Thinking big: large-scale collaborative research in observational epidemiology

Harmonized Phenotypes for Anxiety, Depression, and Attention-Deficit Hyperactivity Disorder (ADHD)

In multi-cohort consortia, the problem often arises that a phenotype is measured using different questionnaires. This study aimed to harmonize scores based on the Child Behaviour Check List (CBCL) and the Strength and Difficulties Questionnaire (SDQ) for anxiety/depression and ADHD. To link the scales, we used parent reports on 1330 children aged 10–11.5 years from the Raine study on both SDQ and CBCL. Harmonization was done based on Item Response Theory. We started from existing CBCL and SDQ scales related to anxiety/depression and ADHD (theoretical approach). Next, we conducted a data-driven approach using factor analysis to validate the theoretical approach. Both approaches yielded similar scales, validating the combination of existing scales. In addition, we studied the impact of harmonized (IRT-based) scores on the statistical power of the results in meta-analytic gene-finding studies. The results showed that the IRT-based harmonized scores increased the statistical power of the results compared to sum scores, even with an equal sample size. These findings can help future researchers to harmonize data from different samples and/or different questionnaires that measure anxiety, depression, and ADHD, in order to obtain the larger sample sizes, to compare research results across subpopulations or to increase generalizability, the validity or statistical power of research results. We recommend using our item parameters to estimate harmonized scores that represent commensurate phenotypes across cohorts, and we explained in detail how other researchers can use our results to harmonize data in their studies.

Systematic Review Reveals Lack of Causal Methodology Applied to Pooled Longitudinal Observational Infectious Disease Studies

OBJECTIVES

Among ID studies seeking to make causal inferences and pooling individual-level longitudinal data from multiple infectious disease cohorts, we sought to assess what methods are being used, how those methods are being reported, and whether these factors have changed over time.

STUDY DESIGN AND SETTING

Systematic review of longitudinal observational infectious disease studies pooling individual-level patient data from 2+ studies published in English in 2009. 2014, or 2019. This systematic review protocol is registered with PROSPERO (CRD42020204104).

RESULTS

Our search yielded 1,462 unique articles. Of these, 16 were included in the final review. Our analysis showed a lack of causal inference methods and of clear reporting on methods and the required assumptions.

CONCLUSION

There are many approaches to causal inference which may help facilitate accurate inference in the presence of unmeasured and time-varying confounding. In observational ID studies leveraging pooled, longitudinal IPD, the absence of these causal inference methods and gaps in the reporting of key methodological considerations suggests there is ample opportunity to enhance the rigor and reporting of research in this field. Interdisciplinary collaborations between substantive and methodological experts would strengthen future work.

Current trends in the application of causal inference methods to pooled longitudinal non-randomised data: a protocol for a methodological systematic review

INTRODUCTION

Causal methods have been adopted and adapted across health disciplines, particularly for the analysis of single studies. However, the sample sizes necessary to best inform decision-making are often not attainable with single studies, making pooled individual-level data analysis invaluable for public health efforts. Researchers commonly implement causal methods prevailing in their home disciplines, and how these are selected, evaluated, implemented and reported may vary widely. To our knowledge, no article has yet evaluated trends in the implementation and reporting of causal methods in studies leveraging individual-level data pooled from several studies. We undertake this review to uncover patterns in the implementation and reporting of causal methods used across disciplines in research focused on health outcomes. We will investigate variations in methods to infer causality used across disciplines, time and geography and identify gaps in reporting of methods to inform the development of reporting standards and the conversation required to effect change.

METHODS AND ANALYSIS

We will search four databases (EBSCO, Embase, PubMed, Web of Science) using a search strategy developed with librarians from three universities (Heidelberg University, Harvard University, and University of California, San Francisco). The search strategy includes terms such as 'pool*', 'harmoniz*', 'cohort*', 'observational', variations on 'individual-level data'. Four reviewers will independently screen articles using Covidence and extract data from included articles. The extracted data will be analysed descriptively in tables and graphically to reveal the pattern in methods implementation and reporting. This protocol has been registered with PROSPERO (CRD42020143148).

ETHICS AND DISSEMINATION

No ethical approval was required as only publicly available data were used. The results will be submitted as a manuscript to a peer-reviewed journal, disseminated in conferences if relevant, and published as part of doctoral dissertations in Global Health at the Heidelberg University Hospital.

Experiences from the harmonization of Finnish national population-based health survey data

Aims: There are several advantages to pooling survey data from individual studies over time or across different countries. Our aim is to share our experiences on harmonizing data from 13 Finnish health examination surveys covering the years 1972–2017 and to describe the challenges related to harmonizing different variable types using two questionnaire variables – blood pressure measurement and total cholesterol assessment – as examples. Methods: Data from Finnish national population-based health surveys were harmonized as part of the research project ‘Projections of the Burden of Disease and Disability in Finland – Health Policy Prospects’, including variables from questionnaires, objective health measurements and results from the laboratory analysis of biological samples. The process presented in the Maelstrom Research guidelines for data harmonization was followed with minor adjustments. Results: The harmonization of data from objective measurements and biomarkers was reasonably straightforward, but questionnaire items proved more challenging. Some questions and response options had changed during the covered time period. This concerned, for example, questionnaire items on the availability and use of medication and diet. Conclusions: The long time period – 45 years – made harmonization more complicated. The survey questions or response options had changed for some topics due to changes in society. However, common core variables for topics that were especially relevant for the project, such as lifestyle factors and certain diseases or conditions, could be harmonized with sufficient comparability. For future surveys, the use of standardized survey methods and the proper documentation of data collection are recommended to facilitate harmonization.

