulties in estimating their usual daily consumption of cereals in the form of breads, breakfast cereals and rice and pasta.
28. The ABS in consultation with the technical working group then examined different options for improving respondent's ability to understand what a serve of fruit or vegetables was. The main focus was on redesigning the prompt card for each question. The purpose of the prompt card was to:
29. When developing the prompt cards it was noted that photographs are a widely used and accepted means for defining and quantifying portion sizes in many national health and dietary surveys around the world (Powers 2000). In consultation with the technical working group, the ABS arranged to produce photographs of single serves of a variety of fruits and photographs of a variety of vegetables. (Refer to Appendix B for the prompt cards used in the Dietary Indicators module).
Table 2. List of photographs of single serves used in the prompt cards for usual fruit and vegetable intake.
30. The interviewer instructions were also modified to encourage respondents to look at the pictures of serves and count the equivalent number of serves they usually consumed. If necessary, the interviewer would prompt the respondent to recall the usual numbers of serves of fruits/vegetables by eating occasion (breakfast, lunch, dinner, in-between meals).
5. RESULTS OF THE 2001 NHS
31. Comparison of Dietary Indicators from the 1995 National Nutrition Survey and the 2001 National Health Survey suggest that there has been some changes in dietary habits by the Australian population. A brief analysis of the six indicators common to both surveys follows. There is also a brief analysis of folate consumption among women aged 18-49 from the 2001 NHS. (Table 3.)
5.1 Usual type of milk consumed (proxy indicator of saturated fat intake)
32. Respondents who report consumption of full cream milk provide a good proxy indicator for their saturated fat intake (Marks et al. 2001). The data on the type of milk usually consumed in 2001 appears to be consistent with the data in 1995, despite some of the methodological differences described earlier (see section 3.4). Approximately 4% of Australians reported consumption of soya milks in 2001. The slight decline in the 'not applicable' category (1.4% in 1995 to 0% in 2001) reflects the improved data quality achieved by changing the methodology of this question from the self-enumerated Food Frequency Questionnaire to an interviewer-based paper questionnaire. However the slight increase (4% in 1995 to 5% in 2001) in the category 'none of the above' suggests that further examination of the types of milk available on the Australian market may be needed to determine if a new milk category needs to be included for the next NHS.
5.2 Usual daily intake of vegetables
33. Usual daily intake of vegetables is an indicator of food diversity. The Australian Guide to Healthy Eating (Smith et al. 1998) recommends that adolescents aged 12 to 18 years consume at least 4 serves (300 grams) and adults 5 serves (375 grams) of vegetables and legumes each day. Results from the 2001 NHS suggest that approximately 11 percent of Australians reported a higher usual daily intake of vegetables compared to the 1995 NNS. The usual daily intake increased for the combined categories of '4 or more serves' (18% in 1995 to 30% in 2001) with a corresponding decrease of 11% in the combined categories of '1-3 serves' (81% in 1995 to 70% in 2001). There are a number of possible reasons for the change in this indicator:
5.3 Usual daily intake of fruit
34. Usual daily intake of fruit is another indicator of food diversity. The Australian Guide to Healthy Eating (Smith et al. 1998) recommends consumption of at least 2 serves of fruit per day (300 grams). Results of the 2001 NHS compared to the 1995 NNS show a small decline of 4.5% of respondents who reported consumption of one serve or less of fruits but a corresponding increase of 3.8% of respondents who reported that they did not eat fruit. This minor shift in categories may be the result of the new methodology used in the 2001 NHS (i.e. the introduction of colour prompt cards and additional interviewer instructions).
5.4 Frequency of discretionary sodium intake
35. This is a direct indicator of dietary behaviour in that the question seeks to measure the proportion of respondents who add salt to their food after cooking. In 1995 this question was included in Section B of the National Nutrition Survey and was administered by a trained interviewer. Therefore this question is methodologically comparable to the 2001 NHS. Results indicate a slight change in dietary behaviour with 55% of respondents reporting that they 'never/rarely' added salt to their food in 2001 compared to 61% in 1995. The shift appears to have occurred from those respondents who reported 'usually' adding salt to their food in 1995 (20%) compared to 2001 (26%).
