In the current climate of increasing drug prices, healthcare decision-makers are increasingly relying on real-world evidence to improve certainty regarding a product’s performance outside of rigidly controlled clinical trials. However, identifying high-quality, contextually relevant real-world data sources remains a significant challenge, particularly outside of major markets or for rare populations. Consequently, researchers often need to leverage real-world data generated in markets outside their target population. This raises important considerations regarding data transportability.
Data transportability is a metric of external validity that assesses the extent to which results from a source population can be generalised to a distinct target population. Employing transportability methods enables the adjustment and extension of evidence, fostering greater confidence in applying findings across diverse populations.
Despite the potential for real-world evidence to assist healthcare decision-makers, there is often hesitancy to fully utilise real-world evidence, or limitations in its availability. For instance, a real-world data survey of European health-technology assessment (HTA) agencies noted that some of the most important barriers to using real-world evidence in reimbursement decisions were the lack of necessary data sources and the delayed data access often associated with real-world data. As such, there are two key reasons why utilising transported data is desirable to researchers:
Transportability methods enable researchers to fulfil evidence requirements without duplicative research
Transportability methods can avoid repetition of research, increase efficiency, and decrease time / resources burdens, accelerating regulatory and HTA decision-making
Although transported data has great potential to increase equity of access, there are important limitations which can impact data comparability and should be acknowledged:
Variability in treatment patterns is influenced by differing clinical guidelines and protocols, medication availability, and reimbursement statuses across different geographies.
Population differences across regions are shaped by demographic, lifestyle, socioeconomic, and epidemiological factors such as disease incidence and progression.
Differences in data quality, completeness, and transparency can affect the accuracy and reliability of results when transporting data. This can include outcome definition discrepancies, data missingness, and potential biases arising from how information is recorded and maintained. The presence of unmeasured confounders also complicates transportability, as it limits the capacity for accurate adjustments.
In the face of these limitations, it is vital that researchers utilise data transportability methods to ensure robustness in their analysis. There are two main classes of data transportability method: weighting methods and outcome regression methods.
Weighting involves reweighting individuals in the source population to better reflect the characteristics of the target population. This process begins with identifying effect modifiers – variables that influence treatment performance – and then calculating weights based on the likelihood that an individual in the source population would be represented in the target population, often using inverse odds of sampling weights.
Those individuals who closely resemble the target population are assigned larger weights, while those with more divergent characteristics receive smaller weights. During analysis, these weights are applied to adjust the estimated treatment effects, ensuring that the final results more accurately represent what would be observed if the treatment were administered directly within the target population.
Outcome regression methods involve developing a predictive model based on source population data to estimate treatment effects within a target population. First, a regression model is constructed to predict outcomes based on treatment effect modifiers identified in the source data. This model is then applied to the characteristics of individuals in the target population, generating predicted outcomes under different treatment scenarios.
By averaging these predicted outcomes, the method estimates the potential effect of the treatment if it were implemented in the target population. In practice, both methods are often combined, ensuring double robustness.
When reporting analyses which have utilised transported data, it is important for researchers to adhere to high-quality, transparent, and credible reporting, thus improving trust in transported real-world data for informing regulatory and HTA decision-making. Standards for reporting are outlined in the NICE framework for real-world evidence.
These comprehensive guidelines emphasise the importance of thoroughly transparent reporting, detailed descriptions of data provenance, careful assessment of data suitability, rigorous quality reporting, and explicit handling of differences and limitations. They also highlight the necessity of clearly describing statistical methods, conducting sensitivity analyses, and openly discussing the interpretation, uncertainty, and potential biases of transported data to ensure validity, credibility, and applicability of the findings.
In summary, transported data can be particularly valuable for jurisdictions with limited local data. Transported real-world data can help to address local evidence gaps, reduce duplication of efforts, and enhance the generalisability and external validity of treatment assessments for patient populations that are underrepresented or excluded in clinical trials.
To ensure reliable application, transparent reporting of transportability methods – including the rationale for data use and detailed documentation of data collection, analysis, and limitations – is essential. In the future, further guidance from payers and greater harmonisation of standards across countries are needed to confidently support the acceptance and use of transported data in decision-making.
On 10th June 2025, Declan Summers, our UK Head of Real-World Evidence, moderated an issue panel at the 2nd Annual IMPACCT Real World Evidence Summit in Europe, exploring whether data transportability can bridge evidence gaps and expedite access to therapies, particularly in markets where high-quality local real-world data are lacking. This session featured an exciting panel, including:
The panel featured a lively discussion regarding the key considerations for the effective use of data transportability methods. The importance of assessing similarities in populations, disease characteristics, and healthcare ecosystems to ensure valid transfer of evidence was highlighted. Speakers emphasised the need for building credibility through transparent communication of assumptions and detailed data provenance. A key discussion focused on the often-underappreciated variations in health systems that can impact the interpretation of transported real-world data. The panel agreed that assessing the comparability of healthcare systems when transporting data is a vital consideration, which is often overlooked.
Regarding the much publicised efforts to harmonise data, there was a consensus amongst the panel that initiatives like the EHDS Regulation will not fully resolve structural differences in data generation, albeit that harmonisation can help reduce issues such as data missingness and improve protocol consistency. The panel also acknowledged that harmonisation of data shifts responsibility to data users to understand and account for underlying data nuances.
The panel agreed that transportability should serve as a complement to local data generation, aiming to strengthen evidence diversity, support smaller markets, and inform health technology assessments in underrepresented populations. However, broader adoption hinges on methodological transparency, cultural shifts among decision-makers, and a more proactive approach in promoting the value of real-world evidence.
If you would like any further information on the themes presented above, please get in touch, or visit our Real-World Evidence page to learn how our expertise can benefit you. Declan Summers (UK Head of Real-World Evidence) created this article on behalf of Costello Medical. The views/opinions expressed are his own and do not necessarily reflect those of Costello Medical’s clients or affiliated partners.
