Diet Quality and Its Potential Cost Savings

  • Clare Collins
  • Alexis Hure
  • Tracy Burrows
  • Amanda Patterson
Part of the Nutrition and Health book series (NH)


Surprisingly, only limited studies have examined the relationship between diet quality and the health care costs to government or to individuals. Additionally, very few studies have examined whether diet quality is able to predict future health care costs.


Diet index Diet quality Diet score Health costs Health service usage Medicare Nutrition survey 



Australian Institute of Health and Welfare


Australian Longitudinal Study on Women’s Health


Australian Recommended Food Score


Disability-adjusted life years


Dietary Questionnaire for Epidemiological Studies


Diet Quality Index


Diet Quality Score


Enhanced Primary Care


Food Frequency Questionnaire


Healthy Diet Indicator


Healthy Eating Index


Healthy Food Index


Health Maintenance Organization


Mediterranean Diet Score


Recommended Food Score


Survey in Europe on Nutrition and the Elderly; a Concerted Action

Key Points

  • Limited studies have determined the relationship between diet quality and health care costs. The current analysis investigates the relationship between Medicare health care costs and claims in Australia with diet quality, in a nationally representative sample of mid-aged Australian women over the 10-year period from 2001 to 2010. Health care utilization data was obtained from Medicare and diet quality was calculated using the Australian Recommended Food Score (ARFS). The ARFS is derived from Food Frequency Questionnaire (FFQ) responses to foods that are consistent with national recommendations detailed in the Dietary Guidelines for Australian Adults.

  • It was found that consuming a greater variety of vegetables predicted lower 10-year cumulative Medicare Charges, Benefit, Gap and a lower number of claims. However, for fruit, dairy and healthy fats component scores this relationship was in the opposite direction. Further evaluations in other studies are needed and will provide the basis for modelling future cost savings and may help to identify appropriate dietary targets associated with reduced health care costs.


Surprisingly, only limited studies have examined the relationship between diet quality and the health care costs to government or to individuals. Additionally, very few studies have examined whether diet quality is able to predict future health care costs.

Optimal dietary quality in the context of this review has been defined as eating patterns that reflect greater adherence to national dietary recommendations. Diet quality can be measured by diet quality or diet index scores. There are over 25 indexes to date that have been developed to assess overall diet quality and/or variety. Wirt and Collins [1] reviewed these in 2009 and found that the major indexes included the Healthy Eating Index (HEI) [2], the Healthy Diet Indicator (HDI) [3], the Healthy Food Index (HFI) [4], the Recommended Food Score (RFS) [5], the Diet Quality Index (DQI) [6], the Diet Quality Score (DQS) [7], and the Mediterranean Diet Score (MDS) [8]. Construction of these diet quality indexes follows three major scoring approaches: (1) based on food groups or specific foods, (2) based on nutrient intakes or (3) derived from combinations of foods and nutrient intakes [9].

A review of the relationship between these diet quality scores and health-related outcomes demonstrated that higher quality eating patterns were associated with lower self-reported indices of health care usage and lower morbidity and mortality from chronic disease [1, 10, 11].

However, whether higher diet quality translates into lower health care costs over time has rarely been examined. One key Australian report evaluated the relationship between fruit and vegetable intake (as a measure of diet quality) and burden of disease. This Australian Institute of Health and Welfare report (AIHW-2003) estimated that 2.1 % of the total burden of disease could be attributed to inadequate fruit and vegetable consumption, making it seventh in a ranking of 14 risk factors studied [12], which together explained 32.2 % of the total burden of disease and injury. Eating an adequate amount of fruits and vegetables has been shown to prevent chronic diseases such as cardiovascular disease, type 2 diabetes and some cancers [12]. The AIHW report found that 69 % of the burden from low fruit and vegetable consumption was due to ischaemic heart disease and that two-thirds of this was experienced by males. Further, the report estimated that 81 % of the burden from low fruit and vegetable consumption was due to mortality, and that the absolute burden from low consumption peaked between 60 and 80 years of age [12].

We previously reported the results of the relationship between diet quality and health service use in a cross-sectional analysis of the Australian Longitudinal Study on Women’s Health (ALSWH) for over 9,000 mid-aged women [13]. Among these women, a higher Australian Recommended Food Score (ARFS, the index of diet quality used) was associated with better self-reported health status (excellent, very good, good or fair/poor), a lower number of visits to a general medical practitioner in the previous year and fewer consultations with a specialist medical practitioner in the previous year. Those with a higher ARFS also had higher intakes of micronutrients and a lower percentage of energy from total or saturated fat [13].