5.5 Food insufficiency
36. As an indicator, this question seeks to identify respondents who may be at risk of a poor diet and nutrition due to their financial incapacity to purchase food. Results from the 2001 NHS are very similar to those in the 1995 National Nutrition Survey. This question was included in Section B of the 1995 NNS and was administered by a trained interviewer.
5.6 Deliberate intake of folate (2001 NHS only)
37. Eleven percent of women (an estimated 509,000 women) aged 18-49 years reported that they had intentionally consumed a folate fortified product (food, drink or vitamin/mineral supplement) in the previous two weeks prior to interview. Consumption of folate fortified foods and vitamin/mineral supplements were slightly more prevalent than the consumption of folate fortified beverages. Of these women, 117,000 (23%) were living as a couple and 191,000 (38%) were living as a couple in a household with children aged 0-14 years. The intentional intake of folate fortified products was highest amongst women aged 30-34 years at 97,000 (19.1%). Only 1.8% of women reported that they did not know whether they had intentionally consumed folate fortified products suggesting that the quality of the data for these questions was high.
5.7 Comparison of results from the 2001 NHS with the 1995 NNS
38. The following table details this comparison.
Table 3. Comparison of results from the 2001 NHS with the 1995 NNS
np not available for publication but included in totals where applicable, unless otherwise indicated.
Table 4. Proportion of women aged 18 to 49 years by intentional intake of Folate
39. The 2001 NHS was conducted using a stratified multistage area sample of private dwellings.
Within selected dwellings a random sub-sample of residents was enumerated as follows:
40. This approach yielded complete interviews from 2,300 children aged 12-17 years and 17,900 adults aged 18 years and over. There were 5,800 women aged 18-49 years in the sample.
41. The 1995 NNS was conducted using a systematically selected sub-sample of Collection Districts from the base 1995 NHS private dwelling sample. Unlike the 1995 NHS sample which involved all people in the household, the NNS was conducted by approaching a maximum of two in-scope people per household in urban areas and three in-scope people in rural areas. This sample design yielded complete interviews from 900 children aged 12-17 years and 10,900 adults aged 18 years and over. (Refer to 1995 NNS Users' Guide, ABS cat. no. 4801.0)
42. Since the estimates in this publication are based on information obtained from a sub-sample of usual residents of a sample of dwellings, they are subject to sampling variability; that is, they may differ from those that would have been produced if usual residents of all dwellings had been included in the survey. One measure of the likely difference is given by the standard error (SE), which indicates the extent to which an estimate might have varied by chance because only a sample of dwellings was included. (Refer to the 2001 NHS Users' Guide).
43. For further information on the 2001 NHS or the 1995 NNS please visit the Health Theme page on this site or contact the Health Section of the ABS.
Australian Bureau of Statistics
Locked Bag 10
Belconnen ACT 2616
Toll free: 1800 060 050
Cognitive testing is generally referred to as "asking respondents questions about questions". Cognitive interviewing identifies difficulties in question comprehension, memory recall strategies, difficulties in selecting a response, interpretation of reference periods and reactions to sensitive questions. Cognitive interviews are conducted on a one-to-one basis between the interviewer and the respondent and are usually recorded on either audio or video tape. The interviews are usually conducted in a standardised environment such as an interview room, office, boardroom or laboratory.
Dietary indicators are short questions on dietary habits that provide a summary indicator on limited aspects of dietary behaviour.
Folate is a water soluble B vitamin which plays an essential role in metabolism and in the division of all body cells including those in blood. It exists in many chemical forms.
Folic acid is the most stable form of folate. It occurs rarely in foods but is the form most often used in vitamin supplements and in fortified foods.
Fortification is the addition of an essential nutrient to a food at a rate higher than would naturally occur in the food.
Food diversity is the consumption of foods from each of the major food groups each day. (Kant et al. 1991).
Food Frequency Questionnaire A self-enumerated questionnaire given to respondents at the end of the interview for the 1995 National Nutrition Survey. Respondents aged 12 years and over were asked to complete this qualitative questionnaire, which collected usual frequency of consumption of 107 food items and 11 vitamin and mineral supplements over the past 12 months.
Food insufficiency is an inadequate amount of food intake due to a lack of money or resources.
