Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Dietary patterns of Australian adults and their association with socioeconomic status: results from the 1995 National Nutrition Survey


Objective: To describe dietary patterns among men and women in the Australian population, and to explore how these varied according to socioeconomic status (SES).

Design: A cross-sectional self-report population survey, the 1995 Australian National Nutrition Survey (NNS), was used.

Setting: Private dwelling sample, covering urban and rural areas across Australia.

Subjects: Data provided by 6680 adults aged 18–64 who participated in the NNS were used in the analyses.

Methods: Factor analyses were used to analyse data from a Food Frequency Questionnaire (FFQ) completed by participants. Associations between SES and dietary pattens were assessed using ANOVA.

Results: Separate factor analyses of the FFQ data for men and women revealed 15 factors, accounting for approximately 50% of the variance in both men's and women's dietary patterns. Several gender and SES differences in food patterns were observed. Lower SES males more frequently consumed ‘tropical fruits’, ‘protein foods’, and ‘offal and canned fish’, while high SES males more often ate ‘breakfast cereals’ and ‘wholemeal bread’. Lower SES females more often ate ‘traditional vegetables’, ‘meat dishes’ and ‘pasta, rice and other mixed foods’, while high SES females more frequently ate ‘ethnic vegetables’ and ‘breakfast cereal/muesli’.

Conclusions: These findings contribute to a better understanding of the dietary patterns that underscore gender-specific SES differences in nutrient intakes. Analyses of the type employed in this study will facilitate the development of interventions aimed at modifying overall eating patterns, rather than specific components of the diet.


Investigating patterns of dietary intake, rather than focusing on individual dietary components, has until recently received relatively little research attention (Hu et al, 1999; Kant, 1996; Nicklas et al, 1989; Randall et al, 1990). However, the analysis of dietary patterns as an approach to investigating links between diet and disease is important, since it recognizes that foods are consumed in many combinations that are likely to be complex, and that nutrient intakes are often highly correlated, with certain nutrients having interactive and synergistic effects (Kant, 1996; Wirfalt et al, 2000). Further, focusing on specific dietary components does not take into account the existence of non-nutrient bioactive substances that may impact on disease risk (Randall et al, 1990). In a practical sense, the examination of dietary patterns is important, since it may more readily enable the identification of those groups in the population at greatest risk of diet-related morbidity and mortality (Huijbregts et al, 1997).

Persons of low socioeconomic status (SES) are one group of concern in terms of nutrition-related health. For example, in previous studies of the Australian population those of lower SES have been shown to consume a greater proportion of their energy as refined sugars (Baghurst et al, 1989), and have a diet higher in fat density (Baghurst et al, 1990; Milligan, et al, 1998; Smith & Owen, 1992; Webb et al, 1999), lower in fibre density (Baghurst et al, 1990; Smith & Owen, 1992; Webb et al, 1999), and lower in density of intake for a range of micronutrients (Baghurst et al, 1990; Milligan et al, 1998; Webb et al, 1999). Socioeconomic differentials in intakes have been observed in other developed countries including the USA (Shimakawa et al, 1994; Wynn, 1987), Sweden (Wallstrom et al, 2000; Wamala et al, 1999), the UK (Bartley et al, 2000; Thompson et al, 1993), and elsewhere in Europe (Irala-Estevez et al, 2000).

While several studies have demonstrated relationships between SES and nutrient intakes, few have assessed the dietary patterns underpinning SES–nutrient relationships. Multivariate methodologies, such as factor analysis, offer potentially useful techniques for describing patterns of dietary intake. Factor analysis is a method that allows exploration and detection of patterns of variables, and offers a means by which to summarize a large number of variables through reduction to a smaller set of underlying factors (Kim, 1975). Factor analysis of food consumption patterns can provide insights into the foods that group together, giving a clearer picture of the patterns of intake. However, only a limited number of studies have utilized multivariate techniques to investigate multiple dietary factors in combination, although recently more studies of this type have emerged.

Several early studies used factor analytic techniques to examine patterns of foods consumed in a single day based on 24 h dietary recall measures (Gex-Fabry et al, 1988; Iizumi & Amemiya, 1986; Schwerin et al, 1981; 1982). For example, Schwerin et al, (1981) conducted factor analyses of 24 h dietary recall data collected in two large population surveys in the USA. Analyses yielded seven different eating patterns, some of which were found to be associated with biochemical deficiencies. However, the use of 24 h recall is not ideal for the assessment of dietary patterns, since there exists substantial day-to-day variability in diet, and 24 h recall data are insufficient to characterize individuals' habitual food intake (Baghurst & Baghurst, 1981).

