Joint Data Analysis in Nutritional Epidemiology: Identification of Observational Studies and Minimal Requirements

Harmonizing Dietary Exposure of Adult and Older Individuals: A Methodological Work of the Collaborative PROMED-COG Pooled Cohorts Study

Objectives: The PROtein-enriched MEDiterranean diet to combat undernutrition and promote healthy neuroCOGnitive ageing in older adults (PROMED-COG) is a European project that investigates the role of nutritional status on neurocognitive ageing. This methodological paper describes the harmonization process of dietary data from four Italian observational studies (Pro.V.A., ILSA, BEST-FU, and NutBrain). Methods: Portion sizes and food frequency consumption within different food frequency questionnaires were retrospectively harmonized across the datasets on daily food frequency, initially analyzing raw data using the original codebook and establishing a uniform food categorization system. Individual foods were then aggregated into 27 common food groups. Results: The pooled cohort consisted of 9326 individuals (40-101 years, 52.4% female). BEST-FU recruited younger participants who were more often smokers and less physically active than those of the other studies. Dietary instruments varied across the studies differing in the number of items and time intervals assessed, but all collected dietary intake through face-to-face interviews with a common subset of items. The average daily intakes of the 27 food groups across studies varied, with BEST-FU participants generally consuming more fruits, vegetables, red meat, and fish than the other studies. Conclusions: Harmonization of dietary data presents challenges but allows for the integration of information from diverse studies, leading to a more robust and statistically powerful dataset. The study highlights the feasibility and benefits of data harmonization, despite inherent limitations, and sets the stage for future research into the effects of diet on cognitive health and aging.

Strengths and limitations of non-disclosive data analysis: a comparison of breast cancer survival classifiers using VisualSHIELD

Preserving data privacy is an important concern in the research use of patient data. The DataSHIELD suite enables privacy-aware advanced statistical analysis in a federated setting. Despite its many applications, it has a few open practical issues: the complexity of hosting a federated infrastructure, the performance penalty imposed by the privacy-preserving constraints, and the ease of use by non-technical users. In this work, we describe a case study in which we review different breast cancer classifiers and report our findings about the limits and advantages of such non-disclosive suite of tools in a realistic setting. Five independent gene expression datasets of breast cancer survival were downloaded from Gene Expression Omnibus (GEO) and pooled together through the federated infrastructure. Three previously published and two newly proposed 5-year cancer-free survival risk score classifiers were trained in a federated environment, and an additional reference classifier was trained with unconstrained data access. The performance of these six classifiers was systematically evaluated, and the results show that i) the published classifiers do not generalize well when applied to patient cohorts that differ from those used to develop them; ii) among the methods we tried, the classification using logistic regression worked better on average, closely followed by random forest; iii) the unconstrained version of the logistic regression classifier outperformed the federated version by 4% on average. Reproducibility of our experiments is ensured through the use of VisualSHIELD, an open-source tool that augments DataSHIELD with new functions, a standardized deployment procedure, and a simple graphical user interface.

Chronic disease outcome metadata from German observational studies - public availability and FAIR principles

Metadata from epidemiological studies, including chronic disease outcome metadata (CDOM), are important to be findable to allow interpretability and reusability. We propose a comprehensive metadata schema and used it to assess public availability and findability of CDOM from German population-based observational studies participating in the consortium National Research Data Infrastructure for Personal Health Data (NFDI4Health). Additionally, principal investigators from the included studies completed a checklist evaluating consistency with FAIR principles (Findability, Accessibility, Interoperability, Reusability) within their studies. Overall, six of sixteen studies had complete publicly available CDOM. The most frequent CDOM source was scientific publications and the most frequently missing metadata were availability of codes of the International Classification of Diseases, Tenth Revision (ICD-10). Principal investigators' main perceived barriers for consistency with FAIR principles were limited human and financial resources. Our results reveal that CDOM from German population-based studies have incomplete availability and limited findability. There is a need to make CDOM publicly available in searchable platforms or metadata catalogues to improve their FAIRness, which requires human and financial resources.

