Sleep and Circadian Informatics Data Harmonization: A Workshop Report from the Sleep Research Society and Sleep Research Network

The increasing availability and complexity of sleep and circadian data are equally exciting and challenging. The field is in constant technological development, generating better high-resolution physiological and molecular data than ever before. Yet, the promise of large-scale studies leveraging millions of patients is limited by suboptimal approaches for data sharing and interoperability. As a result, integration of valuable clinical and basic resources is problematic, preventing knowledge discovery and rapid translation of findings into clinical care. To understand the current data landscape in the sleep and circadian domains, the Sleep Research Society (SRS) and the Sleep Research Network (now a task force of the SRS) organized a workshop on informatics and data harmonization, presented at the World Sleep Congress 2019, in Vancouver, Canada. Experts in translational informatics gathered with sleep research experts to discuss opportunities and challenges in defining strategies for data harmonization. The goal of this workshop was to fuel discussion and foster innovative approaches for data integration and development of informatics infrastructure supporting multi-site collaboration. Key recommendations included collecting and storing findable, accessible, interoperable, and reusable data; identifying existing international cohorts and resources supporting research in sleep and circadian biology; and defining the most relevant sleep data elements and associated metadata that could be supported by early integration initiatives. This report introduces foundational concepts with the goal of facilitating engagement between the sleep/circadian and informatics communities and is a call to action for the implementation and adoption of data harmonization strategies in this domain.

Data Harmonization and Data Pooling from Cohort Studies: A Practical Approach for Data Management

Data pooling from pre-existing multiple datasets can be useful to increase study sample size and statistical power to answer a research question. However, individual datasets may contain variables that measure the same construct differently, posing challenges for data pooling. Variable harmonization, an approach that can generate comparable datasets from heterogeneous sources, can address this issue in some circumstances. As an illustrative example, this paper describes the data harmonization strategies that helped generate comparable datasets across two Canadian pregnancy cohort studies– the All Our Families and the Alberta Pregnancy Outcomes and Nutrition. Variables were harmonized considering multiple features across the datasets: the construct measured; question asked/response options; the measurement scale used; the frequency of measurement; timing of measurement, and the data structure. Completely matching, partially matching, and completely un-matching variables across the datasets were determined based on these features. Variables that were an exact match were pooled as is. Partially matching variables were synchronized across the datasets considering the frequency of measurement, the timing of measurement, and response options. Variables that were completely unmatching could not be harmonized into a single variable. The variable harmonization strategies that were used to generate comparable cohort datasets for data pooling are applicable to other data sources. Future studies may employ or evaluate these strategies. Variable harmonization and pooling provide an opportunity to increase study power and the utility of existing data, permitting researchers to answer novel research questions in a statistically efficient, timely, and cost-efficient manner that could not be achieved using a single data source.