Behavior Problems Following Childhood TBI: The Role of Sex, Age, and Time Since Injury

OBJECTIVE

To examine the frequency of behavioral problems after childhood traumatic brain injury (TBI) and their associations with injury severity, sex, and social environmental factors.

SETTING

Children's hospitals in the Midwestern/Western United States.

PARTICIPANTS

381 boys and 210 girls with moderate (n = 359) and severe (n = 227) TBI, with an average age at injury of 11.7 years (range 0.3-18) who were injured ≤3 years ago.

DESIGN

Secondary data analysis of a multistudy cohort.

MAIN MEASURES

Child Behavior Checklist (CBCL) administered pretreatment.

RESULTS

Thirty-seven percent had borderline/clinical elevations on the CBCL Total Problem Scale, with comparable rates of Internalizing and Externalizing problems (33% and 31%, respectively). Less parental education was associated with higher rates of internalizing, externalizing, and total problems. Time since injury had a linear association with internalizing symptoms, with greater symptoms at longer postinjury intervals. Younger boys had significantly higher levels of oppositional defiant symptoms than girls, whereas older girls had significantly greater attention-deficit hyperactivity disorder symptoms than boys.

CONCLUSIONS

Pediatric TBI is associated with high rates of behavior problems, with lower socioeconomic status predicting substantially elevated risk. Associations of higher levels of internalizing symptoms with greater time since injury highlight the importance of tracking children over time.

Centralizing environmental datasets to support (inter)national chronic disease research: Values, challenges, and recommendations

Whereas environmental data are increasingly available, it is often not clear how or if datasets are available for health research. Exposure metrics are typically developed for specific research initiatives using disparate exposure assessment methods and no mechanisms are put in place for centralizing, archiving, or distributing environmental datasets. In parallel, potentially vast amounts of environmental data are emerging due to new technologies such as high resolution imagery and machine learning.

Demographic and prosocial intrapersonal characteristics of biobank participants and refusers: the findings of a survey in the Netherlands

Research in genetics relies heavily on voluntary contributions of personal data. We aimed to acquire insights into the differences between participants and refusers of participation in a Dutch population-based biobank. Accordingly, we assessed the demographic and prosocial intrapersonal characteristics of respondents who participated (n = 2615) or refused to participate (n = 404) in the Lifelines biobank and databank. Our results indicated that health-related values critically influence participation decisions. The participation threshold for Lifelines was determined by an absence of health-related values and of trust in government. Therefore, considering these factors in communication and recruitment strategies could enhance participation in biomedical research. No indications were found of a stronger general prosociality of participants or their trust in researchers beyond the context of biobanking. This emphasizes the contextual understanding of the decision of participation in biobanking. Our findings may contribute to improving recruitment strategies by incorporating relevant values and/or highlighting prosocial benefits. Moreover, they foreground the need to address trust issues in collaborations between data repositories and commercial companies. Future research should explore how prosocial intrapersonal characteristics drive participation and withdrawal decisions and relate to contextual attributes.

Estimating prevalence of subjective cognitive decline in and across international cohort studies of aging: a COSMIC study

Background

Subjective cognitive decline (SCD) is recognized as a risk stage for Alzheimer’s disease (AD) and other dementias, but its prevalence is not well known. We aimed to use uniform criteria to better estimate SCD prevalence across international cohorts.

Methods

We combined individual participant data for 16 cohorts from 15 countries (members of the COSMIC consortium) and used qualitative and quantitative (Item Response Theory/IRT) harmonization techniques to estimate SCD prevalence.

Results

The sample comprised 39,387 cognitively unimpaired individuals above age 60. The prevalence of SCD across studies was around one quarter with both qualitative harmonization/QH (23.8%, 95%CI = 23.3–24.4%) and IRT (25.6%, 95%CI = 25.1–26.1%); however, prevalence estimates varied largely between studies (QH 6.1%, 95%CI = 5.1–7.0%, to 52.7%, 95%CI = 47.4–58.0%; IRT: 7.8%, 95%CI = 6.8–8.9%, to 52.7%, 95%CI = 47.4–58.0%). Across studies, SCD prevalence was higher in men than women, in lower levels of education, in Asian and Black African people compared to White people, in lower- and middle-income countries compared to high-income countries, and in studies conducted in later decades.

Conclusions

SCD is frequent in old age. Having a quarter of older individuals with SCD warrants further investigation of its significance, as a risk stage for AD and other dementias, and of ways to help individuals with SCD who seek medical advice. Moreover, a standardized instrument to measure SCD is needed to overcome the measurement variability currently dominant in the field.

Leveraging real-world data to investigate multiple sclerosis disease behavior, prognosis, and treatment

Randomized controlled clinical trials and real-world observational studies provide complementary information but with different validity. Some clinical questions (disease behavior, prognosis, validation of outcome measures, comparative effectiveness, and long-term safety of therapies) are often better addressed using real-world data reflecting larger, more representative populations. Integration of disease history, clinician-reported outcomes, performance tests, and patient-reported outcome measures during patient encounters; imaging and biospecimen analyses; and data from wearable devices increase dataset utility. However, observational studies utilizing these data are susceptible to many potential sources of bias, creating barriers to acceptance by regulatory agencies and the medical community. Therefore, data standardization and validation within datasets, harmonization across datasets, and application of appropriate analysis methods are important considerations. We review approaches to improve the scope, quality, and analyses of real-world data to advance understanding of multiple sclerosis and its treatment, as an example of opportunities to better support patient care and research.

