The technique “Joint and Individual Variance Explained” highlights persistent aspects of the diet using longitudinal food frequency data

Dietary Pattern analysis is typically based on dimension reduction and summarizes the diet with a small number of scores. We assess “Joint and Individual Variance Explained” (JIVE) as a method for extracting dietary patterns from longitudinal data that highlights elements of the diet that are associated over time. The Auckland Birthweight Collaborative Study, in which participants completed a food frequency questionnaire at ages 3.5 (n=549), 7 (n=591), and 11 (n=617), is used as an example. Data from each time point is projected onto the directions of shared variability produced by JIVE to yield dietary patterns and scores. We assess the ability of the scores to predict future BMI and blood pressure measurements of the participants, and make a comparison with principal component analysis (PCA) performed separately at each time point. The diet could be summarized with three JIVE patterns. The patterns were interpretable, with the same interpretation across age groups: a vegetable and whole-grain pattern, a sweets and meats pattern, and a cereal vs sweet drinks pattern. The first two PCA-derived patterns were similar across age groups, and similar to the first two JIVE patterns. The interpretation of the third PCA pattern changed across age groups. Scores produced by the two techniques were similarly effective in predicting future BMI and blood pressure. We conclude that when data from the same participants at multiple ages is available, JIVE provides an advantage over PCA by extracting patterns with a common interpretation across age groups.

Dietary patterns and sarcopenia in elderly adults: the TCLSIH study

Sarcopenia is a core contributor to several health consequences, including falls, fractures, physical limitations, and disability. The pathophysiological processes of sarcopenia may be counteracted with the proper diet, delaying sarcopenia onset. Dietary pattern analysis is a whole diet approach used to investigate the relationship between diet and sarcopenia. Here we aimed to investigate this relationship in an elderly Chinese population. A cross-sectional study with 2,423 participants aged more than 60 years was performed. Sarcopenia was defined based on the guidelines of the Asian Working Group for Sarcopenia, composed of low muscle mass plus low grip strength and/or low gait speed. Dietary data were collected using a food-frequency questionnaire that included questions on 100 food items along with their specified serving sizes. Three dietary patterns were derived by factor analysis: sweet pattern; vegetable pattern; animal food pattern. The prevalence of sarcopenia was 16.1%. The higher vegetable pattern score and animal food pattern score were related to lower prevalence of sarcopenia (Ptrend =0.006 and Ptrend <0.001, respectively); the multivariate-adjusted odds ratio (95% confidence interval) of the prevalence of sarcopenia in the highest versus lowest quartiles were 0.54 (0.34, 0.86) and 0.50 (0.33, 0.74), separately. The sweet pattern score was not significantly related to the prevalence of sarcopenia. The present study showed that vegetable pattern and animal food pattern were related to a lower prevalence of sarcopenia in Chinese older adults. Further studies are required to clarify these findings.

Pattern analysis of vegan eating reveals healthy and unhealthy patterns within the vegan diet

Objective: This study aimed to identify the types of foods that constitute a vegan diet and establish patterns within the diet. Dietary pattern analysis, a key instrument for exploring the correlation between health and disease was used to identify patterns within the vegan diet. Design: A modified version of the EPIC-Norfolk food frequency questionnaire (FFQ) was created and validated to include vegan foods and launched on social media. Setting: UK participants, recruited online Participants: A convenience sample of 129 vegans voluntarily completed the FFQ. Collected data was converted to reflect weekly consumption to enable factor and cluster analyses. Results: Factor analysis identified four distinct dietary patterns including: 1) convenience, (22%); 2) health conscious, (12%); 3) unhealthy, (9%); and 4) traditional vegan (7%). Whilst two healthy patterns were defined, the convenience pattern was the most identifiable pattern with a prominence of vegan convenience meals and snacks, vegan sweets and desserts, sauces, condiments and fats. Cluster analysis identified three clusters, cluster one ‘convenience’ (26.8%), cluster two, ‘traditional’ (22%) and cluster 3 ‘health conscious’ (51.2%). Clusters one and two consisted of an array of ultra-processed vegan food items. Together, both clusters represent almost half of participants and yielding similar results to the predominant dietary pattern, strengthens the factor analysis. Conclusions: These novel results highlight a need for further dietary pattern studies with full nutrition and blood metabolite analysis in larger samples of vegans to enhance and ratify these results.