Kant and Schatzkin [11] examined the 10-year association between diet quality (measured by nutrient intakes) with self-reported doctor-advised chronic medical conditions, for a nationally representative sample of adults (>25 years, n = 7,207) in the Continuing Survey of Food Intakes. A greater number of self-reported medical conditions were associated with an increased risk of not consuming 100 % of the Recommended Dietary Allowance (RDA) for some nutrients (vitamins E, B12 and calcium, zinc and iron, P < 0.05). Women were more likely to report poor nutrient intakes than men [11].

In the SENECA (Survey in Europe on Nutrition and the Elderly; a Concerted Action) study on older persons from seven European countries, Haveman-Nies et al. [14] examined the health status and lifestyle behaviours of 216 men and 264 women born between 1913 and 1918, for the 10-year period (1988–1999). Self-rated health and self-care ability declined in men and women with both healthy and unhealthy lifestyle habits during follow-up. There was no relationship between having a healthy, Mediterranean-style diet and deterioration in health status. Inactive and smoking persons did have an increased risk for a decline in health status, compared with active and non-smoking people, and the authors concluded that both physical activity and being a non-smoking delayed deterioration in health status for older people.

Cost-effectiveness studies have modelled the potential cost savings of implementing strategies to promote healthful eating patterns and prevent disease development. Cobiac et al. [15] evaluated the cost-effectiveness of interventions to promote fruit and vegetable consumption to assess the degree to which promotion strategies could potentially reduce population disease burden. They identified 23 interventions that promoted fruit and vegetable intake in adults that also had sufficient information to model the health impacts in disability-adjusted life years (DALY). They modelled the intervention costs and potential cost savings from averting disease treatment, and the cost-effectiveness over the population’s lifetime, from a health sector perspective. They found that only five of the 23 interventions had less than a $50,000 per DALY cost-effectiveness threshold and that the most effective intervention could avert only 5 % of the disease burden attributed to insufficient fruit and vegetable intake. They concluded that further evaluation of population level interventions was needed. While observational studies and modelling estimates do provide a clear rationale for trialling interventions to improve diet quality as a strategy to reduce health care and medical costs, evaluation in real-life settings and databases provides important data in relation to true health care costs.

Others have assessed the cost-effectiveness of interventions to improve diet in specific at risk populations groups. Patrick et al. [16] randomized about 5,000 health maintenance organization (HMO) older Medicare beneficiaries to receive a 2-year preventive service benefit package that included one component aimed at improving diet quality, as measured by percentage energy from fat and grams from daily fibre intakes, or to usual care. At the 2- and 4-year follow-ups, the treatment group reportedly participated in more exercise and reduced their percentage fat intake more than the control group. Although they observed an unexpected greater number of deaths in the treatment group, which they attributed to more individuals over the age of 75 years being randomized to the intervention group, the surviving treatment group participants reported higher satisfaction with health, less decline in self-rated health status, and fewer depressive symptoms than surviving control participants. However, the intervention did not yield a lower cost per quality adjusted life year.

Wolf et al. [17] evaluated, in a 12-month randomized controlled trial, the costs associated with provision of a lifestyle intervention programme on health care expenditure in a high-risk obese population (n = 147) with type 2 diabetes. The intervention group received individual and group education, support and referrals to registered dietitians, while the usual-care group received educational materials only. The net cost of the intervention was $328 per person per year, but the overall mean health plan costs (including the $328) was $3,586 lower in intervention group compared to the usual care group (95 % CI −$8,036 to −$25, P < 0.05). The difference was driven by group differences in medical costs (−$3,316, 95 % CI −$7,829 to −$320, P < 0.05), with significantly fewer inpatient admissions for the intervention group (prevalence 2.8 % in the intervention group vs. 22.5 % in controls, P < 0.001). Authors concluded that adding a modest-cost, registered dietitian-led lifestyle case-management intervention to usual medical care did not increase health care costs, but resulted in modest cost savings for obese patients with type 2 diabetes.