Food frequency questionnaires (FFQ) are a useful tool for assessing usual dietary intake. Only a limited number of studies have factor analyzed food frequency data. Randall et al (1990) conducted factor analysis of 110 food items with 2255 adults aged 50–70 y. Nine dietary-pattern factors, including ‘salad’, ‘fruit’, ‘dessert’ and ‘staple vegetables’, ‘healthful’ (including vegetable items, poultry, fin fish, and unsweetened cereal), and ‘Southern European’ (including spaghetti, eggplant, macaroni, lamb and potatoes) were identified. Other studies have reported differing numbers and types of dietary pattern factors in varied samples, from two major eating patterns among adult men (Webb et al, 1999); four among elderly Japanese (Kumagai et al, 1999); seven among Mexican American mothers (Wolff & Wolff, 1995); to 17 dietary patterns among adolescents and young adults (Nicklas et al, 1989). However, most of these previous studies used relatively small, non-representative samples, and/or a limited number of food groups.

The aims of the present study were, firstly, to describe the structure of eating patterns among men and women in a large, nationally representative population sample, and secondly, to explore how dietary patterns vary according to SES.



Data were derived from the 1995 Australian National Nutrition Survey (NNS) (Australian Bureau of Statistics (ABS), 1998). NNS participants were recruited from the study population of the 1995 National Health Survey (NHS) (ABS, 1995). The NHS is part of a regular 5-y population survey conducted by the Australian Bureau of Statistics, which collects health status information about the Australian population. Recruitment procedures for the 1995 NHS and NNS surveys are described elsewhere (ABS, 1995, 1998). For the NHS, a stratified multi-stage area sampling technique was used to obtain a random, nationwide sample of 23 800 households. Of households selected to participate, 91.5% households responded, with a total of 57 633 persons interviewed. Of those, 22 562 were selected to participate in the NNS. The sample for the NNS was systematically selected from the NHS private dwelling sample covering urban and rural areas across all states and territories of Australia.

While all persons aged 2 y or more were eligible to participate in the NNS, only persons aged 15 and over were eligible to complete a FFQ. Of the 13 858 eligible participants in the NNS, 10 754 were adults. Pregnant women (n=159) were excluded from analyses. The present study uses data provided by 6680 adults (3111 men and 3569 women) of working age (18–64 y) who completed the FFQ.


Trained interviewers personally interviewed participants in the NHS. At the completion of the NHS interview, selected participants were informed of the NNS and agreement to participate was sought. Those agreeing to participate were instructed how to complete the FFQ by the interview staff. Participants self-completed the FFQ and returned it by mail.

Dietary intake

A FFQ was administered as part of the NNS. The FFQ assessed usual frequency of intake of 100 food and non-alcoholic beverage items over the last 12 months. No information was collected on portion sizes. Each item had a choice of nine frequency categories ranging from ‘Never or less than once a month’ to ‘Six or more times per day’. These were converted to daily equivalent frequencies and entered into a factor analysis. Of the 100 food/beverage items, 99 were included in factor analyses. Soy beverages were omitted from the analysis due to the large proportion (93%) of people who consumed them ‘never or less than once a month’. Alcohol consumption was removed from the analysis, since it is considered an important factor contributing to health and in subsequent research a separate adjustment for alcohol consumption may then be made in addition to the dietary factors.

Socioeconomic status

Employment was used as an index of SES. The measure of employment used in the present analyses was a multi-dimensional item derived empirically through gender-specific factor analyses of demographic and socioeconomic variables included in the National Health Survey (Mishra et al, 2001). Items loading on the employment factor included measures of employment status, occupation, and hours worked. The employment factor was split into tertiles, with the lowest representing the most disadvantaged, and the highest representing the most socioeconomically advantaged. Thus the highest tertile comprised those working full-time in managerial or professional occupations, while the lowest tertile included those who were unemployed, or employed in part-time, labouring or manual occupations.

