Maximising the impact of data sharing, for everyone

Georgina Humphreys, Karen Barnes and Halidou Tinto recently published a perspective piece in The American Journal of Tropical Medicine and Hygiene suggesting that the requirement to share data won’t necessarily provide the incentives needed for researchers to prepare and document shared data in a way that makes it usable by others.

A decade of experience has taught the WWARN team that recognition and reward for those who generate and share research data is fundamental for data sharing to be scientifically valuable.

So how did WWARN develop an effective data-sharing model?

WWARN launched in 2009 to support the malaria community to understand and slow the extent and impact of antimalarial drug resistance. The WWARN community quickly realised that defining a targeted research question, and inviting those who had relevant data sets to form a collaborative Study Group, could be the most effective way to answer important questions on antimalarial drug resistance.  

This Study Group approach developed to allow researchers with relevant data to participate in large individual patient data (IPD) analyses and publish results together as a collective. Study Groups begin with the definition of a question that is important to malaria-affected communities. Questions such as how best to protect women from malaria during pregnancy, or the interactions between malnutrition on malaria, to understand how these conditions could influence drug efficacy and patient outcomes. To date the WWARN Network of partners has coordinated 27 Study Groups, 11 of which have published results; a further 16 groups are actively gathering and analysing data. 

Karen Barnes, WWARN Head of Pharmacology explains in our latest perspective piece, “We developed an approach to define a common goal by collaboratively agreeing the best methods to answer a key public health question. This approach reassured colleagues of our intention: to find ways to combine data in order to make the best use of available data. We started to see the benefits of having a much larger data set from which more powerful insights on malaria drug resistance could be derived, particularly in vulnerable populations that are usually under-represented in clinical studies.” (Read: Strength in Numbers: The WWARN Case Study of Purpose-Driven Data Sharing)

WWARN’s Study Groups can answer difficult questions thanks to the input of multiple researchers who share both their data and their expertise. For example, we included over 5,000 malaria episodes from 20 studies in WWARN’s Malaria in Pregnancy Treatment Efficacy Study Group to assess the efficacy of a range of antimalarials used for the treatment of P. falciparum malaria in all trimesters of pregnancy in Africa and Asia. The data set included 90% of pregnant patients enrolled in treatment efficacy studies identified in a search of the published literature. Such a comprehensive IPD analysis would not be possible without collaboration and data contribution from a large group of researchers, mostly from Asia and Africa.

Effective malaria treatment requires special attention to particular subgroups of patients including malnourished children or pregnant women, for example.  This is only feasible when the overall data set is large enough to include significant numbers of these sub-groups.  

Philippe Guerin adds, “By developing large-scale individual patient data analyses, we are supporting the World Health Organization’s Malaria strategy which calls for more targeted intervention approaches.”

We routinely share WWARN Study Group results with experts from the World Health Organization and global health communities, enabling policy makers to use this evidence to adjust treatment policies, develop new surveillance tactics or identify further gaps in the data needed to make decisions.

Oumar Gaye, Head of WWARN’s West Africa Regional Centre explains, “Through analysing large data sets we can also improve our understanding of where there are gaps in data, such as in certain countries or target populations. We can know where to target new trials more efficiently”.

The WWARN’s Parasite Clearance Study Group is an example of how a Study Group can also be used to create and test new tools. This individual-patient data analysis combined more than 6,900 patients enrolled in 24 studies with frequent parasite counts. These results were then used to validate WWARN’s Parasite Clearance Estimator, a tool developed for the research community to standardise estimates of the parasite clearance rates following malaria treatment, the tool has been cited by more than 100 subsequent peer-reviewed publications.

The WWARN publishing model recognises everyone who shared and analysed data by recognising them as authors or data collaborators on the resulting publications. As Georgina Humphreys highlights in her perspective piece, ‘our approach recognises research contribution and incentivises researchers to engage in purpose-driven data sharing, analysis and reuse.’

The Infectious Diseases Data Observatory (IDDO) is building on the experiences of WWARN to develop a collaborative, multi-disease data model for other emerging infections and neglected tropical diseases such as Ebola and visceral leishmaniasis.

In a recent opinion piece published in Lancet Global Health, ‘Secondary analysis and participation of those at the data source’, we emphasize the importance of inviting researchers responsible for data collection to participate in secondary analysis to ensure impactful, ethical, and equitable data sharing.

Laura Merson, Associate Director of IDDO explains, “Building on the Study Group approach, anyone who contributes data to our NTD research groups will be invited to participate in individual patient data meta-analysis and will be recognised with appropriate authorship. IDDO’s Ebola research theme also provides equitable access to data through a timely, transparent data-sharing process managed through an independent Data Access Committee. This approach respects the interests of those who collect the original data.”

Capacity building promotes more standardised research approaches. Tools and resources such as WWARN’s Malaria Toolkit are freely available to help researchers design, plan and collect data effectively and efficiently.

We also offer training and mentorship in data management, statistical analysis and manuscript writing to researchers collecting data so that they can enhance the quality of their research.  For example, we hosted a workshop on Improving Research Data to Inform Better Treatment of Poverty-Related Infectious Diseases in Nairobi, Kenya. The workshop welcomed more than 44 data managers, biostatisticians and researchers, primarily from sub-Saharan Africa.

By working together, recognising everyone’s contributions and supporting researchers working in endemic countries, the infectious diseases community can share data equitably to achieve better health research results from the data that already exists.

We continue to improve and refine our approach to data sharing, for example by supporting the creation of data standards. These tools will support our research partners to create good quality data that can be more easily shared and analysed by the global health community. See the CDISC* Malaria Therapeutic Area User Guide (TAUG-Malaria), which applies the international CDISC standards to malaria data to streamline malaria data capture and enable assessment of antimalarial drug efficacy. We will be releasing training tools to guide researchers on how to use the TAUG-malaria shortly.

We’re also making WWARN data available for other researchers to request for secondary analysis. Researchers can view datasets gathered by WWARN on the Malaria Data Inventory and request access to the data through the newly established independent Data Access Committee (DAC). The DAC is managed by the TDR, Special Programme for Research and Training in Tropical Diseases**, the committee reviews requests for data access on behalf of data contributors. 

Please get in contact with us to share or request access to malaria data.

Please contact us with your questions and suggestions, email info@wwarn.org.

Strength in Numbers: The WWARN Case Study of Purpose-Driven Data Sharing. American Journal of Tropical Medicine and Hygiene.

Secondary analysis and participation of those at the data source, Lancet Global Health.

*Clinical Data Interchange Standards Consortium (CDISC). 

** TDR is hosted at the World Health Organization (WHO), and is sponsored by the United Nations Children’s Fund (UNICEF), the United Nations Development Programme (UNDP), the World Bank and WHO.