The ethical aspects of exposome research: a systematic review

In recent years, exposome research has been put forward as the next frontier for the study of human health and disease. Exposome research entails the analysis of the totality of environmental exposures and their corresponding biological responses within the human body. Increasingly, this is operationalized by big-data approaches to map the effects of internal as well as external exposures using smart sensors and multiomics technologies. However, the ethical implications of exposome research are still only rarely discussed in the literature. Therefore, we conducted a systematic review of the academic literature regarding both the exposome and underlying research fields and approaches, to map the ethical aspects that are relevant to exposome research. We identify five ethical themes that are prominent in ethics discussions: the goals of exposome research, its standards, its tools, how it relates to study participants, and the consequences of its products. Furthermore, we provide a number of general principles for how future ethics research can best make use of our comprehensive overview of the ethical aspects of exposome research. Lastly, we highlight three aspects of exposome research that are most in need of ethical reflection: the actionability of its findings, the epidemiological or clinical norms applicable to exposome research, and the meaning and action-implications of bias.

HDHL-INTIMIC: A European Knowledge Platform on Food, Diet, Intestinal Microbiomics, and Human Health

Studies indicate that the intestinal microbiota influences general metabolic processes in humans, thereby modulating the risk of chronic diseases such as type 2 diabetes, allergy, cardiovascular disease, and colorectal cancer (CRC). Dietary factors are also directly related to chronic disease risk, and they affect the composition and function of the gut microbiota. Still, detailed knowledge on the relation between diet, the microbiota, and chronic disease risk is limited. The overarching aim of the HDHL-INTIMIC (INtesTInal MICrobiomics) knowledge platform is to foster studies on the microbiota, nutrition, and health by assembling available knowledge of the microbiota and of the other aspects (e.g., food science and metabolomics) that are relevant in the context of microbiome research. The goal is to make this information findable, accessible, interoperable, and reusable (FAIR) to the scientific community, and to share information with the various stakeholders. Through these efforts a network of transnational and multidisciplinary collaboration has emerged, which has contributed to further develop and increase the impact of microbiome research in human health. The roles of microbiota in early infancy, during ageing, and in subclinical and clinically manifested disease are identified as urgent areas of research in this knowledge platform.

Identification and Characterization of Human Observational Studies in Nutritional Epidemiology on Gut Microbiomics for Joint Data Analysis

In any research field, data access and data integration are major challenges that even large, well-established consortia face. Although data sharing initiatives are increasing, joint data analyses on nutrition and microbiomics in health and disease are still scarce. We aimed to identify observational studies with data on nutrition and gut microbiome composition from the Intestinal Microbiomics (INTIMIC) Knowledge Platform following the findable, accessible, interoperable, and reusable (FAIR) principles. An adapted template from the European Nutritional Phenotype Assessment and Data Sharing Initiative (ENPADASI) consortium was used to collect microbiome-specific information and other related factors. In total, 23 studies (17 longitudinal and 6 cross-sectional) were identified from Italy (7), Germany (6), Netherlands (3), Spain (2), Belgium (1), and France (1) or multiple countries (3). Of these, 21 studies collected information on both dietary intake (24 h dietary recall, food frequency questionnaire (FFQ), or Food Records) and gut microbiome. All studies collected stool samples. The most often used sequencing platform was Illumina MiSeq, and the preferred hypervariable regions of the 16S rRNA gene were V3-V4 or V4. The combination of datasets will allow for sufficiently powered investigations to increase the knowledge and understanding of the relationship between food and gut microbiome in health and disease.

Dietary Macronutrient Composition in Relation to Circulating HDL and Non-HDL Cholesterol: A Federated Individual-Level Analysis of Cross-Sectional Data from Adolescents and Adults in 8 European Studies

BACKGROUND

Associations between increased dietary fat and decreased carbohydrate intake with circulating HDL and non-HDL cholesterol have not been conclusively determined.

OBJECTIVE

We assessed these relations in 8 European observational human studies participating in the European Nutritional Phenotype Assessment and Data Sharing Initiative (ENPADASI) using harmonized data.