Statistical analysis

With the sample stratified by gender, exploratory factor analysis using the method of principal components and varimax rotation was performed on the 99 items measuring aspects of dietary intake. Items that cross-loaded on several factors (ie items that had loadings of greater than 0.4 on more than one factor), or had loadings of less than 0.4 on all the factors were subsequently eliminated. Inter-item reliability for each factor was assessed by Cronbach's α coefficients for standardized variables. Kaiser's measure of sampling adequacy (MSA) was used to quantify the degree of intercorrelations among the items and the appropriateness of factor analysis is also reported (Hair et al, 1997). In order to assess the stability of the factors, factor structures were compared with the results from the samples after they had been randomly split into two subsamples and the analyses repeated on each half.

Analysis of variance, controlling for age, was performed separately for each of the standardized factor scores for men and women to examine differences in dietary intakes by tertile of employment status (high, middle, low). All analyses were performed using SAS (SAS Institute Inc., 1989).


Table 1 shows the characteristics of the study participants by gender. Approximately 70% of the participants were between 25 and 54 y old. Compared with women, a higher proportion of men were managers, administrators or professionals, smokers, and/or consumed alcohol more than once a week. A higher proportion of women than men were on a weight-reduction or fat-modified diet, and/or consumed low-fat dairy foods frequently.

Table 1 Characteristics of the sample by gender (n=8667)

Exploratory factor analysis of dietary items for men led to the deletion of 35 items due to cross-loadings (capsicums, carrots, rice, stir fry vegetables, tea, fruit juice), or a loading of less than 0.4 (cream, low energy soft drink, milk, other seafoods, jam/marmalades, muesli, yoghurt, confectionary, lentils, mayonnaise, muffins, sweet corn, baked beans, chocolate biscuits, cooked porridge, flavoured milk, mushrooms, pasta, salami, soybeans, cordial, fruit drink, liver, nuts, water, cottage/ricotta cheeses, hot chips, low-energy cordial, peanut butter). Among women, two dietary items were deleted due to cross-loadings (pineapple, plain sweet biscuits) while 31 items were omitted due to a loading of less than 0.4 (cream, milk, sweet potato, wholemeal bread, bacon, canned fish, jam/marmalades, yoghurt, eggs, lentils, mayonnaise, muffin, pizza, roast poultry, sweet corn, stir fry vegetables, baked beans, cooked porridge, flavoured milk, salami, tea, cordial, liver, mixed dishes with poultry, nuts, sausages, water, chopped lamb, mixed lamb, offal, white bread).

Results of the factor analysis revealed 15 distinct, interpretable factors for men and 15 for women (Tables 2 and 3). The eigenvalues (ie the amount of variance explained by each of the factors; the larger the eigenvalue, the more variance is explained by that factor) were examined. Eigenvalues corresponding to these 15 principal components were greater than one, suggesting that the 15-dimensional model was appropriate. The overall sampling adequacies for the main factors were 0.78 and 0.81 for men and women, respectively. For men, the Cronbach's α coefficients ranged from 0.36 to 0.81, and 0.28 to 0.86 for women. In both cases this indicated moderate to excellent internal reliability. In both groups, the 15 factors accounted for 50% of the total variance. Further support for the factor structure was obtained when the analyses were repeated on these split samples and the same structure was found for each gender group.

Table 2 Factor loadings, cumulative percentage of variation and (internal reliabilitya) estimated from responses from 4167 men aged 18–64 y
Table 3 Factor loadings, cumulative percentage of variation and (internal reliabilitya) estimated from responses from 4500 women aged 18–64 y

Provisional names were assigned to the factors for men and women (Tables 2 and 3). A number of similarities in dietary patterns were observed between men and women. For example, ‘salad’ was a major factor for both men and women. Similarly, ‘takeaways’ was another food pattern common to both genders. However, the factor analyses also revealed a number of gender differences. ‘Pasta, rice and other mixed dishes’, ‘ethnic vegetables’, ‘fruit and vegetable juice’, ‘fish and seafoods’, ‘chocolate and confectionary’, and ‘low energy drinks’ emerged amongst women but not amongst men. Conversely, ‘protein foods’ and ‘wholemeal bread’ were found for men, but not women.

Those dietary factors that were significantly associated with employment status are presented in Table 4. Compared with men of high employment status, men of lower status more frequently consumed ‘tropical fruits’, ‘protein foods’, and ‘offal and canned fish’. Men of high employment status more often ate ‘breakfast cereals’ and ‘wholemeal bread’ than did men of lower status, while men in the middle status category consumed ‘traditional vegetables’ and ‘takeaways’ more often than other men. There were fewer differences in dietary intake patterns by employment status among women than there were among men. Compared with women of high employment status, women of in the lower status categories more often ate ‘traditional vegetables’, ‘meat dishes’ and ‘pasta, rice and other mixed foods’. High status women more frequently ate ‘ethnic vegetables’ and ‘breakfast cereal/muesli’ than did other women.