Pavlovich et al. conducted a systematic review of randomized controlled trials on the cost-effectiveness of the provision of outpatient nutrition services [18]. Of the 13 studies included there was relatively consistent support for the cost-effectiveness of nutrition services. Provision of nutrition services was cost-effective in reducing serum cholesterol levels (for example, $20–1,268/mmol/L decrease in serum low-density lipoprotein level), for weight loss ($2.40–10 per pound lost), and in reducing blood glucose ($5/mmol/L decrease). This was evident in target populations with diabetes mellitus and hypercholesterolemia as well as for those without such diagnoses. The authors noted that due to limitations in trial quality and differences in cost perspectives though, that more randomized controlled trials addressing these omissions were needed.

While these approaches provide important insights into potential cost savings through optimizing dietary quality, we have only been able to locate a limited number of studies internationally that have prospectively examined the relationship between diet quality and health care costs. Two studies have examined the national government Medicare charges in relation to aspects of dietary intake. Daviglus et al. [19] looked at US Medicare charges over a 15-year period in relation to fruit and fruit plus vegetable intakes in mid-aged men from the Chicago Western Electric Study, while we examined Australian Medicare costs incurred over a 6-year period by mid-aged women (n > 10,000) from the ALSWH in relation to diet quality as measured by the ARFS [20].

Daviglus et al. demonstrated that in 1,063 men from the Chicago Western Electric Study, those in the highest tertile of fruit and fruit plus vegetable intakes during middle age had lower mean annual and cumulative Medicare charges over a 15-year period [19]. They found that, after adjusting for baseline age, education, total energy intake and baseline risk factors, high fruit and vegetable intakes were associated with lower mean annual and cumulative Medicare charges, including total charges and cardiovascular disease and cancer-related charges, although the P value for trend ranged from 0.019 to 0.862. While many of the results were not statistically significant, the trends do suggest that having higher fruit and vegetable intakes earlier in adulthood has the potential to lower health care costs in older age, as well as improve health status.

While we hypothesized lower costs would be associated with higher diet quality, our Australian data demonstrated a positive association between having a higher ARFS (i.e. better diet quality) and higher cumulative Medicare (Australia) costs. Although those in the highest quintile of ARFS incurred higher Medicare costs, the number of Medicare claims for this quintile was lower than for the lowest quintile of ARFS for the period 2002–2007, P = 0.002.

Firstly, it is important to note that there are some differences between the national Medicare programmes in the USA and Australia. In the USA, Medicare is a social insurance programme administered by the government, which provides health insurance cover to people who are aged 65 and over, or who meet specific criteria. Medicare (USA) covers hospital-related inpatient and outpatient services, skilled nursing facility services and charges related to outpatient services, including emergency room visits, clinic and ambulatory surgery, laboratory tests, radiography, rehabilitation therapy, radiation therapy and renal dialysis. This contrasts with Medicare Australia, who administer a universal health care programme called Medicare for the Australian government. Medicare (Aus) is funded by income tax and an income-related Medicare levy. Medicare (Aus) is the largest source of primary health care spending and covers scheduled fees for out-of-hospital services for doctors (including specialists), tests and examinations by doctors, X-rays and pathology tests, eye tests performed by optometrists, most surgical and other therapeutic procedures performed by doctors, some surgical procedures performed by approved dentists, specified items under the Cleft Lip and Palate Scheme, and specified items for allied health services as part of the Enhanced Primary Care (EPC) programme. In-hospital services are only covered by Medicare (Aus) for treatment as a public patient by doctors and specialists nominated by the hospital. Medicare (Aus) also administers a separate Pharmaceutical Benefits Scheme (PBS), which makes a range of prescription medicines available at affordable prices to Australian residents, through government subsidy.

While the above two studies are the only ones we have found that actually calculate costs, van Baal et al. [21], in three cohorts from the Netherlands, estimated the annual and lifetime medical costs attributed to obesity for people aged 20 years at baseline. They compared obese people to those who were smokers and those with a healthy lifestyle (including following a high quality diet). The obese group were modelled as having the highest annual health expenditure, followed by smokers, but lifetime health expenditure was highest among healthy-living individuals, due to their longer life expectancy. The authors concluded that while obesity prevention is an important and cost-effective way of improving public health, it is not the only strategy for reducing health expenditures.

Other than these studies, we were not able to locate any other studies that have examined associations between diet quality and subsequent cost of health service use. This relationship needs further study with both costs and claims monitored over time to examine patterns of health care usage and costs in the longer term for those with the dietary patterns that adhere more closely to national recommendations.