Overview of Federated Facility to Harmonize, Analyze and Management of Missing Data in Cohorts

Cohorts are instrumental for epidemiologically oriented observational studies. Cohort studies usually observe large groups of individuals for a specific period of time to identify the contributing factors to a specific outcome (for instance an illness) and create associations between risk factors and the outcome under study. In collaborative projects, federated data facilities are meta-database systems that are distributed across multiple locations that permit to analyze, combine, or harmonize data from different sources making them suitable for mega- and meta-analyses. The harmonization of data can increase the statistical power of studies through maximization of sample size, allowing for additional refined statistical analyses, which ultimately lead to answer research questions that could not be addressed while using a single study. Indeed, harmonized data can be analyzed through mega-analysis of raw data or fixed effects meta-analysis. Other types of data might be analyzed by e.g., random-effects meta-analyses or Bayesian evidence synthesis. In this article, we describe some methodological aspects related to the construction of a federated facility to optimize analyses of multiple datasets, the impact of missing data, and some methods for handling missing data in cohort studies.

Meta-analysis of non-linear exposure-outcome relationships using individual participant data: A comparison of two methods

Non-linear exposure-outcome relationships such as between body mass index (BMI) and mortality are common. They are best explored as continuous functions using individual participant data from multiple studies. We explore two two-stage methods for meta-analysis of such relationships, where the confounder-adjusted relationship is first estimated in a non-linear regression model in each study, then combined across studies. The "metacurve" approach combines the estimated curves using multiple meta-analyses of the relative effect between a given exposure level and a reference level. The "mvmeta" approach combines the estimated model parameters in a single multivariate meta-analysis. Both methods allow the exposure-outcome relationship to differ across studies. Using theoretical arguments, we show that the methods differ most when covariate distributions differ across studies; using simulated data, we show that mvmeta gains precision but metacurve is more robust to model mis-specification. We then compare the two methods using data from the Emerging Risk Factors Collaboration on BMI, coronary heart disease events, and all-cause mortality (>80 cohorts, >18 000 events). For each outcome, we model BMI using fractional polynomials of degree 2 in each study, with adjustment for confounders. For metacurve, the powers defining the fractional polynomials may be study-specific or common across studies. For coronary heart disease, metacurve with common powers and mvmeta correctly identify a small increase in risk in the lowest levels of BMI, but metacurve with study-specific powers does not. For all-cause mortality, all methods identify a steep U-shape. The metacurve and mvmeta methods perform well in combining complex exposure-disease relationships across studies.

Online Family Problem-solving Treatment for Pediatric Traumatic Brain Injury

BACKGROUND AND OBJECTIVES

To determine whether online family problem-solving treatment (OFPST) is more effective in improving behavioral outcomes after pediatric traumatic brain injury with increasing time since injury.

METHODS

This was an individual participant data meta-analysis of outcome data from 5 randomized controlled trials of OFPST conducted between 2003 and 2016. We included 359 children ages 5 to 18 years who were hospitalized for moderate-to-severe traumatic brain injury 1 to 24 months earlier. Outcomes, assessed pre- and posttreatment, included parent-reported measures of externalizing, internalizing, and executive function behaviors and social competence.

RESULTS

Participants included 231 boys and 128 girls with an average age at injury of 13.6 years. Time since injury and age at injury moderated OFPST efficacy. For earlier ages and short time since injury, control participants demonstrated better externalizing problem scores than those receiving OFPST (Cohen's d = 0.44; P = .008; n = 295), whereas at older ages and longer time since injury, children receiving OFPST had better scores (Cohen's d = -0.60; P = .002). Children receiving OFPST were rated as having better executive functioning relative to control participants at a later age at injury, with greater effects seen at longer (Cohen's d = -0.66; P = .009; n = 298) than shorter (Cohen's d = -0. 28; P = .028) time since injury.

CONCLUSIONS

OFPST may be more beneficial for older children and when begun after the initial months postinjury. With these findings, we shed light on the optimal application of family problem-solving treatments within the first 2 years after injury.

Examination of Injury, Host, and Social-Environmental Moderators of Online Family Problem Solving Treatment Efficacy for Pediatric Traumatic Brain Injury Using an Individual Participant Data Meta-Analytic Approach

A series of five randomized controlled clinical trials (RCTs) conducted between 2002 and 2015 support the potential efficacy of online family problem-solving treatment (OFPST) in improving both child and parent/family outcomes after pediatric traumatic brain injury (TBI). However, small sample sizes and heterogeneity across individual studies have precluded examination of potentially important moderators. We jointly analyzed individual participant data (IPD) from these five RCTs, involving 359 children and adolescents between the ages of 5 and 18 years, to confirm the role of previously identified moderators (child's age and pre-treatment symptom levels, parental education) and to examine other potential moderators (race, sex, IQ), using IPD meta-analysis. This reanalysis revealed statistically strong evidence that parental education, child age at baseline, IQ, sex, and parental depression level pre-treatment moderated the effect of OFPST on various outcomes. In particular, children of parents with a less than high school education exhibited fewer internalizing problems and better social competence. Children injured at an older age exhibited fewer externalizing behaviors and less executive dysfunction following OFPST. Child IQ moderated the effect of OFPST on social competence, with significantly better competence for children with lower IQ who received OFPST. Lower levels of parental depression followed OFPST among subgroups with lower IQ, boys, and higher parental depression scores at baseline. Our findings indicate that the optimal application of OFPST is likely to involve older children, those with lower IQ scores, or those from families with lower socioeconomic status (SES).

