All data and associated meta-data considered for analysis in the DMS study has been collected and we will not be including any further information in the analysis. However, if you’ve shared data with WWARN, feel free to revisit your space and complete the records we have on your study. Analysis for the DMS study will start by focusing on studies performed on the African continent, and we expect to report on first results by the end of the year.

Appropriate data management (DM) is critical to ascertain the value of research data, especially given the growing trend towards long-term data archiving, sharing and re-use of data sets. Clear, transparent DM helps to ascertain the reliability of data, and secondary users must mostly rely on DM plans or study protocols to interpret the data sets they re-use and to get a sense of their quality. Once a trial is finished, it is difficult to review the study site(s) to assess the robustness of the initial data management system (DMS) and the validity of resulting data thereafter. Thorough documentation of study data and associated processes therefore becomes critical to ensure the re-usability and appropriate interpretation of data over time.

Since its inception in 2009, the WorldWide Antimalarial Resistance Network (WWARN) has grown into a comprehensive data management platform that includes more than 300 antimalarial study datasets. The experience of gathering and handling these data from multiple partners has highlighted significant variability in the DM practices that are initially put in place by researchers to generate, manage, and ultimately share their study data. Variations include the type of initial software and format of the shared dataset, and the levels of detail in the provided study documentation. Thus far, the interlinkages between variability in DM practices and the resulting study data integrity have been little explored; and the reasons underpinning the choice of one DM strategy over another remain unknown. As part of continuous efforts to understand the needs of malaria researchers, and specifically the needs in DM resources and guidance for transparent practices optimising long-term secondary use of clinical research data, WWARN is performing a DM methodology and needs assessment study.

This Data Management Systems Study seeks to scope or identify improvements for resources and tools which could support malaria researchers in their data management and sharing activities – for example, recommended procedures for DM in future trials, or re-usable training materials.

Objectives

• To characterise variations in clinical research data management practices – through systematic examination of datasets, together with their associated meta-data and study documentation (held within the WWARN data-sharing platform of antimalarial trials).
• To explore patterns and correlations between indicators of data integrity, specific DM practices and broader study characteristics – to clarify the preferred approaches of different data owners, and their potential impact on data quality.

All antimalarial clinical trials with data curated and mapped within the WWARN data platform as of 1^st of November 2016 are eligible for this DM methodology and needs assessment study. Data contributors will be contacted in January 2017 to seek their interest in this work: they will be invited to submit additional study documentation (such as protocol, data management plans or variables’ dictionary) if not provided at the time of initial data contribution to WWARN.

For each included study, a mix of categorical and numerical information will be collected to highlight attributes of:

1. the dataset itself – e.g. type of data management software used, frequency of out-of-range or inconsistent values as identified during mapping of the dataset to the WWARN structure;
2.  the associated meta-data – e.g. availability and comprehensiveness of data management plans or variables’ dictionary; and
3. the study – including sponsor/funder, time and place of study, type of study, country of principal investigator, etc.

Such information will be extracted from study supporting documentation and published articles, as well as from reports compiled over the course of data curation by WWARN. Clustering algorithms and statistical approaches will then be used to identify patterns and explore correlations between the three areas of interest – indicators of DM practices, measures of data integrity, and study attributes.

Note: This systematic review of all eligible trial data contributed to WWARN constitutes the quantitative part of a broader project. In parallel, in-depth qualitative case studies are planned at partner sites based in Africa. Those aim at highlighting the local DM practices and constraints faced by research teams operating in Low and Middle Income Countries (LMICs).

The DM Systems Study is coordinated by Amélie Julé and Hazel Ashurst. This work is not using the individual-level participant data, as such we are not forming a WWARN Study Group. Apart from the effort of retrieving and sharing missing information about studies (if applicable), we do not anticipate any further analysis contribution from data contributors. We may have to contact you for additional information, but we will keep this contact to a minimum.

Datasets remain the property of the investigator. Investigators will be informed of the initial findings by mid-2017, and we later aim to publish a report in an open-access journal. Study investigators will be acknowledged in any publication.

Please contact data@wwarn.org if you have any query about this project.

European Commission, 2016. Commission Implementing Regulation (EU) 2016/9 of 5 January 2016 on joint submission of data and data-sharing in accordance with Regulation (EC) No 1907/2006 of the European Parliament and of the Council concerning the Registration, Evaluation, Authorisat. Official Journal of the European Union, 3, pp.41–45.

European Medicines Agency, 2013. European Medicines Agency policy on publication of clinical data for medicinal products for human use,

Institute of Medicine, 2015. Sharing Clinical Trial Data: Maximizing Benefits, Minimizing Risks, Washington DC: The National Academies Press. Available at: https://www.nap.edu/catalog/18998/sharing-clinical-trial-data-maximizing-benefits-minimizing-risk.

Merson, L., Gaye, O. & Guerin, P.J., 2016. Avoiding Data Dumpsters — Toward Equitable and Useful Data Sharing. New England Journal of Medicine, p.NEJMp1605148. Available at: https://www.nejm.org/doi/10.1056/NEJMp1605148 [Accessed June 14, 2016].

Swan, A. & Brown, S., 2008. To share or not to share: Publication and quality assurance of research data outputs. A report commissioned by the Research Information Network. , (June). Available at: https://eprints.soton.ac.uk/266742/.

Taichman, D.B. et al., 2016. Sharing Clinical Trial Data--A Proposal from the International Committee of Medical Journal Editors. The New England journal of medicine, 374(4), pp.384–6. Available at: https://www.ncbi.nlm.nih.gov/pubmed/26786954 [Accessed June 27, 2016].

Walport, M. & Brest, P., 2011. Sharing research data to improve public health. Lancet, 377(9765), pp.537–9. Available at: https://www.sciencedirect.com/science/article/pii/S0140673610622349.

Wellcome Trust, Sharing research data to improve public health: full joint statement by funders of health research. Available at: https://wellcome.ac.uk/what-we-do/our-work/sharing-research-data-improve-public-health-full-joint-statement-funders-health [Accessed June 27, 2016].