A clinical trial conducted by Dr James Lind in the year 1747 on patients with scurvy, made the basis of modern-day RCTs. Since then, there have been many advancements in the art and science of conducting clinical trials. These advancements include: methodological know-how, framing of an organized structure for conducting and reporting of trials, laying down of ethical guidelines and regulatory frameworks etc. Owing to this robust framework, today a well-planned and meticulously executed RCT is considered indispensable in finding the evidence for any given intervention.

Further, some other studies that have human involvement vis-à-vis an expert opinion, case reports, and observational studies also contribute towards evidence building. Though these studies also use a structured framework, owing to fundamental features of RCTs, these are placed quite high in the evidence hierarchy. This hierarchical ladder is popularly known as the “evidence-based pyramid”.

In this pyramid, systematic reviews are placed atop as they help to curate the “net” evidence by combining the information from all relevant clinical trials. The pooled effect from SRs and meta-analysis is frequently used by policymakers and healthcare professionals to frame evidence-based recommendations. Overall, this concept of “evidence-based pyramid” has guided the medical research since decades; and in the quest of finding the evidence for any intervention, different stakeholders including clinicians, academia, and drug-development or healthcare-technology companies have diverted their resources in planning and conducting high quality RCTs.

But despite the inherent qualities of an RCT and its ability to effectively contribute towards evidence generation, there are some concerns and limitations. These include cherry-picking a specific group of patients, enforcing the patients to adopt specific behaviors which may not be always pragmatically possible, choosing only a few objectives while neglecting the impact of the interventions beyond the target research area etc. This has two main repercussions: first, the outcome determination is based only as per data collected in the trial without any consideration to other available data; and second, it limits the generalizability of trial results thereby obscuring its true impact on healthcare stakeholders.

Hence, it is being emphasized to integrate the knowledge from clinical trials with the information gathered from real-life situations, as this might help to generate the “true” evidence. This seems possible too in the present healthcare scenario, as today we routinely collect clinical and other patient-related data which is available in EHR systems, patient registries, drug inventories, claims/billing records etc. Moreover, nowadays enormous data is also being generated at home with the use of connectable and wearable IoT devices. This includes data from smartwatches, pedometers, health Apps, IoT-based food, and exercise trackers and not to miss, App-connected toothbrushes, brushing apps etc. These modern-day sophisticated data collection methods are sensitive, and are becoming increasingly reliable. This data is stored in hospital servers or is available in cloud-based archives. Though all such data is encrypted, it can be accessed after seeking due approvals from regulatory bodies.

Overall, the data collected in such real-life situations, outside the realm of clinical trials, is referred to as real-world data (RWD)1. And with the use of advanced data analytics, and AI-based deep-learning and neural network applications, “this data” can be made to shake-hands with data generated from the clinical trials conducted in pseudo-perfect situations. The evidence curated from this wed-lock is called real-world evidence (RWE). RWE can be used to find various hidden relationships and has the potential to reengineer the evidence hierarchy in the healthcare system. Another utility of RWE is to determine the evidence about an intervention, meant for rare diseases, wherein the studies are mostly limited to case studies or small cohorts. In such situations, exploring RWD from EHRs, disease registries etc. might give an insight into the epidemiology, possible treatments, and outcomes2. It may also hint about their associations with other diseases. Thus, RWE generated from such endeavors would surely help in exploring new treatment pathways and seeking drug approvals for such unique indications.

In yester years, various drug/technology approvals by FDA and other health authorities have included RWD; depicting the adoption of RWE by the regulatory agencies3. In fact, FDA has also released a structural framework for determining RWE and optimizing its utility into practice. Recently a Causal Roadmap framework has also been released to overcome the challenge about difficulty in estimating a causal relationship with data collected outside of the “traditional” clinical trial space4. With so much going on in this space, the day is not far when the inclusion of RWD for determining the evidence would not be optional anymore.

The benefits of this are far beyond imagination. For example, imagine how advantageous it could be if we conduct a trial on a remineralizing toothpaste and have additional RWD from a brushing app about the actual frequency and duration of brushing, or in a trial on the management of periodontitis with some real-world data about patients’ diabetes status, available from EHRs and registries. In both these situations, we can conceptualize how beautifully the trial data and RWD would complement each other and the output might be a bit closer to the “true” evidence.

Owing to its potential to impact evidence-based healthcare practice, I opine that in the times to come, we would have to reshape the present “pyramid” of evidence by including RWE studies within the pyramid. One way of doing this can be by using a split-pyramid design with vertical or horizontal split blocks. The blocks should be attached to each other as jigsaw-based building blocks are attached together. The blocks on one side would represent trials, observational studies etc. while the other side may include RWE studies complementing trials and other related studies. Whichever shape we may choose, one thing is sure that RWD and RWE would soon become an effective arrow in the quiver of evidence-based practice.