Software Application Profile: Opal and Mica: open-source software solutions for epidemiological data management, harmonization and dissemination

Motivation

Improving the dissemination of information on existing epidemiological studies and facilitating the interoperability of study databases are essential to maximizing the use of resources and accelerating improvements in health. To address this, Maelstrom Research proposes Opal and Mica, two inter-operable open-source software packages providing out-of-the-box solutions for epidemiological data management, harmonization and dissemination.

Implementation

Opal and Mica are two standalone but inter-operable web applications written in Java, JavaScript and PHP. They provide web services and modern user interfaces to access them.

General features

Opal allows users to import, manage, annotate and harmonize study data. Mica is used to build searchable web portals disseminating study and variable metadata. When used conjointly, Mica users can securely query and retrieve summary statistics on geographically dispersed Opal servers in real-time. Integration with the DataSHIELD approach allows conducting more complex federated analyses involving statistical models.

Availability

Opal and Mica are open-source and freely available at [www.obiba.org] under a General Public License (GPL) version 3, and the metadata models and taxonomies that accompany them are available under a Creative Commons licence.

Harmonising data on the correlates of physical activity and sedentary behaviour in young people: Methods and lessons learnt from the international Children's Accelerometry database (ICAD)

Background

Large, heterogeneous datasets are required to enhance understanding of the multi-level influences on children’s physical activity and sedentary behaviour. One route to achieving this is through the pooling and co-analysis of data from multiple studies. Where this approach is used, transparency of the methodology for data collation and harmonisation is essential to enable appropriate analysis and interpretation of the derived data. In this paper, we describe the acquisition, management and harmonisation of non-accelerometer data in a project to expand the International Children’s Accelerometry Database (ICAD).

Method

Following a consultation process, ICAD partners were requested to share accelerometer data and information on selected behavioural, social, environmental and health-related constructs. All data were collated into a single repository for cataloguing and harmonisation. Harmonised variables were derived iteratively, with input from the ICAD investigators and a panel of invited experts. Extensive documentation, describing the source data and harmonisation procedure, was prepared and made available through the ICAD website.

Results

Work to expand ICAD has increased the number of studies with longitudinal accelerometer data, and expanded the breadth of behavioural, social and environmental characteristics that can be used as exposure variables. A set of core harmonised variables, including parent education, ethnicity, school travel mode/duration and car ownership, were derived for use by the research community. Guidance documents and facilities to enable the creation of new harmonised variables were also devised and made available to ICAD users. An expanded ICAD database was made available in May 2017.

Conclusion

The project to expand ICAD further demonstrates the feasibility of pooling data on physical activity, sedentary behaviour and potential determinants from multiple studies. Key to this process is the rigorous conduct and reporting of retrospective data harmonisation, which is essential to the appropriate analysis and interpretation of derived data. These documents, made available through the ICAD website, may also serve as a guide to others undertaking similar projects.

Big and disparate data: considerations for pediatric consortia

PURPOSE OF REVIEW

Increasingly, there is a need for examining exposure disease associations in large, diverse datasets to understand the complex determinants of pediatric disease and disability. Recognizing that children's health research consortia will be important sources of big data, it is crucial for the pediatric research community to be knowledgeable about the challenges and opportunities that they will face. The present review will provide examples of existing children's health consortia, highlight recent pooled analyses conducted by children's health research consortia, address common challenges of pooled analyses, and provide recommendations to advance collective research efforts in pediatric research.

RECENT FINDINGS

Formal consortia and other collective-science initiatives are increasingly being created to share individual data from a set of relevant epidemiological studies to address a common research topic under the concept that the joint effort of many individual groups can accomplish far more than working alone. There are practical challenges to the participation of investigators within consortia that need to be addressed in order for them to work.

SUMMARY

Researchers who access consortia with data centers will be able to go far beyond their initial hypotheses and potentially accomplish research that was previously thought infeasible or too costly.

Enhanced Endometriosis Archiving Software (ENEAS): An Application for Storing, Retrieving, Comparing, and Sharing Data of Patients Affected by Endometriosis Integrated in the Daily Practice

Endometriosis is a chronic benign disease affecting women of fertile age, associated with pelvic pain and subfertility, often with negative impacts on quality of life. Meetings involving 5 gynecologists skilled in endometriosis management and 2 informatics technology consultants competent in data management and website administration were enlisted to create an endometriosis databank known as ENEAS (Enhanced Endometriosis Archiving Software). This processing system allows users to store, retrieve, compare, and correlate all data collected in conjunction with different Italian endometriosis centers, with the collective objective of obtaining homogeneous data for a large population sample. ENEAS is a web-oriented application that can be based on any open-source database that can be installed locally on a server, allowing collection of data on approximately 700 items, providing standardized and homogeneous data for comparison and analysis. ENEAS is capable of generating a sheet incorporating all data on the management of endometriosis that is both accurate and suitable for each individual patient. ENEAS is an effective and universal web application that encourages providers in the field of endometriosis to use a common language and share data to standardize medical and surgical treatment, with the main benefits being improved patient satisfaction and quality of life.