METHODS

Dietary macronutrient intake was recorded using study-specific dietary assessment tools. Main outcome measures were lipoprotein cholesterol concentrations: HDL cholesterol (mg/dL) and non-HDL cholesterol (mg/dL). A cross-sectional analysis on 5919 participants (54% female) aged 13-80 y was undertaken using the statistical platform DataSHIELD that allows remote/federated nondisclosive analysis of individual-level data. Generalized linear models (GLM) were fitted to assess associations between replacing 5% of energy from carbohydrates with equivalent energy from total fats, SFAs, MUFAs, or PUFAs with circulating HDL cholesterol and non-HDL cholesterol. GLM were adjusted for study source, age, sex, smoking status, alcohol intake and BMI.

RESULTS

The replacement of 5% of energy from carbohydrates with total fats or MUFAs was statistically significantly associated with 0.67 mg/dL (95% CI: 0.40, 0.94) or 0.99 mg/dL (95% CI: 0.37, 1.60) higher HDL cholesterol, respectively, but not with non-HDL cholesterol concentrations. The replacement of 5% of energy from carbohydrates with SFAs or PUFAs was not associated with HDL cholesterol, but SFAs were statistically significantly associated with 1.94 mg/dL (95% CI: 0.08, 3.79) higher non-HDL cholesterol, and PUFAs with -3.91 mg/dL (95% CI: -6.98, -0.84) lower non-HDL cholesterol concentrations. A statistically significant interaction by sex for the association of replacing carbohydrates with MUFAs and non-HDL cholesterol was observed, showing a statistically significant inverse association in males and no statistically significant association in females. We observed no statistically significant interaction by age.

CONCLUSIONS

The replacement of dietary carbohydrates with fats had favorable effects on lipoprotein cholesterol concentrations in European adolescents and adults when fats were consumed as MUFAs or PUFAs but not as SFAs.

Deep learning in systems medicine

Systems medicine (SM) has emerged as a powerful tool for studying the human body at the systems level with the aim of improving our understanding, prevention and treatment of complex diseases. Being able to automatically extract relevant features needed for a given task from high-dimensional, heterogeneous data, deep learning (DL) holds great promise in this endeavour. This review paper addresses the main developments of DL algorithms and a set of general topics where DL is decisive, namely, within the SM landscape. It discusses how DL can be applied to SM with an emphasis on the applications to predictive, preventive and precision medicine. Several key challenges have been highlighted including delivering clinical impact and improving interpretability. We used some prototypical examples to highlight the relevance and significance of the adoption of DL in SM, one of them is involving the creation of a model for personalized Parkinson's disease. The review offers valuable insights and informs the research in DL and SM.

An Ontology to Standardize Research Output of Nutritional Epidemiology: From Paper-Based Standards to Linked Content

BACKGROUND

The use of linked data in the Semantic Web is a promising approach to add value to nutrition research. An ontology, which defines the logical relationships between well-defined taxonomic terms, enables linking and harmonizing research output. To enable the description of domain-specific output in nutritional epidemiology, we propose the Ontology for Nutritional Epidemiology (ONE) according to authoritative guidance for nutritional epidemiology.

METHODS

Firstly, a scoping review was conducted to identify existing ontology terms for reuse in ONE. Secondly, existing data standards and reporting guidelines for nutritional epidemiology were converted into an ontology. The terms used in the standards were summarized and listed separately in a taxonomic hierarchy. Thirdly, the ontologies of the nutritional epidemiologic standards, reporting guidelines, and the core concepts were gathered in ONE. Three case studies were included to illustrate potential applications: (i) annotation of existing manuscripts and data, (ii) ontology-based inference, and (iii) estimation of reporting completeness in a sample of nine manuscripts.

RESULTS

Ontologies for "food and nutrition" (n = 37), "disease and specific population" (n = 100), "data description" (n = 21), "research description" (n = 35), and "supplementary (meta) data description" (n = 44) were reviewed and listed. ONE consists of 339 classes: 79 new classes to describe data and 24 new classes to describe the content of manuscripts.