Evaluating the Association Between Diet Quality and 10-Year Cumulative Medicare Costs

On this basis, we have recently extended the 1946–1951 (mid-aged) ALSWH cohort follow-up period to 10 years, to examine whether higher diet quality, as measured by the ARFS was still associated with higher health care costs but lower number of claims.

The data for this analysis comes from ALSWH, which was established to investigate multiple factors affecting the health and wellbeing of women over a 20-year period. Women in three age groups (“younger” 18–23, “mid-aged” 45–50 and “older” 70–75 years) were randomly selected from the national health insurance database, Medicare (Aus) that includes all permanent residents of Australia, with over-representation of women living in rural and remote areas. The methods have been previously published [22, 23, 24].

Diet Quality and Health Care Costs in Mid-Aged Women in Australia

Data used in this analysis have been derived from the mid-aged cohort of the ALSWH. Survey 1 (n = 13,716) was conducted in 1996 and the respondents have been shown to be broadly representative of the national population of women in the target age groups [22]. Survey 2 (n = 12,338) was conducted in 1998 and Survey 3 (n = 11,228) was conducted in 2001. The response rate for Survey 3 of the mid-aged cohort was 83 % of women who had completed Survey 1 and had not died or become too ill to complete further surveys. The non-respondents included those who did not complete Survey 3 (7.4 %), withdrew from the study completely (2.8 %) or could not be contacted (6.8 %) [24]. Of the women who completed Survey 3 (then aged 50–55 years), 11,194 completed a usable Food Frequency Questionnaire (FFQ).

Assessment of Dietary Intake

Dietary intake was assessed using the Dietary Questionnaire for Epidemiological Studies (DQES) FFQ. The DQES asks respondents to report their usual consumption of 74 foods and six alcoholic beverages over the preceding 12 months using a 10-point frequency option from “never” up to “three or four times per day”. Portion size photographs are used to adjust the serve size for vegetables, meat and casseroles. Additional questions are asked about total number of daily serves of fruits, vegetables, bread, dairy products, eggs, fat spreads and sugar, as well asking the type of bread, dairy products and fat spreads used. Nutrient intakes are computed from NUTTAB 1995, a food composition database of Australian foods [25], using software developed by the Cancer Council of Victoria. Both the development of the DQES [26] and its validation in mid-aged Australian women have been previously reported [27]. FFQs with greater than four missing values were discarded.

Australian Recommended Food Score

The ARFS was modelled on the RFS by Kant and Thompson [28] and has been previously described [13]. Briefly, it is calculated based on DQES items consistent with national recommendations in the Dietary Guidelines for Australian Adults [29]. In brief, items consumed less than once a week are scored zero and those consumed once a week or more score one. Additional points are awarded for type and variety of core foods consistent with national dietary intake recommendations. A maximum of two points was added for alcohol consumption: one point for moderate frequency and the second point for moderate quantity, when they drank alcohol. The maximum ARFS is 74.

Medicare Data

The health services utilization and cost data was provided by Medicare Australia. In the third survey of the mid-aged cohort, 7,225 out of the 11,226 women (64 %) gave consent for linkage of their Medicare data to their survey. There were significant but small differences between consenters and non-consenters by area of residence, and those consenting to Medicare linkage tended to be better educated, more likely to be able to manage on their available income and more likely to say their health was excellent, very good or good [23, 24]. Only ALSWH respondents who consented to link their Medicare data were used in this analysis. There were 6,781 women who had given both Medicare consent and had an ARFS.

All Medicare data were collected for the 10 years from 2001 to 2010. Annual data was totalled for each woman to determine the amount the women spent on Medicare health care and the “Number of claims”. Women with no claims for a particular year had values of zero imputed into their claims and cost fields for that year, as Medicare Australia only provides data for women for which there are charges recorded. The “Charge” item is the total cost of the treatment. There are two other Medicare variables, the “Benefit” and the “Gap”. The Benefit is what was paid back to the patient, while the Gap is the difference between the Charge and the Benefit (i.e. what the patient paid out of their own pocket). The Charge is almost always higher than the Benefit.

Those with the highest (n = 67) and lowest 1 % (n = 68) of charges were excluded from data analysis to avoid extreme values (<$634 and >$49,395). Therefore, the sample included data for n = 6,646 women. Level of education was recorded at Survey 1 and was missing for n = 39, hence a total sample of n = 6,607 are included in the multivariate regressions with adjustment for baseline area of residence and education. The Medicare data used in this study were from the Medicare Benefits Schedule and did not include data from the PBS.