Comparison of Standardization Methods for the Harmonization of Phenotype Data: An Application to Cognitive Measures

Standardization procedures are commonly used to combine phenotype data that were measured using different instruments, but there is little information on how the choice of standardization method influences pooled estimates and heterogeneity. Heterogeneity is of key importance in meta-analyses of observational studies because it affects the statistical models used and the decision of whether or not it is appropriate to calculate a pooled estimate of effect. Using 2-stage individual participant data analyses, we compared 2 common methods of standardization, T-scores and category-centered scores, to create combinable memory scores using cross-sectional data from 3 Canadian population-based studies (the Canadian Study on Health and Aging (1991-1992), the Canadian Community Health Survey on Healthy Aging (2008-2009), and the Quebec Longitudinal Study on Nutrition and Aging (2004-2005)). A simulation was then conducted to assess the influence of varying the following items across population-based studies: 1) effect size, 2) distribution of confounders, and 3) the relationship between confounders and the outcome. We found that pooled estimates based on the unadjusted category-centered scores tended to be larger than those based on the T-scores, although the differences were negligible when adjusted scores were used, and that most individual participant data meta-analyses identified significant heterogeneity. The results of the simulation suggested that in terms of heterogeneity, the method of standardization played a smaller role than did different effect sizes across populations and differential confounding of the outcome measure across studies. Although there was general consistency between the 2 types of standardization methods, the simulations identified a number of sources of heterogeneity, some of which are not the usual sources considered by researchers.

Harmonization of Food-Frequency Questionnaires and Dietary Pattern Analysis in 4 Ethnically Diverse Birth Cohorts

BACKGROUND

Canada is an ethnically diverse nation, which introduces challenges for health care providers tasked with providing evidence-based dietary advice.

OBJECTIVES

We aimed to harmonize food-frequency questionnaires (FFQs) across 4 birth cohorts of ethnically diverse pregnant women to derive robust dietary patterns to investigate maternal and newborn outcomes.

METHODS

The NutriGen Alliance comprises 4 prospective birth cohorts and includes 4880 Canadian mother-infant pairs of predominantly white European [CHILD (Canadian Healthy Infant Longitudinal Development) and FAMILY (Family Atherosclerosis Monitoring In earLY life)], South Asian [START (SouTh Asian birth cohoRT)-Canada], or Aboriginal [ABC (Aboriginal Birth Cohort)] origins. CHILD used a multiethnic FFQ based on a previously validated instrument designed by the Fred Hutchinson Cancer Research Center, whereas FAMILY, START, and ABC used questionnaires specifically designed for use in white European, South Asian, and Aboriginal people, respectively. The serving sizes and consumption frequencies of individual food items within the 4 FFQs were harmonized and aggregated into 36 common food groups. Principal components analysis was used to identify dietary patterns that were internally validated against self-reported vegetarian status and externally validated against a modified Alternative Healthy Eating Index (mAHEI).

RESULTS

Three maternal dietary patterns were identified-"plant-based," "Western," and "health-conscious"-which collectively explained 29% of the total variability in eating habits observed in the NutriGen Alliance. These patterns were strongly associated with self-reported vegetarian status (OR: 3.85; 95% CI: 3.47, 4.29; r² = 0.30, P < 0.001; for a plant-based diet), and average adherence to the plant-based diet was higher in participants in the fourth quartile of the mAHEI than in the first quartile (mean difference: 46.1%; r² = 0.81, P < 0.001).

CONCLUSION

Dietary data collected by using FFQs from ethnically diverse pregnant women can be harmonized to identify common dietary patterns to investigate associations between maternal dietary intake and health outcomes.

Toward Rigorous Data Harmonization in Cancer Epidemiology Research: One Approach

Cancer epidemiologists have a long history of combining data sets in pooled analyses, often harmonizing heterogeneous data from multiple studies into 1 large data set. Although there are useful websites on data harmonization with recommendations and support, there is little research on best practices in data harmonization; each project conducts harmonization according to its own internal standards. The field would be greatly served by charting the process of data harmonization to enhance the quality of the harmonized data. Here, we describe the data harmonization process utilized at the Fred Hutchinson Cancer Research Center (Seattle, Washington) by the coordinating centers of several research projects. We describe a 6-step harmonization process, including: 1) identification of questions the harmonized data set is required to answer; 2) identification of high-level data concepts to answer those questions; 3) assessment of data availability for data concepts; 4) development of common data elements for each data concept; 5) mapping and transformation of individual data points to common data elements; and 6) quality-control procedures. Our aim here is not to claim a "correct" way of doing data harmonization but to encourage others to describe their processes in order that we can begin to create rigorous approaches. We also propose a research agenda around this issue.

Get real in individual participant data (IPD) meta-analysis: a review of the methodology

Individual participant data (IPD) meta-analysis is an increasingly used approach for synthesizing and investigating treatment effect estimates. Over the past few years, numerous methods for conducting an IPD meta-analysis (IPD-MA) have been proposed, often making different assumptions and modeling choices while addressing a similar research question. We conducted a literature review to provide an overview of methods for performing an IPD-MA using evidence from clinical trials or non-randomized studies when investigating treatment efficacy. With this review, we aim to assist researchers in choosing the appropriate methods and provide recommendations on their implementation when planning and conducting an IPD-MA.

[Harmonisation of databases for the study of frailty in older people: INTAFRADE study]

OBJECTIVES

The main objective of the present work is to evaluate the feasibility of harmonising the available information from different independent databases, in order to build an integrated database to study frailty.