Neural tube defects include three specific abnormalities of development of the central nervous system, namely anencephalus, spina bifida and encephaloccoele.
Nutrition concerns the intake of foods and the substances they contain, and their actions and effects within the body.
Saturated Fat One of the three fatty acids. It is a unit of carbon, hydrogen and oxygen and combines with glycerine to form fat.
Self-enumerated is when a person self-completes a questionnaire in a survey rather than being asked questions directly by an interviewer.
24-hour dietary recall This was the methodology used to collect detailed information on food and nutrient intake in the 1995 National Nutrition Survey. The 24-hour dietary recall collected a list of foods and beverages consumed the previous day from midnight to midnight, the amount consumed, the time of consumption, the name of the eating occasion, the source of the foods and beverages, whether they were consumed in the home and whether they were ever in the home.
Abraham, B., Webb, K. 2001, 'Interim evaluation of the voluntary folate fortification policy', Australian Food and Nutrition Monitoring Unit, Commonwealth of Australia, Canberra.
Australian Bureau of Statistics 1998, 'National Nutrition Survey, 1995, Users' Guide', cat. no. 4801.0, ABS, Canberra.
Australian Bureau of Statistics 1999, 'National Nutrition Survey, Foods Eaten, Australia, 1995', cat. no. 4804.0, ABS, Canberra.
Australian Bureau of Statistics 2002, 'National Health Survey, 2001, Summary of Results', cat. no. 4364.0, ABS, Canberra.
Australian Bureau of Statistics 2003, 'National Health Survey, 2001, Users' Guide', (report on ABS website www.abs.gov.au), Canberra.
Cypel, Y.S. & Tippett, K.S (eds) 1998, 'Design and operation: The Continuing Survey of Food Intakes by Individuals and the Diet and Health Knowledge Survey 1994-96', U.S. Department of Agriculture, Agricultural Research Service, Nationwide Food Surveys Report, USA.
Guenther, P.M., DeMaio, T.J., Ingwersen, L.A.1995, 'The multiple-pass approach for the 24-hour recall' in 'The Continuing Survey of Food Intakes by Individuals (CSFII) 1994-96', International Conference on Dietary Assessment Methods, Boston, USA.
Kant AK, Block G, Schatzkin A, Ziegler RG, Nestle M 1991, 'Dietary diversity in the US population, NHANES II 1976-1980', Journal of the American Dietetic Association, vol 91, pp1526-1531.
Marks, G.C., Webb, K., Rutihauser, I.H.E. & Riley, M.2001, 'Monitoring food habits in the Australian population using short questions', Australian Food and Nutrition Monitoring Unit, Commonwealth of Australia, Canberra.
National Health and Medical Research Council 1992, 'Dietary guidelines for Australians', AGPS, Canberra.
National Health and Medical Research Council 1994, 'Folate fortification', AGPS, Canberra.
National Health and Medical Research Council 1995, 'Dietary guidelines for children and adolescents', AGPS, Canberra.
National Health and Medical Research Council 1995, 'The core food groups: the scientific basis for developing nutrition education tools', AGPS, Canberra.
National Health and Medical Research Council 1999, 'Dietary guidelines for older Australians', AGPS, Canberra.
Powers S. 2000, 'Amount estimation tools (Food Models): Standardization and quality control' 4th International Conference on Dietary Assessment Methods, Tucson, Arizona, USA.
Rutihauser, I.H.E., Webb, K., Abraham, B., Allsopp, R.2001, 'Evaluation of short dietary questions from the 1995 National Nutrition Survey', Australian Food and Nutrition Monitoring Unit, Commonwealth of Australia, Canberra.
Smith, A., Kellett, E., & Schmerlaib Y.1998, 'The Australian Guide to Healthy Eating', Commonwealth Department of Health and Family Services, Canberra.
Strategic Inter-Government Nutrition Alliance (SIGNAL) 2001, 'Eat Well Australia', National Public Health Partnership.
Whitney, E. N., Rolfes, S.R. 1999, 'Understanding Nutrition', 8th Edition, Wadsworth Publishing Co, Belmont, USA.
APPENDIX A: 2001 NHS DIETARY INDICATOR QUESTIONS
Q301. THE NEXT FEW QUESTIONS ARE ABOUT NUTRITION.