The analysis followed the same approach as that described previously [20]. Data manipulation and statistical analyses were performed using Intercooled Stata, version 11 (StataCorp LP, College Station, TX, USA). The distribution of the Medicare data was highly skewed to the right; therefore, median values have been presented. The ARFS was normally distributed. ARFS quintiles were generated using the xtile function in Stata. Generalized linear modelling was performed with area of residence and level of education at baseline used as covariates, to adjust for the sampling frame and for socioeconomic status. P-values <0.01 were considered statistically significant, due to the large sample size.

The ARFS was normally distributed while the Medicare costs and claims were not. The median 10-year (2001–2010) cumulative Medicare Claims and Charge, Benefit and Gap expenses were examined by quintiles of ARFS at Survey 3, with a median (interquartile range, IQR) ARFS of 21 (6) for quintile 1 and 44 (5) for quintile 5 (Table 4.1). After adjustment for area of residence and education at baseline, ARFS quintile 5 had significantly fewer Claims (P = 0.008) compared to ARFS quintile 1. ARFS quintiles 3 and 5 had significantly higher Gap expenses compared to quintile 1; P = 0.004 and P < 0.001, respectively. ARFS quintile was not a significant predictor of Charge (P = 0.583) or Benefit (P = 0.369).
Table 4.1

Median 10-year (2001–2010) cumulative Medicare claims and costs ($AUS) for mid-aged Australian women by quintile of ARFS, where 1 = lowest and 5 = highest quintile

ARFS quintile


ARFS median


Number of claims


Charge ($)


Benefits ($)


Gap ($)










































































aGeneralized linear modelling with adjustment for area of residence and education at baseline shows ARFS quintile is a significant (P < 0.01) predictor of number of Medicare Claims (quintile 5) and Gap expenses (quintiles 3 and 5), compared to quintile 1

ARFS Component Scores

The individual ARFS component scores were then tested as the predictors of 10-year cumulative Medicare claims and costs. Table 4.2 shows that the number of Medicare Claims and Charges was inversely associated with consuming a greater variety of vegetables, independent of total ARFS at Survey 3, and area of residence and education at baseline. There was a positive linear relationship between Medicare claims and charges and the ARFS component scores for fruit, dairy and fat. All other ARFS component scores were not associated with Medicare claims and costs (P > 0.01).
Table 4.2

Coefficients and 95 % confidence intervals for ARFS components that significantly predicta 10-year (2001–2010) cumulative Medicare claims or costs ($AUS) for mid-aged Australian women (n = 6,607)

ARFS component

Maximum points

Coefficients (95 % confidence intervals)

Number of claims

Charge ($)

Benefits ($)

Gap ($)



−2.3 (−3.5, −1.1)

−154 (−233, −76)

−100 (−157, −43)

−54 (−82, −36)



 2.2 (0.8, 3.6)

 162 (72, 252)

 112 (47, 177)

 50 (18, 82)



 6.4 (3.3, 9.4)

 397 (202, 593)

 268 (126, 409)

129 (60, 198)




−98 (−164, −33)



 9.1 (2.8, 15.4)

 672 (270, 1,073)

 421 (131, 712)

250 (108, 393)

aGeneralized linear modelling with adjustment for area of residence and education at baseline, and total ARFS at Survey 3. All significant at P < 0.01

bNon-significant Medicare items and ARFS components are not shown

How Diet Quality Impacts on Health Care Costs

This analysis extends the cumulative Medicare costs and claims in Australia in relation to diet quality in mid-aged Australian women to 10 years. Unlike our previous 5-year cumulative data, which demonstrated higher Medicare costs associated with higher diet quality, this 10-year cumulative data show that Medicare charge and benefit are not related to ARFS (Table 4.1). However, the Medicare Gap, the amount paid by an individual, does increase significantly with higher diet quality (ARFS). This relationship is in the opposite direction to what you would expect, and the 10-year cumulative mean difference for Gap payments from the bottom to the top quintile is substantially greater at $AUS500. This relationship is likely to be confounded by charges incurred for routine screening services related to breast and/or cervical cancer, as those likely to have higher diet quality may also have more routine screening procedures. Interestingly, as for the 5-year data, the total number of claims for health care services over the 10-year period is significantly lower in the highest quintile of ARFS compared to the lowest, by a total of 10 claims. It will be important to revisit this relationship in another 5 years to examine whether there is a point at which having consumed a higher quality diet over an extended period that it does lead to Medicare savings in terms of cumulative health care costs.

