10.12.2024

Beyond emergence: the expanding role of Real-World Evidence in healthcare

The significance of Real-World Evidence in healthcare is increasingly recognised beyond the confines of conventional clinical studies. It serves as a valuable supplement to randomised controlled trials, offering insights into how treatments perform across a wide array of patient demographics. This, in turn, influences the development of healthcare policies, clinical guidelines, and approval protocols, among other areas.

This essay delves into the challenges faced in Real-World Evidence studies, such as issues with data quality, regulatory and ethical considerations, and the complexities involved in interpreting results. It advocates for the adoption of more sophisticated methodologies and study designs to improve the reliability and usefulness of this approach in healthcare settings.

Furthermore, it explores forward-looking avenues such as integrating hybrid study models, applying artificial intelligence in Real-World Evidence studies analysis, and enhancing collaborative data-sharing initiatives. These steps aim to fully realise the potential of Real-World Evidence and Real-World Data studies in advancing patient care and facilitating medical research.

Introduction

Currrent significance of Real-World Evidence

Acknowledging and resolving challenges in RWE studies and RWD

Discussion

Future Directions

Introduction

In healthcare, Real-World Evidence (RWE) is the information and knowledge obtained from observing the outcomes of clinical practices in everyday life, beyond the structured environment of conventional clinical studies. This type of evidence is collected from a range of sources such as electronic health records (EHRs), patient registries, health-related applications, and wearable technology, to name a few (Figure 1).

RWE serves as a valuable adjunct to the data from randomized controlled trials (RCTs), enriching our understanding of the effectiveness of medical treatments across varied populations and real-life situations. The present essay will delve into the current significance, the identified challenges, and the prospective developments of RWE and Real-World Data (RWD), illustrating each of these entities through various medical and surgical situations.

Figure 1: Main Real-World Evidence sources in healthcare (adapted from https://www.mdpi.com/1660-4601/19/16/10159)

Current significance of Real-World Evidence

Firstly, RWE is instrumental in expanding our knowledge of treatment effects. Unlike RCTs with their somewhat narrow inclusion criteria, RWE captures a broader patient demographic, including varied ages, comorbid conditions, and those undergoing multiple therapies. Consequently, RWE offers a more holistic view of treatment outcomes in the broader population (Table 1). Furthermore, as far as RWD is concerned, this approach plays a crucial role in shaping healthcare policies and clinical guidelines by shedding light on treatment efficacy, safety, and impact on patient quality of life in real-life settings. This is especially relevant when RCTs are not feasible or ethical.

Table 1: Comparative strengths and limitations in RCTs and RWE studies (adapted from https://www.mdpi.com/1718-7729/30/2/143

Study type	Strengths	Limitations
RCTs	Gold standard for assessing efficacy Precise measures of treatment outcome under ideal conditions Randomisation promotes even distribution of confounders Well-established methodological rules High internal validity	Costly, cumbersome, and resource-intensive Limited number of patients Lower external validity due to: Highly selected participants due to strict inclusion and exclusion criteria Under-representation of groups such as older adults and people with comorbidities Protocols may not reflect typical care Use of surrogate endpoints that may not be valid measures of patient benefit
RWE studies	High external validity due to inclusion of large numbers of patients in routine care Provides evidence of effectiveness of new treatments in the general population, including those under-represented RCTs Performed relatively quickly and at low cost once data infrastructure is established Large sample sizes allow identification of rare events Facilitates long follow-up	Limited ability to detect rare and long-term side effects Susceptible to confounding and selection bias (interventions are not randomised) Methodologically difficult to do well Variable data quality Data may not include key clinical details, e.g., treatment plans, comorbidities, performance status Higher levels of missing data

Study type

Strengths

Limitations

RCTs

Gold standard for assessing efficacy
Precise measures of treatment outcome under ideal conditions
Randomisation promotes even distribution of confounders
Well-established methodological rules
High internal validity

Costly, cumbersome, and resource-intensive
Limited number of patients
Lower external validity due to:
Highly selected participants due to strict inclusion and exclusion criteria
Under-representation of groups such as older adults and people with comorbidities
Protocols may not reflect typical care
Use of surrogate endpoints that may not be valid measures of patient benefit

RWE studies

High external validity due to inclusion of large numbers of patients in routine care
Provides evidence of effectiveness of new treatments in the general population, including those under-represented RCTs
Performed relatively quickly and at low cost once data infrastructure is established
Large sample sizes allow identification of rare events
Facilitates long follow-up

Limited ability to detect rare and long-term side effects
Susceptible to confounding and selection bias (interventions are not randomised)
Methodologically difficult to do well
Variable data quality
Data may not include key clinical details, e.g., treatment plans, comorbidities, performance status
Higher levels of missing data

Acknowledging this, regulatory authorities such as the Food and Drug Administration (FDA) are increasingly valuing RWE for its contribution to drug development, the approval process, and the extension of existing drug indications, as well as for monitoring post-market safety.

Several practical examples of the RWE significance can be found, for instance in the field of oncology.