Table 4 Age-adjusted mean standardized factor score (and standard deviation) and significance values from analysis of variance of food factors by employment status (only factors for which there are significant differences are shown)


This study is one of only a few population studies of dietary intake to specifically examine eating patterns, rather than nutritional profiles or deficiency/excess of particular dietary components. As far as we are aware, it is the first to examine the dietary profiles of Australian adults through factor analysis of food frequency data. The factor analytic approach used has identified patterns of food groups as they are consumed in the Australian population. This is of practical importance, since interventions aimed at improving diet attempt to influence overall eating patterns, rather than focusing only on specific foods or nutrients. As such, an understanding of patterns of dietary intake is important in order to plan and implement public health initiatives.

The food groups identified by factor analysis are indicative of food items that are consumed together or items that substitute for one another but are consumed with similar frequency. The fact that 15 distinct food groupings were found for men and women reveals the wide diversity of eating patterns among adult Australians. In spite of this common ground, there were a number of clear gender differences in eating patterns, both in terms of the items within food groups and their frequency of consumption. Low energy drinks formed a distinct food group only among women, while men consumed ‘takeaways’ more frequently. For women, similar food items often formed coherent groups such as ‘traditional vegetables’ or ‘fruits’, whereas for men food items in each of these were split over separate groups (eg ‘salad vegetables’, ‘traditional vegetables’, ‘other vegetables’). Overall, the food groups for women appeared more clearly structured into food types, which may reflect greater awareness or deliberate determination of food eating patterns based on food types.

Several of the dietary patterns identified in this study are consistent with those found for previous studies using factor analysis of dietary intake. For example, the factors are similar to several of those reported by Randall et al (1990), including ‘salad’, ‘fruit’, ‘dessert’ and ‘staple vegetables’. However, certain factors reported by Randall and colleagues, such as ‘healthful’ and ‘Southern European’ were not replicated in the present study. Further, Randall's nine factors accounted for only 22 and 23% of the total variance in male and female dietary intake, respectively. In the present study, the 15 dietary-pattern factors identified accounted for a much higher proportion of variance (around 50% for both males and females).

Certain dietary patterns among women, and in particular among men, differed by SES in the present study. Several of these differences are consistent with limited previous findings of SES gradients in food patterns. For example, one study of Australian adults found that lower SES groups reported eating less wholemeal or brown bread (Steele et al, 1991), and less fresh fruit and vegetables. However this earlier study used only a brief (18-item) measure of dietary intake. The present study, which incorporated a much more comprehensive food frequency measure, showed that consumption of different types of fruits and vegetables varied differentially by SES. For example, while low SES women were less likely to consume ‘ethnic vegetables’ (eg zucchini, capsicum), they were more likely than higher SES women to consume traditional vegetables (eg peas, carrots). These findings highlight the complexity of the associations between SES and different components of dietary intake. Results are also consistent with past studies showing socioeconomic gradients in nutrient intakes. For example, the finding that lower SES groups report less frequent consumption of wholemeal breads or breakfast cereals is consistent with past reports of proportionately lower fibre intakes among these groups (Baghurst et al, 1990; Smith & Owen, 1992; Webb et al, 1999).

There were several limitations with the present study. Self-report bias is a problem inherent in studies of dietary intake (Lissner et al, 2000), and responses in the present study may have been confounded by selective under- or over-reporting of particular food items. In addition, portion size of food items was not assessed in the FFQ. Finally, the factor analytic method is limited in that the patterns of consumption of certain food items that were omitted from analyses due to low loadings, or cross-loading on more than one factor, cannot be determined. For example, rice may comprise an important part of dietary intake for men, but in this study it was omitted from the analyses as it was grouped with more than one major factor (cross-loaded). Even though factor structure is dependent on the number of items initially assessed, the large number of items included here should be representative of the broad range of foods consumed in Australia.

The results of the present study have important implications for nutrition promotion. Knowledge of the eating patterns that contribute to gender and SES differences in nutrient intakes is important for efforts aimed at improving the nutrition-related health of high-risk groups. For example, the findings of this study suggest that low SES groups, particularly men, could be targeted in campaigns aimed at encouraging consumption of a wider range of less traditional vegetables, with which low SES groups may be less familiar. Further analyses of the type employed in this study will facilitate such research, and lead to the development of interventions aimed at modifying overall eating patterns, rather than specific components of the diet.