When the ARFS sub-scales were examined, there was a statistically significant relationship with many of the sub-scales scores (Table 4.2). While having greater variety of vegetables predicted lower 10-year cumulative Medicare Charges, Benefit, Gap and a lower number of claims, this was in the opposite direction for fruits, dairy and healthy fats. This finding is different to the results from the Western Electric company study of male employees [19] where after 25 years of follow-up there were lower Medicare (USA) costs for those with the highest reported fruits and vegetables intakes. Although the Medicare costs do represent different things in the USA and Australia, there is now some synergy with the longer follow-up in the Australian data set. Limitations in the analysis include that baseline health status was not adjusted for, although the top and bottom 1 % of extreme Medicare cost values were removed, and not all costs related to medical treatments are captured by the Medicare (Aus) data.

It will be critically important to continue to examine the relationship between diet quality and Medicare costs for the ALSWH, and to also examine these relationships in other cohort studies where data are available to conduct similar analyses. This is particularly so because of a previous review in which we demonstrated that across a range of diet quality scores and studies that higher diet quality is consistently inversely related to all-cause mortality, with a protective effect of moderate magnitude [1], and with associations seemingly stronger for men.

Mid-aged women with lower dietary variety and quality, as measured by the ARFS, do not have increased cumulative Medicare costs, but do have a greater number of Medicare claims. This suggests a need to monitor both costs and number of claims over time to determine whether higher diet quality does eventually lead to monetary savings.


It is surprising that so few economic evaluations of the relationship between diet quality and health care costs have been undertaken. Further evaluations will provide the basis for modelling the cost savings in future interventions and may help to identify appropriate dietary targets associated with reduced health care costs.



The ALSWH is funded by the Australian Government Department of Health and Ageing. CE Collins is supported by an Australian National Health and Medical Research Council NHMRC career development fellowship. We thank all participants in ALSWH for their valuable contribution to this project.


