Viatris

Kelly H. Zou

RWE and Digital

Given the prevalence of many diseaseswith low adherence to treatments and therapies, real-world evidence(RWE) and digital innovationsthat are beyond randomized controlled trials (RCTs) can help providetimely and actionableinsights. RWE can be critical comorbid conditions as risk factors for coronavirus disease 2019 (COVID-19).

RWE, digital innovation and artificial intelligence (AI)are useful for patient-centricity, medicines regulation, and sophisticated algorithms based on machine learning (ML) and deep learning (DL). It is important to harness real-world data (RWD) for regulatory purposes, via quality and accessibility, and under data privacy regulations. For medical, RWE can help enable partnerships with government agencies making authorization and treatment guidelines decisions, deliver enriched content for medical education, and provide critical insights to key opinion leaders. For regulatory, on the other hand, RWE can help prioritize product registration process, support regulatory authorities’ data queries, and provide insights for label expansion. For commercial, RWE can provide information to support pricing and reimbursement decisions by payors, support business strategies and business development plans, enable commercial and due diligence assessments, make growth plans for informed decision-making, and deliver insights for market access.

RWE and RCTs

RCTs are rigorous in assessing a causal effect between treatment options and outcomes. There are several elements their designs:(1) random allocation to intervention groups; (2) patients and trialists should remain unaware of which treatment was given until the study is completed-although such double blind studies are not always feasible or appropriate; (3) all intervention groups are treated identically except for the experimental treatment; (4) patients are normally analyzed within the group to which they were allocated, irrespective of whether they experienced the intended intervention (intention to treat analysis); (5) the analysis is focused on estimating the size of the difference in predefined outcomes between intervention groups.However, despite being the gold standard in terms of evidence hierarchy, a lack ofpatient diversity  among certain groups that are underrepresented is commonplace in RCTs. 

RWD can be fragmented siloed and generally not integrated across different data-capture systems via interoperability. 

According to the United States (US) Food and Drug Administration (FDA), such data can come from a wide variety of sources, such as electronic health records, claims and billing activities, product and disease registries, patient-generated data including in home-use settings, and data gathered from other sources that can inform on health status, such as mobile devices.Bring evidence together systematically and holistically can be valuable.

Patient-Centricity

To analyze the uses of medications, especially for patients who are chronically ill, there are three concepts that are important, i.e., adherence, compliance and persistence (Table 1), AnRWE study design generally consists of a patient-specific index event, which can be either a particular diagnosis or therapy via their diagnosis or drug codes, or a combination of both, per patient. A pre-index period, which can be arbitrarily from 90 days to a year, potentially captures the patient’s comorbid conditions. Subsequently, a post-index period, which can be arbitrarily from a year or longer, follows up on the patient’s medication use, initial fill, refill/refills patterns, non-invasive/invasive procedures, as well as healthcare resource utilization patterns such as hospitalization, as well as emergency department, intensive care unit, and/or outpatient clinic visits. Understanding of adherence is particularly important in the era of COVID-19 since “social determinants of health  are critical elements that can impact not just the likelihood of having an NCD or becoming infected with COVID-19, but also access to healthcare, and a patient's adherence and persistence with their treatments.”

AI, ML and DL

The use of AI, ML and DLis beneficial to patients, for example, for biomarker validation to improve diagnostic accuracy. There are various techniques to optimally linear or nonlinear combine biomarkers to improve diagnostic accuracy, and data science methods, e.g., classification and regression trees and cluster analysis, as well as classical logistic regression, may be applied.

In summary, RCTs, RWE and digital innovation can jointly be highly useful to advance medical sciences, understand disease burdens, and optimize treatments precisely, timely and effectively. Beyond detections and diagnoses, RWD of high quality and representativeness may help comprehensively map patient journeys. Such information can help evaluate diagnoses, therapies, and prognoses holistically and comprehensively. Furthermore, beyond single-pill approaches,integrated care and cutting-edge technologies rely on advances in outcomes research and data science.