Real-world data (RWD) and real-world evidence (RWE) are not genuinely new concepts. They emerged in scientific literature three decades ago, following the global transition to keeping electronic healthcare records (EHRs).
Over the years, pharma companies worldwide have used patient data from real-world settings for retrospective efficacy analyses and to support future R&D efforts.
However, the efforts to conduct RWD collection on a global scale have been challenged with limited technological capabilities as well as with regulatory, logistical, and legal issues.
Today, with the rise of new cutting-edge technologies, it is possible to attain RWD more quickly, assess it more thoroughly, so that its therapeutic relevance could be put to use fluidly.
In such a perspective, the potential of RWD to improve drug development, accelerate registration, and improve healthcare outcomes is becoming virtually limitless.
Real-World Data vs. Real-World Evidence: Definitions
Real-World Data (RWD) in medicine is data obtained from many independent sources that follow healthcare outcomes in a diverse population.
The FDA defines it as data relating to patient health status and/or the delivery of health care, routinely collected from various sources.
RWD is mainly observational and pertinent to actual clinical practice, in contrast to strictly controlled experimental data acquired in randomized clinical trials. It is primarily sourced from real-world medical environments and reflects realistic healthcare scenarios.
RWD examples include data derived from electronic health records (EHRs), pharmacy registries, patient surveys, medical claims, billing data, product and disease registries, e-health devices, clinical trials, and observational studies.
Real-World Evidence (RWE) is the clinical evidence about the proper usage, benefits, and harms of a therapeutic or medical device obtained from RWD.
It is extracted from RWD to form an analysis for the potential clinical study, base rationale, and engineer proper trial infrastructure.
The Challenge of Randomized Clinical Trials
Randomized clinical trials (RCT) are the fastest way of obtaining regulatory approval for a new therapeutic or medical device. Their design can vary depending on the trial purpose and the desired outcomes.
Randomized, placebo-controlled, double-blind clinical trials are the golden standard in the industry, especially when assessing the efficacy and safety of a new therapeutic. They are designed to prevent study bias and obtain accurate data relevant to the study outcome.
Given that they operate under controlled conditions, they answer the questions only within a trial-defined framework and, as such, do not necessarily provide clinical relevance for the real-world environment.
This is why in the biomedical world, there is a growing initiative to pursue creating a clinical trial framework based on the evidence obtained from real-world clinical settings.
The Advantages and Use of RWD
When properly collected, structured, and analyzed, the RWD can show unique insights into populations, diseases, and the therapeutical efficacy of medications and procedures.
According to experts, one of the most significant values of RWD is that data collection does not take place under strictly controlled conditions, thus allowing the inclusion of patients that would be considered as outliers in classical RCT scenarios.
Furthermore, data obtained from real-world settings are exceptionally valuable in patient care, especially in the case of a rare or novelty disease when clinicians cannot account for the entire afflicted population. In clinical settings, the use of RWE would help mitigate uncertainties by filling knowledge gaps that RCT did not otherwise address.
RWD could reveal how patient lifestyle and inherent characteristics affect health outcomes, which could aid in early disease detection and the development of prevention protocols.
Pharma and biotech industries could use RWE to build studies and to increase efficacy in R&D departments, ultimately accelerating their products to the market. RWD is useful when assessing the effectiveness of a product, or it can be used to facilitate decision-making in RCT, for example, in hypothesis generation, identifying biomarkers of a study, mapping of comorbidities, assessing feasibility, etc. When deployed correctly, RWD sources can be used to develop analysis to support many types of study designs or to obtain evidence for interventional studies.
When using RWE, the authorities could more easily assess the medication efficacy in diverse populations and follow the adverse event progression. Pharmaceutical companies and health insurance payers mainly utilize RWE to understand disease progression mechanisms and provide relevant assessments and appropriate care. In the US, FDA uses RWD and RWE to monitor post-marketing activities and adverse events to make regulatory decisions.
The ultimate benefit of RWD is to get the adequate treatment to the patient at the right time, measure the right outcomes in a proper way, and demonstrate the power of interventions.
Challenges in RWD Use
Despite the increasing recommendations to adopt the RWD framework, there are still challenges and limitations that complicate the collection, structuring, and effective use of this data.
Considering that this data has to be sourced from multiple repositories, the collection of such creates discrepancies between various sources. RWD repositories are extensive and seldom used in their native form. In order to be put to purpose, the data needs to be appropriately structured and analyzed.
Uneven Data Quality
In RCT settings, there are very stringent regulations that apply for on-site data acquisition. The quality controls for the RWD collection are yet to be officially established.
The on-site data quality varies significantly, considering that there are inconsistencies in EHR structures and content across institutions, which is even more complicated with irregular timing of clinical visits and testing. Furthermore, most of the clinical data lacks some differentiating information, such as genomic profiling and patient-reported outcomes.
Limited Access
Most of the clinics worldwide fill their EHR repositories with unstructured data that is not machine-readable, and acquisition of such presents a logistical and legal challenge.
Additionally, in the case of rare diseases, there is an extremely small and globally distributed patient population, which requires access to multiple specialty hospitals.
Acquisition of RWD also requires attaining patient consent and resolving ethical and privacy concerns for data collection and sharing.
Lack of Standardization
Lack of standardization of data between different healthcare systems leads to poor quality analyses, limited transparency in to laboratory results and therapeutic methods, and creates bias in results.
One of the contributing factors that limit RWD collection worldwide is the lack of integration of disparate data sources for both prospective and retrospective studies.
Mitigating The Challenges with Technology
Today, with the rise of new advanced technologies in healthcare, prevalently the emergence of fast data exchange through the adoption of interoperability standards, RWD collection projects can be streamlined to the maximally desirable extent.
New, advanced RWE generation methodologies surpass the traditional ones by efficacy and cost-effectiveness.
Traditional RWE approaches implied doing descriptive analyses to characterize patients and established matching techniques to compare groups of patients with similar characteristics.
Advanced RWE generation frameworks include predictive models, machine learning technologies, probabilistic causal models, and unsupervised algorithms to extract deeper insights.
Furthermore, in order to reduce disparities and ensure data standardization in RWD collection, novelty software use or reference global code sets, such as SNOMED CT, ICD-10, and LOINC.
There is an increasingly growing managerial initiative to adopt these international codes in local healthcare settings, facilitating worldwide data exchange and more quickly feeding the RWD database with valuable information.