WHAT TYPE OF MILK DO YOU USUALLY CONSUME?
Q302. THIS QUESTION IS ABOUT YOUR USUAL CONSUMPTION OF VEGETABLES, INCLUDING FRESH, FROZEN AND TINNED VEGETABLES.
Interviewer: Show Prompt Card 9
HOW MANY SERVES OF VEGETABLES DO YOU USUALLY EAT EACH DAY?
Q303. THIS QUESTION IS ABOUT YOUR USUAL CONSUMPTION OF FRUIT, INCLUDING FRESH, DRIED, FROZEN AND TINNED FRUIT.
Interviewer: Show Prompt Card 10
HOW MANY SERVES OF FRUIT DO YOU USUALLY EAT EACH DAY?
Q304. HOW OFTEN DO YOU ADD SALT TO YOUR FOOD AFTER IT IS COOKED?
IS IT, NEVER, RARELY, SOMETIMES OR USUALLY?
Q305. Sequence Guide:
. If respondent is female AND aged less than 50 -------> Q.306
. Otherwise ----------------------------------------------------------> Q.309
Q306. THE NEXT FEW QUESTIONS ARE ABOUT FOLATE OR FOLIC ACID WHICH CAN BE ADDED TO SOME FOODS, BEVERAGES, AND VITAMIN OR MINERAL SUPPLEMENTS.
IN THE LAST 2 WEEKS, HAVE YOU EATEN ANY FOOD PRODUCTS BECAUSE THEY HAD FOLATE ADDED TO THEM?
Q307. IN THE LAST 2 WEEKS, DID YOU DRINK ANY BEVERAGES BECAUSE THEY HAD FOLATE ADDED TO THEM?
Q308. IN THE LAST 2 WEEKS, HAVE YOU TAKEN ANY VITAMIN OR MINERAL SUPPLEMENTS BECAUSE THEY CONTAINED FOLATE OR FOLIC ACID?
Q309. IN THE LAST 12 MONTHS, WERE THERE ANY TIMES THAT YOU RAN OUT OF FOOD AND COULDN’T AFFORD TO BUY MORE?
APPENDIX B: 2001 NHS DIETARY INDICATOR PROMPT CARDS
APPENDIX C: 2001 NHS DIETARY INDICATOR INTERVIEWER INSTRUCTIONS
1.1 Nutrition (Q.301-Q.309)
A serve of vegetables equals half a cup of cooked vegetables or one cup of salad vegetables. This is equivalent to approximately 75 grams of vegetables.
Respondents should be encouraged to look at the pictures of vegetable ‘serves’ provided on Prompt Card 9 to help jog their memory and improve their understanding of the definition of a serve. Respondents should add up the vegetables they have eaten (including snacks) and convert them into serves. Once the respondent reaches the ‘6 serves or more’ category, they should be prompted to stop counting. However, respondents should be reminded that it is usual consumption we are after. Counting the serves is only used to assist the respondent.
If the respondent is still having difficulties answering this question, it may help if you prompt the respondent to describe each serve of vegetables she/he usually eats for each meal or snack. For example,
“Can you tell me what vegetables you usually eat for breakfast?”
“Can you describe the serves of vegetables that you usually have for lunch?
“What vegetables do you usually have for dinner?”
“Do you usually eat any vegetables in between meals?” If “Yes”, then ask, “What vegetables are they?”
A serve of fruit equals one medium piece or two small pieces of fruit or one cup of diced fruit pieces or one quarter of a cup of sultanas or four dried apricot halves. Note that a smaller quantity of dried fruit, such as apricots or sultanas, is counted as a serve. This is equivalent to approximately 150 grams of most fruits except dried where it is only 50 grams.
Respondents should be encouraged to look at the pictures of ‘serves’ provided on Prompt Card 10. They should add up the fruits they have eaten (including snacks) and convert them into serves. As with the previous question, once the respondent reaches the ‘6 serves or more’ category, she/he should be prompted by the interviewer to stop counting.
If the respondent is still having difficulties answering this question, it may help if you prompt the respondent to describe each serve of fruit she/he usually eats for each meal or snack. For example,
“Can you tell me how many pieces of fruit you usually eat for breakfast?”
“Can you describe the serves of fruit you usually have for lunch?