MATERIAL AND METHODS

This work is based on the European project, Integral Approach to the Transition between Frailty and Dependence on older adults: Patterns of occurrence, identification tools and model of care (INTAFRADE), developed by 4 groups, 3 in Spain and one in France. Each partner provided their databases related to the study of frailty. As a previous step to the creation of an integrated database the characteristics and variables included in each study were mapped, specifying whether their harmonisation was possible or not.

RESULTS

A total of 30 different variables that corresponded to 8 dimensions were identified: Sociodemographic and social characteristics, health status, lifestyle habits, anthropometric measures, other physical measurements, use of health services, and adverse health results. Of them all, 28 (93%) variables were harmonisable, although only 20% were present in all databases, with 47% in 3 of them. In relation to the frailty instruments, all of them were lacking at least 50% of the items. The harmonisation process will allow us to jointly analyse information available on 2,361 people.

CONCLUSIONS

The European INTAFRADE study will allow a deeper understanding of the frailty process in older people by harmonising information from heterogeneous databases.

Statistical approaches to harmonize data on cognitive measures in systematic reviews are rarely reported

OBJECTIVES

To identify statistical methods for harmonization, the procedures aimed at achieving the comparability of previously collected data, which could be used in the context of summary data and individual participant data meta-analysis of cognitive measures.

STUDY DESIGN AND SETTING

Environmental scan methods were used to conduct two reviews to identify (1) studies that quantitatively combined data on cognition and (2) general literature on statistical methods for data harmonization. Search results were rapidly screened to identify articles of relevance.

RESULTS

All 33 meta-analyses combining cognition measures either restricted their analyses to a subset of studies using a common measure or combined standardized effect sizes across studies; none reported their harmonization steps before producing summary effects. In the second scan, three general classes of statistical harmonization models were identified (1) standardization methods, (2) latent variable models, and (3) multiple imputation models; few publications compared methods.

CONCLUSION

Although it is an implicit part of conducting a meta-analysis or pooled analysis, the methods used to assess inferential equivalence of complex constructs are rarely reported or discussed. Progress in this area will be supported by guidelines for the conduct and reporting of the data harmonization and integration and by evaluating and developing statistical approaches to harmonization.

World Endometriosis Research Foundation Endometriosis Phenome and Biobanking Harmonization Project: II. Clinical and covariate phenotype data collection in endometriosis research

OBJECTIVE

To harmonize the collection of nonsurgical clinical and epidemiologic data relevant to endometriosis research, allowing large-scale collaboration.

DESIGN

An international collaboration involving 34 clinical/academic centers and three industry collaborators from 16 countries on five continents.

SETTING

In 2013, two workshops followed by global consultation, bringing together 54 leaders in endometriosis research.

PATIENTS

None.

INTERVENTION(S)

Development of a self-administered endometriosis patient questionnaire (EPQ), based on [1] systematic comparison of questionnaires from eight centers that collect data from endometriosis cases (and controls/comparison women) on a medium to large scale (publication on >100 cases); [2] literature evidence; and [3] several global consultation rounds.

MAIN OUTCOME MEASURE(S)

Standard recommended and minimum required questionnaires to capture detailed clinical and covariate data.

RESULT(S)

The standard recommended (EPHect EPQ-S) and minimum required (EPHect EPQ-M) questionnaires contain questions on pelvic pain, subfertility and menstrual/reproductive history, hormone/medication use, medical history, and personal information.

CONCLUSION(S)

The EPQ captures the basic set of patient characteristics and exposures considered by the WERF EPHect Working Group to be most critical for the advancement of endometriosis research, but is also relevant to other female conditions with similar risk factors and/or symptomatology. The instruments will be reviewed based on feedback from investigators, and-after a first review after 1 year-triannually through systematic follow-up surveys. Updated versions will be made available through http://endometriosisfoundation.org/ephect.

Data harmonization and federated analysis of population-based studies: the BioSHaRE project

Abstracts

[Epidemiological studies based on medical and administrative databases : a potential strength in France]

Population-based epidemiological cohorts may include nowadays hundreds of thousands of subjects, followed-up during decades. France has a major potential strength: nationwide medical and social databases set up for administrative purposes. The main databases useful for epidemiology are the social security database which contains individual medical data from different sources, and the retirement fund database on employment and social benefits. These databases have several advantages: they cover the whole French population, with no lost to follow-up, data are often of good quality and it is possible to link them with individual surveys. However medical data are not always ascertained and an important methodological and practical work has to be done, and some legal and practical problems have to be solved for an optimal use.

Age and gender differences in physical capability levels from mid-life onwards: the harmonisation and meta-analysis of data from eight UK cohort studies

Using data from eight UK cohorts participating in the Healthy Ageing across the Life Course (HALCyon) research programme, with ages at physical capability assessment ranging from 50 to 90+ years, we harmonised data on objective measures of physical capability (i.e. grip strength, chair rising ability, walking speed, timed get up and go, and standing balance performance) and investigated the cross-sectional age and gender differences in these measures. Levels of physical capability were generally lower in study participants of older ages, and men performed better than women (for example, results from meta-analyses (N = 14,213 (5 studies)), found that men had 12.62 kg (11.34, 13.90) higher grip strength than women after adjustment for age and body size), although for walking speed, this gender difference was attenuated after adjustment for body size. There was also evidence that the gender difference in grip strength diminished with increasing age,whereas the gender difference in walking speed widened (p<0.01 for interactions between age and gender in both cases). This study highlights not only the presence of age and gender differences in objective measures of physical capability but provides a demonstration that harmonisation of data from several large cohort studies is possible. These harmonised data are now being used within HALCyon to understand the lifetime social and biological determinants of physical capability and its changes with age.

