Harnessing the True Transformational Power of Real-World Data in Healthcare - Digital Solutions, IT Services & Consulting - Payoda

Harnessing the True Transformational Power of Real-World Data in Healthcare

Real-World Data (RWD) has become a terminology that has gained serious gravity in the Medical and Pharmaceutical circles since 2018. Its popularity and necessity have grown multifold over the years. Of late, due to the increasing complexity, cost, and power of medicines, the U.S. Food and Drug Administration and the EU have started evaluating the newly discovered drugs and results of clinical trials by relying heavily on Real-World Data and Real-World Evidence. According to the FDA, RWD is the data related to the usage, potential risks, or benefits of administered drugs and treatment procedures that are derived from sources other than the conventional randomized clinical trials.

The FDA now views data generated from Real-world studies as a mandatory and critical complement to Randomized Clinical Trials (RCT). The European Medical Regulations require Medical Technology and Drug Development companies to submit more RWD in order for there to be a better chance of approvals.

Real-World Data for Healthcare is collected from a variety of sources. It is the data relating to patient health status, treatments, drugs, and dosages offered, the improvements, adverse effects, and other related statistics. RWD is usually high-volume because it is routinely collected from a multitude of healthcare providers.

RWD is worth its weight in gold, as it provides critical insights into the health situation and treatment effectiveness in a more representative and diverse group of patients. It also allows for the generation of hypotheses regarding rare conditions, effects, and biomarkers that are out of the ordinary.

How can RWD help according to the U.S. FDA:

• Generating a theory for testing in randomized controlled trials that can make the results from trials more dependable.

• Develop new tools for drug development, which includes the vital biomarker identification process.

• Assessing the feasibility of clinical trials by examining the impact of planned inclusion/exclusion criteria in the relevant cohort, both within a geographical area or at a particular trial site.

• Identification and prior probability distributions in Bayesian statistical models.

• Determining prognostic markers or patient baseline characteristics for better classification and enrichment of treatments.

• Assembling geographically distributed research groups (e.g., in drug development for rare diseases or targeted therapeutics).

Sources of Real-World Data:

RWD can be derived from EHR, medical billing and claims data, data from disease registries, pharmacy data, social media, patient-generated data and from the data gathered from health monitoring devices and wearables. Real-World Evidence (RWE) is the clinical evidence concerning the potential benefits or the risks involved in the usage of a medical product derived from the analysis of RWD. RWD can be transformed into RWE with the help of analytics platforms powered by Big-Data Analytics or Machine Learning algorithms that are capable of ingesting and processing huge caches of data and extracting the essential information that is capable of providing actionable insights.

Transforming RWD to RWE:

A data analytics and statistics presentation platform that deals with standardized data accelerates the healthcare data flow by employing common data models such as Fast Healthcare Interoperability Resources (FIHR), Observational Medical Outcomes Partnership (OMOP), or other similar ones. The platform should be data agnostic and should be capable of ingesting, standardizing, processing, and leveraging the input data to provide actionable evidence.

RWD/RWE use case in the drug life cycle:

RWE can be leveraged to make clinical trials more effective by providing support in patient recruitment and drug repurposing. For example, during clinical trials, RWE gathered from the studies of a currently marketed product in a similar pharmacology family can have a positive effect on the product portfolio by bringing to light the previously unknown, unforeseen, positive side effects as new potential indicators. The most famous example of such a drug repurposing being Viagra, which was initially developed to be a drug to lower blood pressure, but an evidence-based discovery was the side effect that led to the drug ultimately being approved as a short term remedy for erectile dysfunction. The below image describes the phases in a drug life cycle and how RWE works.

Gaps in Healthcare Real-World Data and the role of IoHT:

The following are considered as gaps in Healthcare RWD:

  1. Unreliable data between encounters
  2. Artificial settings can sometimes result in anomalous data.
  3. Most of the data is subjective.

Data from wearable health devices such as Fitbit, Jawbone, Apple Watch, and others come under the umbrella of the Internet of Healthcare Things. The data from such devices and the genomic datasets close the gaps present inherently in RWD. Capturing health data in the patient’s natural setting, processing real-time and continuous data from large population results in objectivity. The below picture shows how IoHT data is true, the enriched version of RWD.

The confluence of RWD, Big-Data Analytics, and Machine Learning Algorithms:

Leveraging the all-round capabilities of Big-Data Analytics and incorporating Machine Learning algorithms, if a collaborative data repository is developed it would help healthcare providers and pharmaceutical companies utilize alternative analytical techniques, drug repurposing options, optimizing patient journey, and adopt better treatment approaches.

An integrated data warehouse can be created upon which the ML algorithmic magic needs to be applied. Data from various sources such as from recently conducted clinical trials and treatments (from a SAS data set), Genetic data, Lab test results such as blood, EEG, ECG, x-ray and pet scans, insurance claims, and medical literature needs to be included as part of a wholesome RWD.