“How many serves of fruit do you usually have for dinner?”
“Do you usually eat any fruit in between meals?” If “Yes”, then ask, “What sort of fruit is that?”
However, respondents should be reminded that it is usual consumption we are after. Counting serves is only used to assist the respondent.
Q.306, Q.307 and Q.308 aim to measure the intentional consumption of fortified (enriched) folate products and to identify women of childbearing age (defined here as 18 to 50 years) who have deliberately consumed folate or folic acid, fortified (enriched) food or taken vitamins or minerals in the two weeks prior to the interview because they had folate added.
Folate is a B vitamin also known as folic acid, folacin or pteroylglutamic acid. It is found naturally in certain foods such as fresh vegetables and fruit, orange juice, legumes, nuts, liver and yeast. Scientific studies have shown that consumption of folate by women who are pregnant, or trying to get pregnant, significantly reduces the risk of neural tube defects in babies. The most common form of neural tube defect is spina bifida.
Q.306, Q.307, and Q.308 refer to the deliberate consumption of foods, beverages and supplements which have been fortified with folate or folic acid. That is, they have had folate added to them, or their existing folate content has been increased. We are not trying to measure whether the respondent consumed foods or beverages which naturally have a high folate content.
The information from this question will be linked to other data items, such as dietary habits and health status indicators, to produce an indicator of food insufficiency.
APPENDIX D: METHODOLOGY OF THE 1995 NNS
Information on diet was not collected in the 1995 NHS, but was obtained in the 1995 National Nutrition Survey (NNS), which was conducted in association with the NHS. At the end of the NHS interview, people selected for the NNS were advised that the ABS was also collecting information about the nutritional patterns of Australians and their agreement was sought for a further interview. It was explained that a professional nutritionist would arrange an appointment in the home within a few weeks to collect some information about what they eat and would take some physical measurements, the results of which would be forwarded to them. The NNS interview was structured to collect information on:
The 24-hour recall questionnaire collected detailed information on all foods and beverages consumed on the day prior to interview, from midnight to midnight. Information collected included the time of consumption, the name of the eating occasion (e.g. breakfast), detailed food descriptions to allow for accurate food coding, the amount eaten, the source of the foods and whether the foods were consumed at home. A multiple-pass recall method was used in the NNS which was adapted by the Department of Health and Family Services from that used in the Continuing Survey of Food Intakes by Individuals 1994-96 of the United States Department of Agriculture (USDA). Components of the 24-hour recall questionnaire included:
The methodology was developed by the Agricultural Research Service of the USDA to maximise the ability of respondents to remember what had been eaten and drunk. (Cypel & Tippett 1997; Guenther 1995).
Standardised measuring guides were used to assist the respondent to estimate the amount of food actually consumed. These were:
For additional information see National Nutrition Survey: Users' Guide, Australia 1995 (ABS cat. no. 4801.0).
After spending an average of 26 minutes completing the IFIQ, the respondents in the 1995 NNS had developed a good understanding of the type and amount of foods and beverages they had consumed in the previous 24 hour period. Their recall had been enhanced by the multiple pass method of the IFIQ and the use of aids to assist them in reporting the dimension and volumes of the food they had consumed. This context was lacking for respondents answering questions on dietary habits collected by interview in the 2001 NHS.
APPENDIX E: RESULTS OF TESTING ON DIETARY INDICATORS MODULE
The main objective of cognitive interviewing is to identify "errors" in the questionnaire and the source of errors through obtaining in-depth knowledge of respondents' thoughts when they answer the questions. Cognitive interviewing identifies difficulties in question comprehension, memory recall strategies, difficulties in selecting a response, interpretation of reference periods and reactions to sensitive questions. Cognitive interviewing generally involves the respondent providing extra information about their interpretation of the questions in a questionnaire and/or extra information about how they came to their answers.
Testing focussed on assessing respondents' conceptual understanding of the folate questions (as part of the diet module to be included in the main NHS questionnaire) using a standard cognitive interview method. A number of cognitive testing techniques were used to explore the processes of memory storage and retrieval, and how the information required was processed by the respondent (e.g. "think aloud", retrospective and concurrent probing).