Cardiovascular and metabolic influences of fetal smoke exposure

Many epidemiological studies showed associations of low birth weight with cardiovascular disease, type 2 diabetes and obesity. The associations seem to be consistent and stronger among subjects with a postnatal catch up growth. It has been suggested that developmental changes in response to adverse fetal exposures might lead to changes in the fetal anatomy and physiology. These adaptations may be beneficial for short term, but may lead to common diseases in adulthood. Maternal smoking during pregnancy is one of the most important adverse fetal exposures in Western countries, and is known to be associated with a 150-200 g lower birth weight. An accumulating body of evidence suggests that maternal smoking during pregnancy might be involved in pathways leading to both low birth weight and common diseases, including cardiovascular disease, type 2 diabetes and obesity, in adulthood. In this review, we discuss epidemiological studies focused on the associations of maternal smoking with fetal growth and development and cardiovascular and metabolic disease in later life. We also discuss potential biological mechanisms, and challenges for future epidemiological studies.

Cardiovascular and metabolic influences of fetal smoke exposure

Many epidemiological studies showed associations of low birth weight with cardiovascular disease, type 2 diabetes and obesity. The associations seem to be consistent and stronger among subjects with a postnatal catch up growth. It has been suggested that developmental changes in response to adverse fetal exposures might lead to changes in the fetal anatomy and physiology. These adaptations may be beneficial for short term, but may lead to common diseases in adulthood. Maternal smoking during pregnancy is one of the most important adverse fetal exposures in Western countries, and is known to be associated with a 150–200 g lower birth weight. An accumulating body of evidence suggests that maternal smoking during pregnancy might be involved in pathways leading to both low birth weight and common diseases, including cardiovascular disease, type 2 diabetes and obesity, in adulthood. In this review, we discuss epidemiological studies focused on the associations of maternal smoking with fetal growth and development and cardiovascular and metabolic disease in later life. We also discuss potential biological mechanisms, and challenges for future epidemiological studies.

Invited commentary: consolidating data harmonization--how to obtain quality and applicability?

It is recognized that very large sample sizes capable of providing adequate statistical power are required to properly investigate and understand the role and interaction of genetic, lifestyle, and environmental factors in modulating the risk and progression of chronic diseases. However, very few one-off studies provide access to very large numbers of participants, and the collection of high-quality data necessitates a major investment of resources. The scientific community is thus increasingly engaged in collaborative efforts to facilitate harmonization and synthesis of data across studies. Complementary harmonization approaches may be adopted to support these efforts. In the current issue of the American Journal of Epidemiology, Hamilton et al. (Am J Epidemiol. 2011;174(3):253-260) present the consensus measures for Phenotypes and eXposures (PhenX) Toolkit, which promotes the use of identical data collection tools and procedures to support harmonization across emerging studies. Data synthesis is greatly facilitated by the use of common measures and procedures. However, the "stringent" criteria required by PhenX can limit its utilization. The opportunity to make use of rigorous but more "flexible" harmonization approaches should also be considered.

Is rigorous retrospective harmonization possible? Application of the DataSHaPER approach across 53 large studies

BACKGROUND

Proper understanding of the roles of, and interactions between genetic, lifestyle, environmental and psycho-social factors in determining the risk of development and/or progression of chronic diseases requires access to very large high-quality databases. Because of the financial, technical and time burdens related to developing and maintaining very large studies, the scientific community is increasingly synthesizing data from multiple studies to construct large databases. However, the data items collected by individual studies must be inferentially equivalent to be meaningfully synthesized. The DataSchema and Harmonization Platform for Epidemiological Research (DataSHaPER; http://www.datashaper.org) was developed to enable the rigorous assessment of the inferential equivalence, i.e. the potential for harmonization, of selected information from individual studies.

METHODS

This article examines the value of using the DataSHaPER for retrospective harmonization of established studies. Using the DataSHaPER approach, the potential to generate 148 harmonized variables from the questionnaires and physical measures collected in 53 large population-based studies (6.9 million participants) was assessed. Variable and study characteristics that might influence the potential for data synthesis were also explored.

RESULTS

Out of all assessment items evaluated (148 variables for each of the 53 studies), 38% could be harmonized. Certain characteristics of variables (i.e. relative importance, individual targeted, reference period) and of studies (i.e. observational units, data collection start date and mode of questionnaire administration) were associated with the potential for harmonization. For example, for variables deemed to be essential, 62% of assessment items paired could be harmonized.

CONCLUSION

The current article shows that the DataSHaPER provides an effective and flexible approach for the retrospective harmonization of information across studies. To implement data synthesis, some additional scientific, ethico-legal and technical considerations must be addressed. The success of the DataSHaPER as a harmonization approach will depend on its continuing development and on the rigour and extent of its use. The DataSHaPER has the potential to take us closer to a truly collaborative epidemiology and offers the promise of enhanced research potential generated through synthesized databases.