Next-Gen Implementation of Big Data in Healthcare: An Analytical Ecosystem and its capabilities:

Capabilities of such a platform would include:

  1. Mapping common pathways across billions of events, providing a well averaged causal inference and cognizance.
  2. Visualization of the relationships between millions of entities from different phases of the healthcare journey.
  3. Detection of anomalies within large data sets.
  4. Accelerating the development of next-gen predictive models with distributed computing.
  5. Deriving insights from unstructured physician notes using NLP and text mining.

How does RWD differ from data obtained from RCT:

Data collected through Randomized Clinical Trials and RWD differ in several ways. RCT data is documented based on the observations recorded from patients who are in a controlled setting, whereas RWD records data from the patient’s natural setting. The number of patients from which the data is derived in RCT is less, whereas in RWD it is exponentially greater. RCT data is based on a limited number of trials and patients due to the cost involved, complexity, and time taken. In order to record rare adverse events, the data must be derived from a much larger population, which is why RWD is considered an essential source of data that complements or at times overrides data from RCT.

Outcomes of Adopting Real-World Data:

For Healthcare Providers:

With the aid of RWD, healthcare provides gain the ability to augment their intelligence on patient profiles, diagnosis, treatment pathways, prognosis prediction, and potential adverse effects of medications. By having evidence-based systems and methodologies, providers are able to leverage more efficient, clinical decisions.

For Payers:

Insurance companies can leverage RWD to determine how a medicinal product is actually used by the patients and also gain insights that help them to achieve personalized reimbursement based on usage, value, and outcomes.

For Patients:

With RWD being widely adopted, patients will have more transparency regarding healthcare data processing and will play the role of key enablers of RWD and pave the way for next-gen approaches such as personalized medicine.

For Regulatory Authorities:

Regulators such as the FDA and EMA use RWD for post-market safety and benefit/risk studies. The 21st Century Care Act clearly specifies RWE as the key enabler for regulatory decisions and market approvals. The FDA has released a framework that explains how RWE is incorporated and used as a primary marker to support the process for drug regulation, submission, and approval. The FDA has proposed to conduct a full post-market safety study using just RWE very soon.

In Drug Development and Approval:

RWD has proven itself as a game-changer for drug manufacturing companies as it has shown greater success rates with better outcomes in almost all the critical stages of drug development.

In Precision Medicine:

RWD in combination with AI can be used to improve clinical trial design and patient recruitment for the trials. AI transforms RWD into actionable insights, which is the driving factor behind Precision Medicine.

It is possible for MedTech companies to provide personalized medical decision support for specific patient groups by combining RWE based on patient outcomes and genomics data which enable physicians to assess the treatments and drug efficacy based on patient characteristics.

Future of RWD:

The ultimate goal of leveraging RWD is to streamline the healthcare processes in order to promote value-based care, develop new medical devices and medications that improve patient outcomes. But for this to happen the following must be adhered to.

  1. The strength of the RWE churned out from the RWD depends on the quality of the clinical study methodology and the reliability (based on data accrual and data quality control results) and the appropriateness of the underlying data.
  2. Healthcare organizations must strive to achieve a truly collaborative platform to get a holistic view of the patients and their data. The data should be highly interoperable.

Global health research networks are providing a platform to exchange RWD combined with AI and Big-Data Analytics to health research and pharma companies to optimize several of their processes. In the past couple of years, established IDNs have started to set up their own data analytics platforms in order to source and offer RWD access to their patients and other relevant stakeholders. These platforms are also bolstered with tools needed to protect patient privacy and gain their consent and hence remaining compliant to HIPAA, GDPR standards.

Challenges and Opportunities in RWD:

  • Development of user interfaces that help clinicians capture all the relevant data required for a comprehensive RWD feed in a more efficient manner.
  • Infusing AI, ML, and Big-Data Analytics to enrich RWD and convert them into trustable RWE.
  • Data interoperability challenges need to be solved by adopting common data models. Of late several IDNs are looking toward Observational Medical Outcomes Partnership (OMOP) as well as Clinical Data Interchange Standards Consortium’s (CDISC) Study Data Tabulation Model (STDM) as standards to be followed. And across the board, the FIHR is gaining the greatest traction as the most acceptable model for EHR exchange. Implementing and linking these standards will increase data fluidity, access, and availability.

The true objective of RWD is to provide actionable insights at every point of the healthcare life cycle. But with technological improvements and infusions, there is no doubt that RWD can self-enrich and provide multichannel value in innovative and unforeseen ways for the greater good of the entire healthcare system.

Leave a Reply

Your email address will not be published. Required fields are marked *

two × 2 =