CONCLUSION

ONE is a resource to automate data integration, searching, and browsing, and can be used to assess reporting completeness in nutritional epidemiology.

Reducing waste in nutritional epidemiology: review and perspectives

We discuss efforts in improving the value of nutrition research. We organised the paper in five research stages: Stage 1: research priority setting; Stage 2: research design, conduct and analysis; Stage 3: research regulation and management; Stage 4: research accessibility and Stage 5: research reporting and publishing. Along the stages of the research cycle, varied initiatives exist to improve the quality and added value of nutrition research. However, efforts are focused on single stages of the research cycle without vision of the research system as a whole. Although research on nutrition research has been limited, it has potential to improve the quality of nutrition research and develop new tools and instruments for this purpose. A comprehensive assessment of the magnitude of research waste in nutrition and consensus on priority actions is needed. The nutrition research community at large needs to have open discussions on the usefulness of these tools and lead suitable efforts to enhance nutrition research across the stages of the research cycle. Capacity building is essential and considerations of nutrition research quality are vital to be integrated in training efforts of nutrition researchers.

Dietary Behavior: An Interdisciplinary Conceptual Analysis and Taxonomy

Background: Dietary behavior encompasses many aspects, terms for which are used inconsistently across different disciplines and research traditions. This hampers communication and comparison across disciplines and impedes the development of a cumulative science. We describe the conceptual analysis of the fuzzy umbrella concept "dietary behavior" and present the development of an interdisciplinary taxonomy of dietary behavior. Methods: A four-phase multi-method approach was employed. Input was provided by 76 scholars involved in an international research project focusing on the determinants of dietary behavior. Input was collected from the scholars via an online mind mapping procedure. After structuring, condensing, and categorizing this input into a compact taxonomy, the result was presented to all scholars, discussed extensively, and adapted. A second revision round was then conducted among a core working group. Results: A total of 145 distinct entries were made in the original mind mapping procedure. The subsequent steps allowed us to reduce and condense the taxonomy into a final product consisting of 34 terms organized into three main categories: Food Choice, Eating Behavior, and Dietary Intake/Nutrition. In a live discussion session attended by 50 of the scholars involved in the development of the taxonomy, it was judged to adequately reflect their input and to be a valid and useful starting point for interdisciplinary understanding and collaboration. Conclusion: The current taxonomy can be used as a tool to facilitate understanding and cooperation between different disciplines investigating dietary behavior, which may contribute to a more successful approach to tackling the complex public health challenges faced by the field. The taxonomy need not be viewed as a final product, but can continue to grow in depth and width as additional experts provide their input.

ONS: an ontology for a standardized description of interventions and observational studies in nutrition

Background

The multidisciplinary nature of nutrition research is one of its main strengths. At the same time, however, it presents a major obstacle to integrate data analysis, especially for the terminological and semantic interpretations that specific research fields or communities are used to. To date, a proper ontology to structure and formalize the concepts used for the description of nutritional studies is still lacking.

Results

We have developed the Ontology for Nutritional Studies (ONS) by harmonizing selected pre-existing de facto ontologies with novel health and nutritional terminology classifications. The ONS is the result of a scholarly consensus of 51 research centers in nine European countries. The ontology classes and relations are commonly encountered while conducting, storing, harmonizing, integrating, describing, and searching nutritional studies. The ONS facilitates the description and specification of complex nutritional studies as demonstrated with two application scenarios.

Conclusions

The ONS is the first systematic effort to provide a solid and extensible formal ontology framework for nutritional studies. Integration of new information can be easily achieved by the addition of extra modules (i.e., nutrigenomics, metabolomics, nutrikinetics, and quality appraisal). The ONS provides a unified and standardized terminology for nutritional studies as a resource for nutrition researchers who might not necessarily be familiar with ontologies and standardization concepts.

Electronic supplementary material

The online version of this article (10.1186/s12263-018-0601-y) contains supplementary material, which is available to authorized users.