Comparison of formaldehyde exposure levels in two multi-industry occupational exposure databanks using multimodel inference

The recent increase in international multicenter epidemiologic studies of occupational hazards and the current development of risk analysis tools based on multinational exposure data provide an incentive to evaluate the extrapolation of exposure data across countries. We compared formaldehyde measurements in the French (COLCHIC) and U.S. (IMIS) occupational exposure databases for 1986-2001 using multimodel inference. In this approach, conclusions are based on a set of plausible models rather than on a single model. Modeled variables included data source, industry, year, and sample type (short-term or long-term). The model set included 72 models corresponding to testing all variables and 5 interactions and estimation of 93 coefficients. As it is impossible to classify non-detects (ND) as short-term or long-term samples, they were removed from IMIS prior to analysis. Respectively, 3143 and 2646 IMIS and COLCHIC data, spread across 28 common industries, were analyzed. The full model explained 21% of the total variability. All fixed effects and the source-industry interaction were shown as strong predictors of exposure. The temporal trend (5% yearly decrease) and difference between short-term and long-term data (short-term greater than long-term by two-fold) were similar across the two databanks. Industry-specific differences between IMIS and COLCHIC were detected only for health-related activities for which COLCHIC levels were higher (~ four-fold). Sensitivity analyses were conducted to assess the impact of excluding NDs from IMIS. They showed initial predicted industry-specific geometric means for IMIS data potentially multiplied by factors from 0.42 to 0.98 (long-term data) and from 0.11 to 1.4 (short-term data) when NDs were included with various distributions across the short-term and long-term categories. The most realistic scenario yielded 0.82 for long-term (18% overestimation) data and 1.05 (5% underestimation) for short-term data. Although tempered by a probable non-detect bias, our analysis showed that both databanks provide a similar multi-industry portrait of formaldehyde exposure despite a potential for very different occupational settings. Our results offer encouraging insight about extrapolation of exposure data across countries.

The Shanghai Changfeng Study: a community-based prospective cohort study of chronic diseases among middle-aged and elderly: objectives and design

The Shanghai Changfeng Study is a community-based prospective cohort study of chronic diseases ongoing since February 2009 in Shanghai, China. The study focuses on multiple chronic diseases, including obesity and metabolic syndrome, diabetes, osteoporosis, liver diseases, cardiovascular diseases and neurologic diseases. 15,000 subjects of 40 years or over are planned to be recruited. The rationale, objectives and design of this study are described in this paper.

The Generation R Study: design and cohort update 2010

The Generation R Study is a population-based prospective cohort study from fetal life until young adulthood. The study is designed to identify early environmental and genetic causes of normal and abnormal growth, development and health during fetal life, childhood and adulthood. The study focuses on four primary areas of research: (1) growth and physical development; (2) behavioural and cognitive development; (3) diseases in childhood; and (4) health and healthcare for pregnant women and children. In total, 9,778 mothers with a delivery date from April 2002 until January 2006 were enrolled in the study. General follow-up rates until the age of 4 years exceed 75%. Data collection in mothers, fathers and preschool children included questionnaires, detailed physical and ultrasound examinations, behavioural observations, and biological samples. A genome wide association screen is available in the participating children. Regular detailed hands on assessment are performed from the age of 5 years onwards. Eventually, results forthcoming from the Generation R Study have to contribute to the development of strategies for optimizing health and healthcare for pregnant women and children.

Quality, quantity and harmony: the DataSHaPER approach to integrating data across bioclinical studies

BACKGROUND

Vast sample sizes are often essential in the quest to disentangle the complex interplay of the genetic, lifestyle, environmental and social factors that determine the aetiology and progression of chronic diseases. The pooling of information between studies is therefore of central importance to contemporary bioscience. However, there are many technical, ethico-legal and scientific challenges to be overcome if an effective, valid, pooled analysis is to be achieved. Perhaps most critically, any data that are to be analysed in this way must be adequately 'harmonized'. This implies that the collection and recording of information and data must be done in a manner that is sufficiently similar in the different studies to allow valid synthesis to take place.

METHODS

This conceptual article describes the origins, purpose and scientific foundations of the DataSHaPER (DataSchema and Harmonization Platform for Epidemiological Research; http://www.datashaper.org), which has been created by a multidisciplinary consortium of experts that was pulled together and coordinated by three international organizations: P³G (Public Population Project in Genomics), PHOEBE (Promoting Harmonization of Epidemiological Biobanks in Europe) and CPT (Canadian Partnership for Tomorrow Project).

RESULTS

The DataSHaPER provides a flexible, structured approach to the harmonization and pooling of information between studies. Its two primary components, the 'DataSchema' and 'Harmonization Platforms', together support the preparation of effective data-collection protocols and provide a central reference to facilitate harmonization. The DataSHaPER supports both 'prospective' and 'retrospective' harmonization.

CONCLUSION

It is hoped that this article will encourage readers to investigate the project further: the more the research groups and studies are actively involved, the more effective the DataSHaPER programme will ultimately be.

Analysis of individual drug use as a time-varying determinant of exposure in prospective population-based cohort studies

In pharmaco-epidemiology, the use of drugs is the determinant of interest when studying exposure-outcome associations. The increased availability of computerized information about drug use on an individual basis has greatly facilitated analyses of drug effects on a population-based scale. It seems likely that many negative findings in the early days of pharmaco-epidemiology can be explained by non-differential misclassification because of too simple (yes/no) exposure measures. In this paper, the authors discuss the importance of an adequate definition of drug exposure in pharmaco-epidemiological research and how this time-varying determinant can be analyzed in cohort studies. To reduce the risk of non-differential misclassification, a precise definition of exposure is mandatory and it is important to distinguish the complete follow-up period of a population into mutually exclusive episodes of non-use, past use and current use for each individual. By analyzing exposure to drugs as a time-dependent variable in a Cox regression model, cohort studies with complete coverage of all filled prescriptions can provide us with valid and precise risk estimates of drug-outcome associations. However, such estimates may be biased in the presence of time-dependent confounders which are themselves affected by prior exposure.