Advances in dietary pattern analysis in nutritional epidemiology

Background and Purpose It used to be a common practice in the field of nutritional epidemiology to analyze separate nutrients, foods, or food groups. However, in reality, nutrients and foods are consumed in combination. The introduction of dietary patterns (DP) and their analysis has revolutionized this field, making it possible to take into account the synergistic effects of foods and to account for the complex interaction among nutrients and foods. Three approaches of DP analysis exist: (1) the hypothesis-based approach (based on prior knowledge regarding the current understanding of dietary components and their health relation), (2) the exploratory approach (solely relying on dietary intake data), and (3) the hybrid approach (a combination of both approaches). During the recent past, complementary approaches for DP analysis have emerged both conceptually and methodologically. Method We have summarized the recent developments that include incorporating the Treelet transformation method as a complementary exploratory approach in a narrative review. Results Uses, peculiarities, strengths, limitations, and scope of recent developments in DP analysis are outlined. Next, the narrative review gives an overview of the literature that takes into account potential relevant dietary-related factors, specifically the metabolome and the gut microbiome in DP analysis. Then the review deals with the aspect of data processing that is needed prior to DP analysis, particularly when dietary data arise from assessment methods other than the long-established food frequency questionnaire. Lastly, potential opportunities for upcoming DP analysis are summarized in the outlook. Conclusion Biological factors like the metabolome and the microbiome are crucial to understand diet-disease relationships. Therefore, the inclusion of these factors in DP analysis might provide deeper insights.

A review of statistical methods for dietary pattern analysis

Background Dietary pattern analysis is a promising approach to understanding the complex relationship between diet and health. While many statistical methods exist, the literature predominantly focuses on classical methods such as dietary quality scores, principal component analysis, factor analysis, clustering analysis, and reduced rank regression. There are some emerging methods that have rarely or never been reviewed or discussed adequately. Methods This paper presents a landscape review of the existing statistical methods used to derive dietary patterns, especially the finite mixture model, treelet transform, data mining, least absolute shrinkage and selection operator and compositional data analysis, in terms of their underlying concepts, advantages and disadvantages, and available software and packages for implementation. Results While all statistical methods for dietary pattern analysis have unique features and serve distinct purposes, emerging methods warrant more attention. However, future research is needed to evaluate these emerging methods’ performance in terms of reproducibility, validity, and ability to predict different outcomes. Conclusion Selection of the most appropriate method mainly depends on the research questions. As an evolving subject, there is always scope for deriving dietary patterns through new analytic methodologies.

Salivary Protein Profile and Food Intake: A Dietary Pattern Analysis

Saliva research has gained interest due to its potential as a source of biomarkers. One of the factors inducing changes in saliva, in the short term, is food intake, and evidence exist about changes in salivary proteome induced by some food components. Since this topic of research is in its early stages, it was hypothesized that saliva protein composition could be associated with different levels of adherence to dietary patterns that contain higher amounts of plant products. The aim of the present study was to test this hypothesis, in adults, by comparing salivary protein electrophoretic profiles of individuals with different diet characteristics, particularly dietary patterns (DP) that exhibit different proportions of animal and plant-based products. Dietary habits were assessed in 122 adults (61 from each sex, with ages ranging from 20 to 59 years) using Food Frequency Questionnaires. To identify the dietary patterns, a principal component analysis was used. Individual’s non-stimulated saliva was evaluated for flow rate, pH, protein concentration, α-amylase activity, and electrophoretic protein profiles. Seven dietary patterns (DP) were identified. Salivary amylase enzymatic activity was positively associated with animal-based and starchy foods DP, and with plant-based fatty foods without wine DP. At the same time, protein bands containing amylase and type S cystatins were positively associated with the cheese/yoghurt and wine DP. Our results support the association of salivary proteomics and different dietary patterns and highlight the need of considering food consumption habits in studies using saliva, since this is a factor associated with variations in the composition of this fluid.