  1. Australian Bureau of Statistics. 1995 National Health Survey: Users' Guide Catalogue no. 4363.0 Canberra: Australian Bureau of Statistics

  2. Australian Bureau of Statistics. 1998 National Nutrition Survey: User's Guide, 1995 Catalogue no. 4801.0 Canberra: Australian Bureau of Statistics

  3. Baghurst K, Baghurst P . 1981 The measurement of usual dietary intakes in individuals and groups Trans. Menzies Found. 3: 139–160

    Google Scholar 

  4. Baghurst K, Record S, Syrette J, Crawford D, Baghurst P . 1989 Intakes and sources of a range of dietary sugars in various Australian populations Med. J. Aust. 151: 512–518

    CAS  PubMed  Google Scholar 

  5. Baghurst K, Record S, Baghurst P, Syrette J, Crawford D, Worsley A . 1990 Socioeconomic determinants in Australia of the intake of food and nutrients implicated in cancer aetiology Med. J. Aust. 153: 444–452

    CAS  PubMed  Google Scholar 

  6. Bartley M, Fitzpatric R, Firth D, Marmot M . 2000 Social distribution of cardiovascular disease risk factors: change among men in England 1984–1993 J Epidemiol. Community Health 54: 806–814

    CAS  Article  Google Scholar 

  7. Gex-Fabry M, Raymond L, Jeanneret O . 1988 Multivariate analysis of dietary patterns in 939 Swiss adults: sociodemographic parameters and alcohol consumption profiles Int. J. Epidemiol. 17: 548–555

    CAS  Article  Google Scholar 

  8. Hair JF Jr, Anderson RE, Tatham RL, Black WC . 1997 Multivariate Data Analysis with Readings 4th edn New Jersey: Prentice Hall

    Google Scholar 

  9. Hu FB, Rimm E, Smith-Warner SA, Feskanich D, Stampfer MJ, Ascherio A, Sampson L, Willett WC . 1999 Reproducibility and validity of dietary patterns assessed with a food-frequency questionnaire Am. J. Clin. Nutr. 69: 243–249

    CAS  Article  Google Scholar 

  10. Huijbregts P, Feskens E, Rasanen L, Fidanza F, Nissinen A, Menotti A, Kromhout D . 1997 Dietary pattern and 20 y mortality in elderly men in Finland, Italy, and the Netherlands: longitudinal cohort study Br. Med. J. 314: 13–18

    Article  Google Scholar 

  11. Iizumi H, Amemiya T . 1986 Eleven-year follow-up of changes in individuals' food consumption patterns Int. J. Vit. Nutr. Res. 56: 399–409

    CAS  Google Scholar 

  12. Irala-Estevez J, Groth M, Johansson L, Oltersdorf U, Prattala R, Martinez-Gonzalez M . 2000 A systemic review of socio-economic differences in food habits in Europe: consumption of fruit and vegetables Eur. J. Clin. Nutr. 54: 706–714

    CAS  Article  Google Scholar 

  13. Kant AK . 1996 Indexes of overall diet quality: a review J. Am. Diet. Assoc. 96: 785–792

    CAS  Article  Google Scholar 

  14. Kim J . 1975 Factor analysis In Statistical Package for the Social Sciences ed. NH Nie, CH Hull, JG Jenkins, K Steinbrenner & DH Bent New York: McGraw-Hill

    Google Scholar 

  15. Kumagai S, Shibata H, Watanabe S, Suzuki T, Haga H . 1999 Effect of food intake pattern on all-cause mortality in the community elderly: a 7-year longitudinal study J. Nutr. Health Aging 3: 29–33

    CAS  PubMed  Google Scholar 

  16. Lissner L, Heitmann BL, Bengtsson C . 2000 Population studies of diet and obesity Br. J. Nutr. 83: S21–S24

    CAS  Article  Google Scholar 

  17. Milligan R, Burke V, Beilin L, Dunbar D . 1998 Influence of gender and socio-economic status on dietary patterns and nutrient intakes in 18 y old Australians Aust. NZJ. Public Health 22: 485–493

    CAS  Article  Google Scholar 

  18. Mishra GD, Ball K, Dobson AJ, Byles JE, Warner-Smith P . 2001 The measurement of socioeconomic status: investigation of gender- and age-specific indicators in Australia National Health Survey '95. Social Indicators Research 56: 73–89

