©
World Bank
Fred McElwee from the Economics of Population Health Research Centre, and Andrew Farlow from the Department of Economics at Oxford University, used two individual patient data meta-analyses from the WorldWide Antimalarial Resistance Network (WWARN) to illustrate this approach.
Both of these studies analysed pseudonymised individual-level data from multiple existing clinical and observational studies, brought together into a curated, harmonised data repository. This approach is distinct from standard aggregated meta-analyses of published data, which do not have access to the individual-level data. The results of these studies helped bring about changes to malaria treatment guidelines for children, which had previously relied on a few and scattered clinical studies.
Financial incentives for developing medicines at accurate dosing are often weaker for diseases of poverty, especially for paediatric populations: the ability to pay for drugs is limited, markets are smaller, and intellectual property laws may be weak or poorly enforced.
“Carrying out a prospective clinical trial, whether it’s testing a new drug, formulation or dosing, can be expensive, especially in babies and children” says Fred McElwee, first author of the study. “There are, to some extent, financial and regulatory incentives for pharmaceutical companies to perform these studies for drugs sold in high-income countries. But the situation is more challenging for drugs which treat diseases that mainly affect poorer regions.”
Government and philanthropic funders have instead attempted to plug this gap, with collaborative product development partnerships such as the Medicines for Malaria Venture (MMV) and Drugs for Neglected Diseases Initiative (DNDi) (both WWARN/IDDO collaborators). Economists categorise these efforts as ‘push mechanisms’, which provide direct incentives for investment in drug research and development, and ‘pull mechanisms’ that incentivise commercial drug developers by funding the purchase of medicines (for example, through advanced market commitments).
Andrew Farlow says “Data sharing offers a third, complementary approach: investments in collecting, curating and then analysing individual-level patient data can be thought of as a pulley mechanism. Just as a pulley can amplify a small force to lift a heavy weight, data sharing maximises the utility of data that already exists, thus leveraging comparatively small investments into evidence that goes on to create large benefits for global health.”
Previous WWARN individual patient data meta-analyses have generated policy-changing scientific evidence, while still being an order of a magnitude cheaper than primary data collection. The team are now working on generating more precise estimates of these costs, to be reported in future publications.
Dr Farlow adds, “Efforts to generate good evidence, to base paediatric dosing guidelines on, must also be complemented by affordable access to formulations that are appropriate for babies and children.”
IDDO Director and co-author Professor Philippe Guerin says, “It takes time and effort to pool, curate and harmonise often heterogeneous datasets from multiple investigations, all using different standards. But the result is a much larger, richer dataset which enables us to spot patterns in populations too small to be seen in individual studies.
“The IDDO data repository can catalyse research that helps to ensure all patients, including babies and children, are given the right dose of drug for them, in the right form, and at the right time. This is the fruit of a collective effort”
The IDDO data repository currently makes over one million individual patient records available for reuse by the infectious disease research community, to generate the missing evidence that improves treatment and outcomes for patients.