Firstly, in terms of clinical trials. A major oncology drug developer (Pfizer) used RWE from EHRs to create a more practical dosing study protocol for metastatic cancer, leading to less burdensome, cost-effective trials with reliable outcomes. The RWE approach also facilitated smaller, faster, and cheaper trials by enabling external control arms, aligning with regulatory standards. Secondly, as regards regulatory processes. Post-approval, RWE is increasingly critical for regulatory decisions. The FDA's Breakthrough Therapy Designation Pathway, for instance, has seen the incorporation of RWE to fulfill requirements for postmarketing studies. A specific example includes the FDA’s conditional approval of osimertinib for EGFR T790M+ non-small cell lung cancer, where AstraZeneca was requested to provide overall response-rate data from real-world cohorts. This highlights the evolving regulatory mechanisms to incorporate RWE in addressing evidence gaps from traditional clinical trials.

https://www.mckinsey.com/industries/life-sciences/our-insights/real-world-evidence-driving-a-new-drug-development-paradigm-in-oncology

The added value brought up by RWE studies has also been repeatedly substantiated in the recent scientific literature. To quote a few examples:

In lung cancer, RWE was shown to add to lacking information on outcome of immunotherapy in elderly patients (>75 years) or patients with ECOG performance status. While RWE studies support the use of immunotherapy in elderly patients, the data cautions on the application of checkpoint inhibitors in performance status 2 patients due to a lacking survival benefit. Such lacks being detected by RWE research call for RCTs matching real-world situations more closely. https://pubmed.ncbi.nlm.nih.gov/32446182/; https://pubmed.ncbi.nlm.nih.gov/30995172/
In patients with HPV-positive oropharyngeal squamous cell carcinomas, a causal inference analysis of RWD complemented the evidence from the published RCTs, by confirming the important role of cisplatin in the management of this patient population. https://pubmed.ncbi.nlm.nih.gov/34311328/

Thus, RWE helps to speed up approval processes, bringing new therapies to the patients, especially in the orphan and oncological drug development companies increasingly include an external RWD control arm in their clinical trials in order to accelerate therapies being provided to patients.

The increasing recognition of the importance of RWE is also supported by cancer registries. Currently, there are over 700 cancer registries globally that are investigating RWE in relation to different types of cancer, patient populations, standard oncological treatments, and the connections between these factors (Cancer Registries, ICCP Portal. International Cancer Control Platform; iccp-portal.org). Looking ahead, ensuring the data quality, accuracy, completeness, origin, and traceability of registry data will be crucial. This data serves as the primary source in observational studies or parts of clinical trials, and it is vital for the advancement of RWE.

https://pubmed.ncbi.nlm.nih.gov/35485775/

Acknowledging and resolving challenges in RWE studies and RWD

The primary challenges facing RWE and RWD fall into three categories: inconsistencies in data quality, concerns over regulations and ethics, and the complexities of interpreting results.

Firstly, the issue with RWD lies in its diverse sources and the fact that it is often gathered without initial research intent, leading to variations in data quality, completeness, and uniformity. To counter these problems, effective data management and analysis are crucial to ensure RWE's dependability.

Secondly, employing RWE requires careful consideration of privacy laws, patient consent, and ethical guidelines. It's vital to maintain patient privacy and obtain explicit consent for using their data in research, addressing these ethical prerequisites.

Thirdly, RWE studies, unlike RCTs, where conditions are tightly controlled, have to deal with numerous external variables. Thus, to draw accurate conclusions, meticulous study designs and advanced statistical techniques are necessary to navigate these complexities and avoid misinterpretation. In this, causal inference plays a pivotal role in RWE studies, for it enables the assessment of cause-and-effect relationships within observational data.

https://pubmed.ncbi.nlm.nih.gov/37900353/

This is crucial for translating observational insights into actionable knowledge, particularly in healthcare, where understanding the impact of interventions on patient outcomes is essential. By employing causal inference methodologies, RWE studies can offer robust evidence to support clinical and policy decision-making, bridging the gap between research findings and real-world applications.

Practical examples of challenges identified throughout the use of RWE have recently been reported in the scientific literature.

In China, the early efforts to generate RWE faced issues with data accessibility and sharing, the accuracy, completeness, and consistency of data, and lengthy ethical approval processes for retrospective database studies. These challenges highlight the need for explicit policies for data access, consent, privacy, and improved methods and standards for collecting and using RWE. https://www.bmj.com/content/360/bmj.j5262
Unlike RCTs, where methodologies are well developed, RWE often lacks standardized practices. There is a pressing need for rigorous, yet practical, methods and practices to define how collecting, analyzing, and reporting RWD should be conducted to ensure its broader use.https://www.mckinsey.com/industries/life-sciences/our-insights/real-world-evidence-from-activity-to-impact-in-healthcare-decision-making
The third example refers to privacy concerns and ethical issues regarding access to large datasets, which have restricted both the collection and sharing of data. Despite stronger support for access by independent academic research compared to industry-sponsored science, data access remains a primary hurdle to advancing the science of RWE analytics. https://www.mckinsey.com/industries/life-sciences/our-insights/real-world-evidence-from-activity-to-impact-in-healthcare-decision-making

These three examples illustrate the complex landscape of RWE production and utilization, underscoring the need for concerted efforts to overcome these barriers and leverage RWE effectively in healthcare: while RWE offers significant potential to enhance clinical decision-making and healthcare policy, addressing the challenges of data quality, standardization, privacy, and ethical considerations is crucial for realizing its full potential.

