Measuring diet quality: development of a new index and comparative evaluation with other indices in Australia

Rasoul Zarrin (2010). Measuring diet quality: development of a new index and comparative evaluation with other indices in Australia PhD Thesis, School of Population Health, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Thesis_submission_form.pdf Thesis_submission_form.pdf application/pdf 289.04KB 0
s40737562_PhD_Abstract.pdf Abstract application/pdf 16.49KB 0
s40737562_PhD_finalthesis.pdf Final thesis application/pdf 1.60MB 16
Author Rasoul Zarrin
Thesis Title Measuring diet quality: development of a new index and comparative evaluation with other indices in Australia
School, Centre or Institute School of Population Health
Institution The University of Queensland
Publication date 2010-03
Thesis type PhD Thesis
Supervisor A/Prof. Geoffrey Marks
Dr. Torukiri Ibiebele
Prof. Gail Williams
Total pages 255
Total colour pages 6
Total black and white pages 249
Subjects 11 Medical and Health Sciences
Abstract/Summary Background and aims Developing a valid index of overall diet quality and using the measure to predict chronic disease risk has been a focus of nutritional epidemiology; however, there has been little research on the validation of such indices for use in Australia. This research aimed to critically investigate the literature on diet quality indices to develop a valid diet quality index based on dietary guidance for Australians and to examine its validity using data from Australian population based cross-sectional and longitudinal studies. Construct validity of the index was also assessed by comparing the Dietary Guideline Index (DGI) and Australian Healthy Eating Index (Aust-HEI). Methods A critical literature review was undertaken to guide design and validity assessment of the Aussie-DQI. A new categorisation method of diet quality indices based on their purpose and which explains differences in diet quality indices was introduced. Dietary guidance for Australians was used to guide the selection of components and development of age- and sex-specific cut-offs of the new index. For the purpose of assessing content validity, a checklist was used to ensure that components of the new index addressed the key aspects of the guidelines. To assess construct validity of the Aussie-DQI, the association of the Aussie-DQI with socio-economic, demographic and health behaviour characteristics and intakes of desirable/undesirable components adjusted for sex and age was examined. People with implausible energy intake, and breastfeeding and lactating women were excluded from the analyses. We used the 1995 National Nutrition Survey (NNS), a nationally representative data with 10,851 participants aged 19 years and over which used 24/h recall for dietary assessment. Physical measurements included anthropometric measurements and blood pressure, and interviewer administered questionnaires were used to collect information on the above mentioned characteristics. We used a general linear regression model to examine the association between dietary intakes and the above mentioned characteristics (exposure) across categories of the Aussie-DQI scores (outcome) using data from the 24/h recall. The analysis was repeated using dietary data obtained from a validated food frequency questionnaire (FFQ) also administered during the 1995 NNS. Criterion validity was assessed based on the capacity of the Aussie-DQI to predict mortality over 16 year follow-up. Longitudinal data were used from the Nambour Skin Cancer Study, with 1621 participants aged between 25 and 75 years at baseline. Information relating to age, sex, education, smoking, physical activity and history of selected medical conditions was collected using a self-administered health behaviour questionnaire at baseline. The Cox proportional hazard model was used to examine the association between the index (exposure) and all-cause and specific-cause mortality (outcome). Results Analysis of dietary information from the 24/h recall of the 1995 NNS shows that females (F:55.3±12.8 vs. M:52.5±13.6), higher age (F:57.2±12.5 vs. 49.2±12.7; M:56.8±13.2 vs. 45.3±12.9), non-smokers (F:56.5±12.6 vs. 51.3±12.5; M:53.3±13.4 vs. 47.3±12.8), lower BMI (F:54.8±12.7 vs. 53.4±12.8; M:52.8±13.8 vs. 49.2±13.0), higher physical activity (F:55.8±12.4 vs. 52.0±12.8 ; M:52.4±14.0 vs. 49.7±13.4), higher income (M:52.8±12.9 vs. 50.0±13.7) and higher schooling years (M:51.4±13.5 vs. 50.4±13.9), are more likely to have higher diet quality scores (P for trend < 0.05). The intake of meat and dairy products, calcium (men) and zinc generally decreased as diet quality scores increased, but no statistically significant decrease was found for calcium intake among women. Similar results were achieved when dietary information from the FFQ was used in the analysis. When compared with the two recent Australian diet quality indices, similar associations were found between the score of the three indices and intake of most food groups and nutrients. Only the Aussie-DQI showed a significant inverse association with BMI among males, while the DGI scores were significantly higher among females in the highest BMI group, adjusted for energy and other characteristics. A significant inverse association between diet quality score and the intake of total sugar was observed only with the Aussie-DQI in both genders (P for trend < 0.05). In the Cox proportional hazard analysis, after adjustment for potential confounding variables the Aussie-DQI was inversely associated with the risk of cancer mortality among men (HR= 0.30, 95% CI: 0.11, 0.83; P for trends = 0.06). No association was found for CVD mortality or cancer mortality for females. Conclusion A critical literature review on diet quality indices has been provided by modelling a systematic approach for developing and validating diet quality indices. The new approach explains differences in diet quality indices and provides guidance on several stages of their development. The Aussie-DQI, which has been developed and validated to measure diet quality in Australia, is different from other indices, particularly concerning components, approaches used for validation, flexibility of application to different studies and types of dietary information and consideration of overconsumption. The findings of the assessment of construct validity are comparable with other well-conducted studies. The assessment of criterion validity of the Aussie-DQI was associated with decreased risk of cancer mortality among men.
Keyword diet, quality, index, validity, dietary guidelines, all-cause, specific-cause mortality
Additional Notes Colour pages: 13, 20, 50, 51, 167, 178 Landscape pages: 26, 50-55, 60- 69, 72, 73, 75, 78, 82, 83, 86, 134, 135, 138, 139, 141-145, 163-166, 169, 170, 171, 209-238, 240-242

Citation counts: Google Scholar Search Google Scholar
Created: Mon, 04 Apr 2011, 19:18:42 EST by Mr Rasoul Zarrin on behalf of Library - Information Access Service