Respondents demonstrated that they had little difficulty responding to the questions on usual type of milk consumed, salt added to foods after cooking, and the folate questions. It was suggested that an added category on soy milk be included. However the cognitive testing revealed difficulties with the questions on usual intake and food insufficiency, as discussed below.
Usual vegetable intake
Respondents had difficulty answering this question. Many had trouble calculating the number of serves, since most people don't measure out their vegetables and those who do often use other measures, such as "spoonfuls". Despite being shown the prompt card, some respondents still appeared to be equating one serve with one meal (e.g. lunch is one serve, dinner is another, rather than in terms of 1/2 cup measures). Another source of confusion was that some respondents counted having three types of vegetables in a meal as having three serves, without appearing to consider the quantity of vegetables consumed. Some respondents responsible for cooking the family meal appeared to be thinking in terms of the vegetables consumed by their whole family, rather than what they might have consumed as an individual. Other respondents commented that it was difficult to answer since they did not have vegetables every day and the amount of vegetables they ate also varied depending on the season.
The majority of respondents spent time calculating their answers, generally thinking about the cup measurement. Many respondents worked through a whole day breaking it down into meal times to calculate a total per day. On several occasions this resulted in a larger number of serves reported and respondents were more confident of their answers. There was a question about whether potatoes were included. As a result of the cognitive testing, the question was clarified to ensure that respondents were made aware that fresh, frozen, and tinned vegetables were in scope, and that vegetable juice was not. A new prompt card was designed which listed different types of vegetables with a brief description of the size of a serve of vegetables to assist the respondent to calculate the usual number of serves eaten.
Usual intake of fruit
The majority of respondents spent time calculating their answers. A small number used meals as a guide to provide an answer. Many commented that there was a seasonal influence in what and how much they ate. Several respondents questioned whether to include fruit juice, canned fruit and dried fruit. Once again, several respondents commented that the amount they eat varies from day to day. As a result of the cognitive testing, the question was clarified to ensure that respondents were made aware that fresh, frozen, tinned and dried fruit were in scope, and that fruit juice was not. A new prompt card was designed which included a brief description of the size of a serve of fruit to assist the respondent to calculate the usual number of serves eaten.
Usual cereal intake
Respondents had similar difficulties answering this question as they did with the question on vegetable intake - particularly with regard to calculating number of serves consumed. Further examination of the respondent's initial responses compared with their response after probing showed that they inaccurately estimated their cereal consumption. In 85% of cases of mis-reporting the bias was towards under-reporting and was large enough to change their response category after probing. Respondents appeared to have difficulty understanding what foods should be included as part of their usual cereal intake. When respondents were prompted to spend more time thinking about what they had included in their answer their reported consumption increased. As a result of the cognitive testing, the question was clarified to ensure that respondents were made aware that breads of all types should be included: english style muffins, crumpets and flat breads; breakfast cereals including porridge; pasta and rice products. A new prompt card was designed which listed different types of cereals with a brief description of the size of a serve of each type of cereals (breakfast cereals, bread, rice and pasta) to assist the respondent to calculate the usual number of serves eaten.
When asked if they had run out of food in the last 12 months and could not afford to buy more, most respondents answered "no". Probing of respondents revealed that many did not recall the last 12 months and thought only about running out of food. Some respondents queried the placement of the question in the middle of the Dietary indicators module. As a result of cognitive testing the 'food insufficiency' question was moved to the end of the Dietary indicators module.
After providing additional explanatory information in the interviewer's instructions and moving the food insufficiency question to the end of the Dietary indicators module, a full pilot test for the complete 2001 NHS was conducted on 300 households which provided approximately 400 respondents. Feedback from the pilot test and interviewer debriefings highlighted the problems with the usual intake questions, particularly the question on cereal intake.
Respondents continued to have difficulty defining a serve. Interviewers needed considerable time to prompt them. In terms of reporting their 'usual' behaviour most respondents commented that they do not eat fruit, vegetables or cereals every day. They tended to focus on main meals and thus excluded foods consumed as snacks when formulating their answers. Even with a detailed definition at the beginning of the usual cereal intake question and a prompt card, respondents appeared to not include all types of cereals. The impression of the interviewers was that respondents defined 'cereals' as 'breakfast cereals' only.
These documents will be presented in a new window.