  1. 1.
    Wirt A, Collins C. Diet quality—what is it and does it matter? Public Health Nutr. 2009;12:2473–92.PubMedCrossRefGoogle Scholar
  2. 2.
    Hann C et al. Validation of the Healthy Eating Index with use of plasma biomarkers in a clinical sample of women. Am J Clin Nutr. 2001;74:479–86.PubMedGoogle Scholar
  3. 3.
    Huijbregts P et al. Dietary pattern and 20 year mortality in elderly men in Finland, Italy, and the Netherlands: longitudinal cohort study. Br Med J. 1997;315:13–7.CrossRefGoogle Scholar
  4. 4.
    Osler M et al. Dietary patterns and mortality in Danish men and women: a prospective observational study. Br J Nutr. 2001;85:219–25.PubMedCrossRefGoogle Scholar
  5. 5.
    Kant A et al. A prospective study of diet quality and mortality in women. J Am Med Assoc. 2000;283(16):2109–15.CrossRefGoogle Scholar
  6. 6.
    Seymour J et al. Diet Quality Index as a predictor of short-term mortality in the American Cancer Society Cancer Prevention Study II Nutrition Cohort. Am J Epidemiol. 2003;157(11):980–8.PubMedCrossRefGoogle Scholar
  7. 7.
    Fitzgerald A, Dewar R, Veugelers P. Diet quality and cancer incidence in Nova Scotia, Canada. Nutr Cancer. 2002;43(2):127–32.PubMedCrossRefGoogle Scholar
  8. 8.
    Trichopoulou A et al. Diet and survival in elderly people. Br Med J. 1995;311:1457–60.CrossRefGoogle Scholar
  9. 9.
    Kant A. Indexes of overall diet quality. J Am Diet Assoc. 1996;96:785–91.PubMedCrossRefGoogle Scholar
  10. 10.
    Kant A. Dietary patterns and health outcomes. J Am Diet Assoc. 2004;104:615–35.PubMedCrossRefGoogle Scholar
  11. 11.
    Kant A, Schatzkin A. Relation of age and self-reported chronic medical condition status with dietary nutrient intake in the US population. J Am Coll Nutr. 1999;18:69–76.PubMedCrossRefGoogle Scholar
  12. 12.
    Begg S, et al. The burden of disease and injury in Australia 2003. Canberra: Australian Institute of Health and Welfare; 2007.Google Scholar
  13. 13.
    Collins C, Young A, Hodge A. Diet quality is associated with higher nutrient intake and self-rated health in mid-aged women. J Am Coll Nutr. 2008;27:146–57.PubMedCrossRefGoogle Scholar
  14. 14.
    Haveman-Nies A, De Groot L, Van Staveren W. Relation of dietary quality, physical activity, and smoking habits to 10-year changes in health status in older Europeans in the SENECA study. Am J Public Health. 2003;93(2):318–23.PubMedCrossRefGoogle Scholar
  15. 15.
    Cobiac L, Vos T, Veerman J. Cost-effectiveness of interventions to promote fruit and vegetable consumption. PLoS One. 2010;5(11):e14148.PubMedCrossRefGoogle Scholar
  16. 16.
    Patrick D et al. Cost and outcomes of Medicare reimbursement for HMO preventive services. Health Care Financ Rev. 1999;20(4):25–43.PubMedGoogle Scholar
  17. 17.
    Wolf A et al. Effects of lifestyle intervention on health care costs: Improving Control with Activity and Nutrition (ICAN). J Am Diet Assoc. 2007; 107(8):1365–73.PubMedCrossRefGoogle Scholar
  18. 18.
    Pavlovich W et al. Systematic review of literature on the cost-effectiveness of nutrition services. J Am Diet Assoc. 2004;104(2):226–32.PubMedCrossRefGoogle Scholar
  19. 19.
    Daviglus M et al. Relationship of fruit and vegetable consumption in middle-aged men to Medicare expenditures in older age: the Chicago Western Electric Study. J Am Diet Assoc. 2005;105:1735–44.PubMedCrossRefGoogle Scholar
  20. 20.
    Collins C, Patterson A, Fitzgerald D. Higher diet quality does not predict lower Medicare costs but does predict number of claims in mid-aged Australian women. Nutrients. 2011;3(1):40–8.PubMedCrossRefGoogle Scholar
  21. 21.
    van Baal P et al. Lifetime medical costs of obesity: prevention no cure for increasing health expenditure. PLoS Med. 2008;5(2):e29.PubMedCrossRefGoogle Scholar
  22. 22.
    Brown W et al. Women’s Health Australia: recruitment for a national longitudinal cohort study. Women Health. 1998;28:23–40.PubMedGoogle Scholar
  23. 23.
    Brown W, Young A, Byles J. Tyranny of distance? The health of mid-age women living in five geographical areas of Australia. Aust J Rural Health. 1999;7: 148–54.PubMedCrossRefGoogle Scholar
  24. 24.
    Lee C et al. Cohort profile: the Australian Longitudinal Study on Women’s Health. Int J Epidemiol. 2005;34: 987–91.PubMedCrossRefGoogle Scholar
  25. 25.
    National Food Authority. NUTTAB95 database. Canberra: National Food Authority; 1995.Google Scholar
  26. 26.
    Giles G. Dietary Questionnaire for Epidemiological Studies (version 2). Melbourne: The Cancer Council Victoria; 1996.Google Scholar
  27. 27.
    Hodge A et al. The Anti Cancer Council of Victoria FFQ: relative validity of nutrient intakes compared with weighed food records in young to middle-aged women in a study of iron supplementation. Aust N Z J Public Health. 2000;24:576–83.PubMedCrossRefGoogle Scholar
  28. 28.
    Kant A, Thompson F. Measures of overall diet quality from a food frequency questionnaire: National Health Interview Survey. Nutr Res. 1992;17:1443–56.CrossRefGoogle Scholar
  29. 29.
    NHMRC. Dietary guidelines for Australian adults. Canberra: National Health and Medical Research Council; 2003.Google Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Clare Collins
    • 1
  • Alexis Hure
    • 2
  • Tracy Burrows
    • 3
  • Amanda Patterson
    • 1
  1. 1.School of Health SciencesThe University of Newcastle, AustraliaNewcastleAustralia
  2. 2.School of Medicine and Public Health, Research Centre for Gender, Health and AgeingUniversity of NewcastleCallaghanAustralia
  3. 3.School of Health SciencesUniversity of Newcastle, Priority Research Centre in Physical Activity and NutritionCallaghanAustralia

Personalised recommendations