    Article  Google Scholar 

  19. Nicklas TA, Webber LS, Thompson B, Berenson GS . 1989 A multivariate model for assessing eating patterns and their relationship to cardiovascular risk factors: the Bogalusa Heart Study Am. J. Clin. Nutr. 49: 1320–1327

    CAS  Article  Google Scholar 

  20. Randall E, Marshall JR, Graham S, Brasure J . 1990 Patterns in food use and their associations with nutrient intakes Am. J. Clin. Nutr. 52: 739–745

    CAS  Article  Google Scholar 

  21. SAS Institute Inc.. 1989 SAS/STAT User's Guide (Version 6, 4th edn, Vol 2) Cary, NC: SAS Institute Inc.

  22. Schwerin HS, Stanton JL, Riley AM Jr, Schaefer AE, Leveille GA, Elliott JG, Warwick KM, Brett BE . 1981 Food eating patterns and health: a reexamination of the Ten-State and HANES 1 surveys Am. J. Clin. Nutr. 34: 568–555

    CAS  Article  Google Scholar 

  23. Schwerin HS, Stanton JL, Smith JL, Riley AM, Brett BE . 1982 Food, eating habits, and health: a further examination of the relationship between food eating patterns and nutritional health Am. J. Clin. Nutr. 35: 1319–1325

    CAS  Article  Google Scholar 

  24. Shimakawa T, Sorlie P, Carpenter M, Dennis B, Tell G, Watson R, Williams O . 1994 Dietary intake patterns and sociodemographic factors in the atherosclerosis risk in communities study. ARIC study investigators Prev. Med. 23: 769–780

    CAS  Article  Google Scholar 

  25. Smith A, Owen N . 1992 Associations of social status and health-related beliefs with dietary fat and fiber densities Prev. Med. 21: 735–745

    CAS  Article  Google Scholar 

  26. Steele P, Dobson A, Alexander H, Russell A . 1991 Who eats what? A comparison of dietary patterns among men and women in different occupational groups Aust. J. Public Health 14: 286–295

    Google Scholar 

  27. Thompson R, Margetts B, Speller V, McVey D . 1999 The health education authority's health and lifestyle survey 1993: who are the low fruit and vegetable consumers? J. Epidemiol. Community Health 53: 294–299

    CAS  Article  Google Scholar 

  28. Wallstrom P, Wirfalt E, Janzon L, Mattisson I, Elmstahl S, Johansson U, Berglund G . 2000 Fruit and vegetable consumption in relation to risk factors for cancer: a report from the Malmo Diet and Cancer study Public Health Nutr. 3: 263–271

    CAS  Article  Google Scholar 

  29. Wamala S, Mittleman M, Schenck-Gustafsson K, Orth-Gomer K . 1999 Potential explanations for the educational gradient in coronary heart disease: a population-based case-control study of Swedish women Am. J. Public Health 89: 315–321

    CAS  Article  Google Scholar 

  30. Webb K, Schofield W, Lazarus R, Smith W, Mitchell P, Leeder S . 1999 Prevalence and socio-demographic predictors of dietary goal attainment in an older population Aust. N.Z. J. Public Health 23: 578–584

    CAS  Article  Google Scholar 

  31. Wirfalt E, Mattisson I, Gullberg B, Berglund G . 2000 Food patterns defined by cluster analysis and their utility as dietary exposure variables: a report from the Malmo Diet and Cancer Study Public Health Nutr. 3: 159–173

    CAS  Article  Google Scholar 

  32. Wolff CB, Wolff HK . 1995 Maternal eating patterns and birth weight of Mexican American infants Nutr. Health 10: 121–134

    CAS  Article  Google Scholar 

  33. Wynn A . 1987 Inequalities in nutrition Nutr. Health 5: 79–94

    CAS  Article  Google Scholar 

Download references


Dr Kylie Ball is supported by a Public Health Research Fellowship from the National Health and Medical Research Council (ID 136925). Dr David Crawford is supported by a Nutrition Research Fellowship from the National Heart Foundation of Australia.

Author information



Corresponding author

Correspondence to K Ball.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Mishra, G., Ball, K., Arbuckle, J. et al. Dietary patterns of Australian adults and their association with socioeconomic status: results from the 1995 National Nutrition Survey. Eur J Clin Nutr 56, 687–693 (2002).

Download citation


  • dietary patterns
  • socioeconomic status
  • factor analysis
  • population study

Further reading


Quick links