Association between diet and periodontitis: a cross-sectional study of 10,000 NHANES participants

ABSTRACT Background Periodontitis is a major cause of tooth loss globally. Risk factors include age, smoking, and diabetes. Intake of specific nutrients has been associated with periodontitis risk but there has been little research into the influence of overall diet, potentially more relevant when formulating dietary recommendations. Objectives We aimed to investigate potential associations between diet and periodontitis using novel statistical techniques for dietary pattern analysis. Methods Two 24-h dietary recalls and periodontal examination data from the cross-sectional US NHANES, 2009–2014 (n = 10,010), were used. Dietary patterns were extracted using treelet transformation, a data-driven hierarchical clustering and dimension reduction technique. Associations between each pattern [treelet component (TC)] and extent of periodontitis [proportion of sites with clinical attachment loss (CAL) ≥ 3 mm] were estimated using robust logistic quantile regression, adjusting for age, sex, ethnicity, education level, smoking, BMI, and diabetes. Results Eight TCs explained 21% of the variation in diet, 1 of which (TC1) was associated with CAL extent. High TC1 scores represented a diet rich in salad, fruit, vegetables, poultry and seafood, and plain water or tea to drink. There was a substantial negative gradient in CAL extent from the lowest to the highest decile of TC1 (median proportion of sites with CAL ≥ 3 mm: decile 1 = 19.1%, decile 10 = 8.1%; OR, decile 10 compared with decile 1: 0.67; 95% CI: 0.46, 0.99). Conclusions Most dietary patterns identified were not associated with periodontitis extent. One pattern, however, rich in salad, fruit, and vegetables and with plain water or tea to drink, was associated with lower CAL extent. Treelet transformation may be a useful approach for calculating dietary patterns in nutrition research.

Dietary pattern analysis among stone formers: resemblance to a DASH-style diet

Recent epidemiological studies have shown that dietary patterns may have a more persistent impact on the risk of stone formation than single nutrients of the diet. Dietary Approaches to Stop Hypertension (DASH), a low-sodium and fruits/vegetables-rich diet, has been associated with a lower risk of nephrolithiasis, due to altered urinary biochemistry. This observational study aimed to investigate whether the dietary pattern of stone formers (SF) resembled a DASH-diet and its influence on urinary lithogenic parameters. Anthropometric data, fasting serum sample, 24-h urine samples, and a 3-day food intake record under an unrestricted diet were obtained from 222 SF and compared with 136 non-SF subjects (controls). The DASH-diet food portions were determined from the food records whereas intakes of sodium chloride (NaCl) and protein (protein equivalent of nitrogen appearance, PNA) were estimated from 24-hr urinary sodium and urea. A dietary profile close to a DASH-diet was not observed in any of the groups. NaCl intake and PNA were significantly higher in SF versus non-SF (12.0 ± 5.2 v.s. 10.1 ± 3.4 g/day, p = 0.01 and 1.8 ± 0.1 v.s. 1.4 ± 0.1 g/kg/day, p = 0.03). SF exhibited a positive correlation of NaCl intake and PNA with urinary calcium, oxalate and uric acid, and of PNA with urinary sodium. SF consumed more vegetables and legumes, but less fruits and low-fat dairy items than non-SF. The present series presented a dietary profile characterized by low calcium and high salt and protein contents, not reflecting an ideal DASH-style diet pattern.

Dietary pattern longitudinality during 8 years in children: results from the European Longitudinal Study of Pregnancy and Childhood (ELSPAC–CZ)

Objective: Dietary pattern analysis constitutes a suitable method for identifying complex food preferences as well as a useful tool for comparing dietary behaviour across individual populations. In addition to a lack of information on Central European dietary patterns, dietary data featuring a longitudinal aspect are likewise largely unavailable for the region. Our study thus strives to address this gap by analysing children’s dietary patterns, their stability and possible changes at 7, 11 and 15 years in the Czech part of the European Longitudinal Study of Pregnancy and Childhood (ELSPAC–CZ). Design: We analysed dietary data based on the self-reported semi-quantitative FFQ obtained in 1998, 2002 and 2006. Dietary patterns were derived using factor analysis for each period, followed by the determination of dietary pattern stability across the individual periods. Setting: The analysis of dietary patterns was based on longitudinal children’s dietary data from the geographical region that was undergoing massive socio-economic changes at the time of birth of the study subjects. Participants: All participants were children. At 7 years the analysis included 3220 children, at 11 years the analysis included 2509 children and at 15 years the analysis included 1589 children. Results: Two stable children’s dietary patterns labelled as ‘prudent’ and ‘junk food’ were identified across all three time points (7, 11 and 15 years). Conclusions: This study identifies stable longitudinal trends in the dietary behaviour of children enrolled in the ELSPAC-CZ study.