Discussion

The role of RWE in healthcare sees a significant transformation, underscoring its increasing value in enhancing patient care and medical research. RWE's integration into the healthcare landscape is leading to a more nuanced understanding of medical treatments' effectiveness and safety across diverse patient populations.

This broader perspective is crucial, particularly in areas where RCTs may not fully capture the complexities of real-world patient experiences. The expansion of RWE contributes not only to drug development and approval processes but also to the evolution of clinical guidelines, thereby supporting a more evidence-based approach to patient care (Figure 2).

Figure 2: Potential improvements derived from Real-World Evidence and Real-Word Data studies (adapted from https://www.mdpi.com/1660-4601/19/16/10159)

However, harnessing the full potential of RWE is not without its challenges. Data quality, privacy concerns, and the complexity of interpreting results from varied and unstructured data sources remain significant hurdles. These issues necessitate innovative approaches and methodologies, such as Federated Data Networks (FDNs), to ensure the reliability and usability of RWE.

https://marksmanhealthcare.com/2022/05/28/federated-data-networks-fdns-enhancing-the-quality-of-rwe-research/

Moreover, the application of advanced analytical techniques and the development of hybrid study designs represent forward-looking solutions that aim to address these challenges. As the healthcare sector continues to evolve towards more personalised and efficient models of care, the strategic integration of RWE into clinical and policy decision-making processes is anticipated to play a pivotal role, marking a new era of evidence-driven medicine.

It is essential to guarantee that researchers conducting RWE studies have the necessary methodological skills and the capability to precisely apply these methods. Important design aspects to consider include formulating a well-defined research question within a causal inference framework, selecting an appropriate data source tailored to the study's needs, incorporating new treatment users with comparator groups that closely match, and correctly classifying person-time while determining appropriate censoring strategies.

Future Directions

The growing importance of RWE studies in the armament of clinical research also paves the way for various innovative solutions.

Firstly, there's a growing trend towards hybrid study designs that incorporate elements of both RCTs and real-world studies. These designs aim to combine the rigour of RCTs with the generalisability of RWE, offering a more nuanced understanding of treatment outcomes.

https://www.frontiersin.org/articles/10.3389/frhs.2022.1053496/full

For instance, the Covid-19 pandemic has shown that the demand for immediate, efficient care can be met through the conditional approval of a new care principle. The Pragmatic Controlled Trial (PCT), a twin study using the Bayesian principle instead of randomisation, enables care under non-standardised everyday conditions. By describing the endpoint-specific risks of each individual patient and classifying the interventions, this approach creates an unbiased evaluation in a non-experimental, but risk-stratified and controlled study.

https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-1819-6237

Secondly, the application of artificial intelligence and machine learning techniques to RWD can uncover patterns and insights that would be difficult to detect through traditional analysis methods, potentially leading to breakthroughs in understanding disease progression and treatment responses.

https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-021-01403-2

Thirdly, more collaborative efforts for data sharing are being put forth. Enhancing the utility of RWE indeed requires overcoming data silos through a tighter, inter-sectorial collaboration among healthcare providers, researchers, and regulators. Initiatives that facilitate data sharing while respecting patient privacy will significantly boost the quality and impact of RWE.

https://www.mdpi.com/1660-4601/17/9/3046

Regarding RWD, while it has the potential to streamline research and development timelines, biases can arise when relying on a single data source. These biases are linked to diagnostic and therapeutic equipment, the diversity of patient groups, and limited training models. Moreover, the personal and sensitive nature of health data necessitates adherence to privacy rights and regulatory standards. To overcome these challenges and fully leverage RWD, innovative approaches are essential. Federated Data Networks represent one such strategy aimed at enhancing the utility and integrity of RWD.

https://marksmanhealthcare.com/2022/05/28/federated-data-networks-fdns-enhancing-the-quality-of-rwe-research/

In summary, Real-World Evidence is transforming the field of clinical medicine by providing valuable insights into how medical treatments perform in a wide range of patients, enhancing their effectiveness and safety. Despite facing obstacles, with thoughtful management and innovative strategies, Real-World Evidence studies can greatly enhance our comprehension of medical interventions, leading to better treatment and management outcomes and more streamlined healthcare operations. As we progress, incorporating Real-World Evidence into clinical and policy decisions will become more prevalent, heralding a new era of personalised, evidence-driven medical care.

Clinique de Genolier

Hospitals

Centres

For patients

About us

Hospitals

Centres

For patients

About us

How can we help you?

Top searches

Beyond emergence: the expanding role of Real-World Evidence in healthcare

Introduction

Current significance of Real-World Evidence

Acknowledging and resolving challenges in RWE studies and RWD

Discussion

Future Directions

Dr. med. Jacques Bernier

Science and Innovation